NO341404B1 - Monoclonal antibodies and method for their preparation, as well as hybridoma, polynucleotide, vector, cell and composition for the treatment of cancer - Google Patents

Description

FIELD OF THE INVENTION

The invention described here relates to monoclonal antibodies and methods for their production, as well as hybridoma, polynucleotide, vector, cell and composition for the treatment of cancer.

BACKGROUND OF THE INVENTION

Cancer is the second leading cause of human death after coronary disease. Millions of people worldwide die from cancer every year. In the United States alone, as indicated by the American Cancer Society, cancer causes the death of well over half a million people annually, with over 1.2 million new cases diagnosed per year. year. While deaths from heart disease are on the decline significantly, those resulting from cancer are generally on the rise. In the early part of the next century, cancer is predicted to become the leading cause of death.

Worldwide, many forces are the leading cause of death. In particular, carcinomas of the lung, prostate, breast, colon, pancreas, ovary and bladder represent the primary cause of cancer death. These and virtually all other carcinomas share a common lethal feature. With very few exceptions, metastatic disease from a carcinoma is fatal. Furthermore, even for those cancer patients who initially survive their primary cancer, common experience has shown that their lives are dramatically changed. Many cancer patients experience strong anxiety driven by the awareness of the potential for recurrence or treatment failure. Many cancer patients experience physical weakness following treatment. Furthermore, many cancer patients experience a recurrence.

Prostate cancer is the fourth most common cancer in men worldwide. In North America and northern Europe, it is by far the most common cancer in men and is the second leading cause of cancer death in men. In the US alone, well over 30,000 men die annually from this disease - second only to lung cancer. Despite the magnitude of these numbers, there is still no effective treatment for metastatic prostate cancer. Surgical prostatectomy, radiation therapy, hormone ablation therapy, surgical castration and chemotherapy continue to be the main treatment modalities. Unfortunately, these treatments are ineffective for many and are often associated with unwanted consequences.

On the diagnostic front, the lack of a prostate tumor marker that can accurately detect early-stage localized tumors remains a significant limitation in the diagnosis and treatment of this disease. However, although the serum prostate specific antigen (PSA) test is a very useful tool, its specificity and overall utility are widely considered lacking in many important respects.

Advances in the identification of additional specific markers for prostate cancer have been enhanced by the generation of prostate cancer xenografts that can recapitulate different stages of the disease in mice. The LAPC (Los Angeles Prostate Cancer) xenografts are prostate cancer xenografts that have survived passage in severe combined immunodeficiency (SCID) mice and have shown the capacity to mimic the transition from androgen dependence to androgen independence (Klein et al, 1997, Nat. Med. 3 :402). Later identified prostate cancer markers include PCTA-1 (Su et al, 1996, Proe. Nati. Acad. Sei. USA 93: 7252), prostate-specific membrane (PSM) antigen (Pinn i et al, Clin Cancer Res 1996 Sep 2 (9) : 1445-51), STE AP (Hubert, et al, Proe Nati Acad Sei USA. 1999 Dec 7; 96(25): 14523-8) and prostate stem cell antigen (PSCA) (Reiter et al, 1998, Proe. Nati. Acad . Sei. USA 95: 1735).

While previously identified markers such as PSA, PSM, PCTA and PSCA have facilitated efforts to diagnose and treat prostate cancer, there is a need for the identification of additional markers and therapeutic targets for prostate and related cancers to further improve diagnosis and therapy.

Renal cell carcinoma (RCC) accounts for approximately 3 percent of adult malignancies. Once adenomas reach a diameter of 2 to 3 cm, malignant potential exists. In adults, the two principal malignant kidney tumors are renal cell adenocarcinoma and transitional cell carcinoma of the renal pelvis or ureter. The incidence of renal cell adenocarcinoma is estimated at more than 29,000 cases in the United States and more than 11,600 patients died of this disease in 1998. Transitional cell carcinoma is less frequent, with an incidence of approximately 500 cases per year. year in the United States.

Surgery has been the primary therapy for renal cell adenocarcinoma for many decades. Until recently, metastatic disease has been refractory to any systemic therapy. With recent developments in systemic therapies, especially immunotherapies, metastatic renal cell carcinoma can be attacked aggressively in appropriate patients with a possibility of durable responses. Still, there is a remaining need for effective therapies for these patients.

Of all new cases of cancer in the United States, bladder cancer represents about 5 percent in men (fifth most common neoplasm) and 3 percent in women (eighth most common neoplasm). The incidence is increasing slowly, at the same time as an increasing elderly population. 1 In 1998, there were an estimated 54,500 cases, including 39,500 in men and 15,000 in women. The age-adjusted incidence in the USA is 32 per 100,000 for men and eight per 100,000 in women. The historical male/female ratio of 3:1 may be decreasingly related to smoking patterns in women. There were an estimated 11,000 deaths from bladder cancer in 1998 (7,800 in men and 3,900 in women). Bladder cancer incidence and mortality increase strongly with age and will be a growing problem as the population ages.

Most bladder cancers return in the bladder. Bladder cancer is managed with a combination of transurethral resection of the bladder (TUR) and intravesical chemotherapy or immunotherapy. The multifocal and recurrent nature of bladder cancer highlights the limitations of TUR. Most muscle-invasive cancers are not cured by TUR alone. Radical cystectomy and urinary debridement are the most effective means of removing the cancer, but carry an indescribable impact on urinary and sexual function. There continues to be a significant need for treatment modalities beneficial to bladder cancer patients.

An estimated 130,200 cases of colorectal cancer occurred in 2000 in the United States, comprising 93,800 cases of colon cancer and 36,400 of rectal cancer. Colorectal cancer is the third most common cancer in men and women. Incidence rates decreased significantly during 1992-1996 (-2.1% per year). Research indicates that these declines are due to increased screening and polyp removal, preventing polyp progression to invasive forces. There were an estimated 56,300 deaths (47,700 from colon cancer, 8,600 from rectal cancer) in 2000, which accounts for approx. 11% of all U.S. cancer deaths.

To date, surgery is the most common form of therapy for colorectal cancer and for cancers that have not spread, it is often curative. Chemotherapy, or chemotherapy plus radiation, is given before or after surgery to most patients where the cancer has deeply perforated the bowel wall or has spread to the lymph nodes. A permanent colostomy (creation of an abdominal opening for the elimination of body waste) is occasionally necessary for colon cancer and is rarely necessary for rectal cancer. There continues to be a need for effective diagnostic and treatment modalities for colorectal cancer.

There were an estimated 164,100 new cases of lung and bronchial cancer in 2000, accounting for 14% of all U.S. cancer diagnoses. The incidence rate of lung and bronchial cancer is significantly decreasing in men, from a high of 86.5 per 100,000 in 1984 to 70.0 in 1996.11 In the 1990s the rate of increase among women began to decline. 1 In 1996, the incidence rate in women was 42.3 per 100,000.

Lung and bronchial cancer caused an estimated 156,900 deaths in 2000, which accounts for 28% of all cancer deaths. During 1992-1996, mortality from lung cancer decreased significantly among men (-1.7% per year), while rates for women were still significantly increasing (0.9% per year). Since 1987, more women have died each year from lung cancer than breast cancer, which for over 40 years was the main cause of cancer death in women. Declining lung cancer incidence and mortality rates likely resulted from reduced smoking rates over the previous 30 years; however, declining smoking patterns among women lag behind those of men. Importantly, although declines in adult tobacco use have slowed, youth tobacco use is increasing again.

Treatment options for lung and bronchial cancer are determined by the type and stage of the cancer and include surgery, radiation therapy and chemotherapy. For many, localized cancer surgery is usually the treatment of choice. Because the disease has usually spread by the time it is discovered, radiation therapy and chemotherapy are often necessary in combination with surgery. Chemotherapy alone or combined with radiation is the treatment of choice for small cell lung cancer; on this regimen, a large percentage of patients experience remission, which in some cases is long lasting. However, there is an ongoing need for effective treatment and diagnostic approaches for lung and bronchial cancer.

An estimated 182,800 new invasive cases of breast cancer were expected to occur among women in the United States during 2000.1 In addition, approx. 1,400 new cases of breast cancer were expected to be diagnosed in men in 2000. After an increase of approx. 4% per years in the 1980s, breast cancer incidence rates in women leveled off in the 1990s to approx. 110.6 cases per 100,000.

In the U.S. alone, there were an estimated 41,200 deaths (40,800 women, 400 men) in 2000 due to breast cancer. Breast cancer ranks second among cancer deaths in women. According to the most recent data, death rates declined significantly during 1992-1996 with the largest reductions in younger women, both white and black. These reductions were likely the result of earlier detection and improved treatment.

Taking into account the medical circumstances and the patient's preferences, treatment of breast cancer may involve lumpectomy (local removal of the tumor) and removal of the lymph nodes under the arm; mastectomy (surgical removal of the breast) and removal of the lymph nodes under the arm; radiation therapy; chemotherapy; or hormone therapy. Often two or more methods are used in combination. A number of studies have shown that for early-stage disease, the long-term survival rate after lumpectomy plus radiotherapy is similar to the survival rate after modified radical mastectomy. Significant advances in reconstructive techniques provide many options for breast reconstruction after mastectomy. Recently, such reconstruction was performed at the same time as the mastectomy.

Local excision of ductal carcinoma in situ (DCIS) with sufficient amounts of surrounding normal breast tissue can prevent the local recurrence of DCIS. Radiation to the breast and/or tamoxifen can reduce the chance of DCIS occurring in the remaining breast tissue. This is important because DCIS, if left untreated, can progress to invasive breast cancer. Still, there are serious side effects or consequences to these treatments. There is therefore a need for effective breast cancer treatments.

There were an estimated 23,100 new cases of ovarian cancer in the USA in 2000. This accounts for 4% of all cancers among women and ranks second among gynecological cancers. During 1992-1996, ovarian cancer incidence rates were significantly decreasing. Following ovarian cancer, 14,000 deaths were estimated in 2000. Ovarian cancer causes more deaths than any other cancer of the female reproductive system.

Surgery, radiation therapy and chemotherapy are treatment options for ovarian cancer. Surgery usually involves the removal of one or both ovaries, the fallopian tube (salpingo-oophorectomy) and the uterus (hysterectomy). In some very early tumors, only the ovary involved will be removed, especially in younger women who want to have children. In advanced disease, an attempt is made to remove all intra-abdominal disease to improve the effect of chemotherapy. There continues to be an important need for effective treatment options for ovarian cancer.

There were an estimated 28,300 new cases of pancreatic cancer in the United States in 2000.1 Over the past 20 years, pancreatic cancer rates have decreased in men. Rates among women have remained roughly constant but may be beginning to decline. Pancreatic cancer caused an estimated 28,200 deaths in 2000 in the United States. During the last 20 years, there has been a slight but significant reduction in mortality rates among men (approximately -0.9% per year), while rates have increased slightly among women.

Surgery, radiation therapy and chemotherapy are treatment options for pancreatic cancer. These treatment options may prolong survival and/or alleviate symptoms in many patients but are unlikely to produce a cure for most. There is a significant need for additional therapeutic and diagnostic options for cancer. These include the use of antibodies, vaccines and small molecules as treatment modalities. In addition, there is also a need to use these modalities as research tools to diagnose, detect, monitor and further the state of the field in all areas of cancer treatment and research.

The therapeutic utility of monoclonal antibodies (mAbs) (G. Kohler and C. Milstein, Nature 256:495-497 (1975)) has been realized. Monoclonal antibodies have now been approved as therapies in transplantation, cancer, infectious disease, cardiovascular disease and inflammation. Different isotypes have different effector functions. Such differences in function are reflected in distinct 3-dimensional structures for the different immunoglobulin isotypes (P.M. Alzari et al, Annual Rev. Immunol., 6:555-580 (1988)).

Because mice are suitable for immunization and recognize most human antigens as foreign, mAbs against human targets with therapeutic potential have typically been of murine origin. However, murine mAbs have inherent disadvantages as human therapeutic agents. They require more frequent dosing as mAbs have a shorter circulating half-life in humans than human antibodies. More critically, the repeated administration of murine antibodies to the human immune system causes the human immune system to respond by recognizing the mouse protein as foreign and generating a human anti-mouse antibody (HAMA) response. Such a HAMA response can result in an allergic reaction and the rapid removal of the murine antibody from the system thereby renders the treatment with murine antibody worthless. To avoid such influences, attempts have been made to create human immune systems in mice.

Initial efforts hoped to create transgenic mice capable of responding to antigens with antibodies having human sequences (See Bruggemann et al, Proe. Nat'l. Acad. Sei. USA 86:6709-6713 (1989)), but was limited by the amount of DNA that could be kept stable by available cloning constituents. The use of yeast artificial chromosome (YAC) cloning vectors led the way to the introduction of large germline fragments of the human Ig locus into transgenic mammals. Substantially a majority of the human V, D and J region genes arranged in the same distribution as found in the human genome and the human constant regions were introduced into mice using YACs. One such transgenic mouse strain is known as xenomouse ("XenoMouse(r)") mice and is commercially available from Abgenix, Inc. (Fremont CA).

SUMMARY OF THE INVENTION

The present invention comprises a monoclonal antibody or an antigen-binding fragment thereof, as it comprises an antigen binding site that binds specifically to a PSCA protein comprising the amino acid sequence SEQ ID NO: 2 and where the monoclonal antibody comprises the amino acid sequence of the VH region of SEQ ID NO: 35 and the VL region of SEQ ID NO: 37.

The present invention further comprises antibody or antigen-binding fragment according to claim 1, the monoclonal antibody being produced by the hybridoma deposited under American Type Culture Collection (ATCC) Accession No. PTA-6699.

The present invention further comprises antibody or fragment according to claim 1 or 2, the fragment being a Fab, F(ab')2, Fv or scFv fragment.

The present invention further comprises antibody or fragment according to any one of claims 1, 2 or 3, the antibody or fragment being linked to a detectable marker.

The present invention further comprises hybridoma, in that it produces the monoclonal antibody according to claim 1 or 2.

The present invention further comprises the hybridoma according to claim 5, the hybridoma being deposited under American Type Culture Collection (ATCC) Accession No. PTA-6699.

The present invention further comprises polynucleotide, in that it codes for a sequence comprising a light chain variable region and a heavy chain variable region of the antibody according to any one of claims 1, 2 or 3.

The present invention further comprises the oolynucleotide according to claim 7, in that it codes for the light chain and the heavy chain of the antibody according to any one of claims 1, 2 or 3.

The present invention further comprises vector, as it comprises the polynucleotide according to claim 7 or claim 8.

The present invention further comprises the cell, as it has been transfected with the vector according to claim 9.

The present invention further comprises composition, as it comprises the antibody, or Fab, F(ab')2, Fv or scFv fragment thereof according to any one of claims 1 to 4.

The present invention further comprises a method for producing an antibody according to claim 1 or an antigen-binding fragment thereof, as it comprises:

cultivation of a cell, where:

(a) the cell has been transfected with a polynucleotide encoding a sequence comprising a light chain variable region of the antibody according to claim 1 and a polynucleotide encoding a sequence comprising a heavy chain variable region of the antibody according to claim 1; or (b) the cell according to claim 10;

whereby the antibody or an antigen-binding fragment thereof is produced.

Accordingly, the invention provides antibodies, as well as binding fragments thereof, which bind to PSCA proteins.

It is possible to detect the presence and status of PSCA polynucleotides and proteins in various biological samples, as well as perform methods to identify cells expressing PSCA. The invention further provides compositions that can be used for the treatment of cancers expressing PSCA such as cancers in tissues listed in Table I, including therapies directed at inhibiting the transcription, translation, processing or function of PSCA as well as cancer vaccines.

BRIEF DESCRIPTION OF THE FIGURES

Figure 1. The cDNA and amino acid sequence of PSCA (also designated "PSCA-v.1" or "PSCA variant 1") is shown in Figure IA. The starting methionine is underlined. The open reading frame spans nucleic acid 18-389 including the stop codon.

The cDNA and amino acid sequence of PSCA variant 2 (also designated "PSCA-v.2") is shown in Figure 1B. The codon for the starting methionine is underlined. The open reading frame spans nucleic acid 56-427 including the stop codon.

The cDNA and amino acid sequence of PSCA variant 3 (also designated "PSCA-v.3") is shown in Figure IC. The codon for the starting methionine is underlined. The open reading frame spans nucleic acid 423-707 including the stop codon.

The cDNA and amino acid sequence of PSCA variant 4 (also designated "PSCA-v.4") is shown in Figure ID. The codon for the starting methionine is underlined. The open reading frame spans nucleic acid 424-993 including the stop codon.

The cDNA and amino acid sequence of PSCA variant 5 (also designated "PSCA-v.5") is shown in Figure IE. The codon for the starting methionine is underlined. The open reading frame spans nucleic acid 910-1479 including the stop codon.

The cDNA and amino acid sequence of PSCA variant 6 (also designated "PSCA-v.6") is shown in Figure IF. The codon for the starting methionine is underlined. The open reading frame spans nucleic acid 83-427 including the stop codon.

Figure 1 G. SNP variants of PSCA-v.2, PSCA-v.7 through v. 18. The PSCA-v.7 through v. 18 proteins have 123 amino acids. Variants PSCA-v.7 through v. 18 are variants with a single nucleotide difference from PSCA-v.2 and encode the same protein as v.2. Although these SNP variants are shown separately, they can also occur in any combination and in any of the transcript variants listed above in Figures 1A through 1F. Figure 1H. SNP variants of PSCA-v.4, PSCA-v.l9 through v.30. The PSCA v.l9 through v.30 proteins have 189 amino acids. Variants PSCA-v.19 through v.30 are variants with a single nucleotide difference from PSCA-v.4. PSCA-v.9, v. 10, v.l 1, v.24 and v.25 proteins differ from PSCA-v.l by one amino acid. PSCA-v.23, v.28, v.29 and v.30 encode the same protein as v.4. Although these SNP variants are shown separately, they can also occur in any combination and in any of the v.3 and v.4 transcript variants. Figure II. Expression of PSCA variants. (ll(a)) Primers were designed to differentiate between variants PSCA-v.l/v.2/v.4, PSCA-v.3 and PSCA-v.5. Primers A and B, indicated by small arrows above exons in the figure result in a PCR product of 425 bp for PSCA-v. l/v.2/v.4, a 300 bp PCR product for PSCA-v.3 and a 910 bp PCR product for PSCA-v.5. (ll(b)) First strand cDNA was prepared from normal bladder, brain, heart, kidney, liver, lung, prostate, spleen, skeletal muscle, testis, pancreas, colon, stomach, collections of prostate cancer, bladder cancer, kidney cancer, colon cancer, lung cancer , ovarian cancer, breast cancer, cancer metastasis and pancreatic cancer. Normalization was performed by PCR using primers for actin. Semiquantitative PCR using the variant-specific primers was performed at 30 cycles of amplification. Results show expression of PSCA-v.5 mainly in breast cancer, cancer metastasis and pancreatic cancer and at a lower level in colon cancer and lung cancer. PSCA-v.l/v.2/v.4 PCR product was detected in prostate cancer, bladder cancer, kidney cancer, colon cancer, lung cancer, ovarian cancer, breast cancer, cancer metastasis and pancreatic cancer. Among normal tissues, PSCA-v. l/v.2/v.4-PCR product detected only in prostate, stomach and at a lower level in kidney and lung, while PSCA-v.5 was not detected in any normal tissues. PSCA-v.3-PCR-detected product was not detected in any of the samples tested. Figure 1J. Expression of PSCA-v.4 and PSCA-v.5. 1 J(a) Primers were designed to differentiate between PSCA-v.4 and PSCA-v.5 as indicated by the arrows labeled B and C in the figure. Primers specific for PSCA-v.4 lead to a PCR product of 460 bp, while primers specific for PSCA-v.5 lead to a PCR product of 945 bp in size. 1 J(b) First-strand cDNA was prepared from normal bladder, brain, heart, kidney, liver, lung, prostate, spleen, skeletal muscle, testis, pancreas, colon, stomach, prostate cancer collections, bladder cancer and multi-xenograft collection (prostate cancer- , kidney cancer and bladder cancer xenografts). Normalization was performed by PCR using primers for actin. Semiquantitative PCR using the variant-specific primers was performed at 30 cycles of amplification. Results show expression of PSCA-v.4 in prostate cancer, bladder cancer and multi-xenograft collection, normal kidney and prostate. PSCA-v.5 was detected only in normal prostate and bladder cancer. Figure 2. Amino acid sequences of PSCA antibodies. Figure 2A. Amino acid sequence of Hal-4,117-VH. Underlined is the constant region of the heavy chain. Figure 2B. Amino acid sequence of Hal-4,117-VL. Underlined is the constant region of the light chain. Figure 2C. Amino acid sequence of Hal- 4,120-VH. Figure 2D. Amino acid sequence of Hal-4,120-VL. Underlined is the constant region of the light chain. Figure 2E. Amino acid sequence of Hal-5.99-VH. Underlined is the constant region of the heavy chain. Figure 2F. Amino acid sequence of Hal-5.99-VL. Underlined is the constant region of the light chain. Figure 2 G. Amino acid sequence of Hal-4.121-VH. Underlined is the constant region of the heavy chain. Figure 2H. Amino acid sequence of Hal-4.121-VL-c.5. Underlined is the constant region of the light chain. Figure 21. Amino acid sequence of Hal-4.121-VL-c.26. Underlined is the constant region of the light chain. Figure 2J. Amino acid sequence of Hal-1.16-VH. Underlined is the constant region of the heavy chain. Figure 2K. Amino acid sequence of Hal-1.16-VL. Underlined is the constant region of the light chain. Figure 2L. Amino acid sequence of Hal-4.5-VH. Underlined is the constant region of the heavy chain. Figure 2M. Amino acid sequence of Hal-4.5-VL. Underlined is the constant region of the light chain. Figure 2N. Amino acid sequence of Hal-4.40-VH. Underlined is the constant region of the heavy chain. Figure 20. Amino acid sequence of Hal-4.40-VL. Underlined is the constant region of the light chain. Figure 2P. Amino acid sequence of Hal-4.37-VH. Underlined is the constant region of the heavy chain. Figure 2Q. Amino acid sequence of Hal-4.37-VL. Underlined is the constant region of the light chain. Figure 2R. Amino acid sequence of Hal-1.43-VH. Underlined is the constant region of the heavy chain. Figure 2S. Amino acid sequence of Hal-1.43-VL. Underlined is the constant region of the light chain. Figure 2T. Amino acid sequence of Hal-1,152-VH. Underlined is the constant region of the heavy chain. Figure 2U. Amino acid sequence of Hal-1,152-VL. Underlined is the constant region of the light chain. Figure 3. Nucleotide and amino acid sequences for PSCA antibodies. Figure 3A. The cDNA and amino acid sequence of Hal-4.117-VH. Underlined is the constant region of the heavy chain. Figure 3B. The cDNA and amino acid sequence of Hal-4.117-VL. Underlined is the constant region of the light chain. Figure 3C. The cDNA and amino acid sequence of Hal-4,120-VH. Underlined is the constant region of the heavy chain. Figure 3D. The cDNA and amino acid sequence of Hal-4,120-VL. Underlined is the constant region of the light chain. Figure 3E. The cDNA and amino acid sequence of Hal-5.99-VH. Underlined is the constant region of the heavy chain. Figure 3F. The cDNA and amino acid sequence of Hal-5.99-VL. Underlined is the constant region of the light chain. Figure 3 G. The cDNA and amino acid sequence of Hal-4,121-VH. Underlined is the constant region of the heavy chain. Figure 3H. The cDNA and amino acid sequence of Hal-4.121-VL-c.5. Underlined is the constant region of the light chain. Figure 31. The cDNA and amino acid sequence of Hal-4.121-VL-c.26. Underlined is the constant region of the light chain. Figure 3J. The cDNA and amino acid sequence of Hal-1.16-VH. Underlined is the constant region of the heavy chain. Figure 3K. The cDNA and amino acid sequence of Hal-1.16-VL. Underlined is the constant region of the light chain. Figure 3L. The cDNA and amino acid sequence of Hal-4.5-VH. Underlined is the constant region of the heavy chain. Figure 3M. The cDNA and amino acid sequence of Hal-4.5-VL. Underlined is the constant region of the light chain. Figure 3N. The cDNA and amino acid sequence of Hal-4.40-VH. Underlined is the constant region of the heavy chain. Figure 30. The cDNA and amino acid sequence of Ha 1-4.40-VL. Underlined is the constant region of the light chain. Figure 3P. The cDNA and amino acid sequence of Hal-4.37-VH. Underlined is the constant region of the heavy chain. Figure 3Q. The cDNA and amino acid sequence of Hal-4.37-VL. Underlined is the constant region of the light chain. Figure 3R. The cDNA and amino acid sequence of Hal-1.43-VH. Underlined is the constant region of the heavy chain. Figure 3S. The cDNA and amino acid sequence of Hal-1.43-VL. Underlined is the constant region of the light chain. Figure 3T. The cDNA and amino acid sequence of Hal-1,152-VH. Underlined is the constant region of the heavy chain. Figure 3U. The cDNA and amino acid sequence of Hal-1,152-VL. Underlined is the constant region of the light chain. Figure 4. Array of PSCA antibodies to germlin V-D-J sequences. Figure 4A. Alignment of Hal-4.117-VH (SEQ ID NO: 13) to human VH4-31. Figure 4B. Alignment of Hal-4.117-VL (SEQ ID NO: 14) to human LI9. Figure 4C. Alignment of Hal-4.120-VH (SEQ ID NO: 15) to human VH4-31. Figure 4D. Alignment of Hal-4.120-VL (SEQ ID NO: 16) to human 02. Figure 4E. Alignment of Hal-5.99-VH (SEQ ID NO: 17) to human VH4-34. Figure 4F. Alignment of Hal-5.99-VL (SEQ ID NO: 18) to human A27. Figure 4 G. Alignment of Hal-4.121-VH (SEQ ID NO: 19) to human-VH4-34. Figure 4H. Alignment of Hal-4.121-C.5-VL (SEQ ID NO: 20) to human 08. Figure 41. Alignment of Hal-4.121-C.26-VL (SEQ ID NO: 21) to human A3. Figure 4J. Alignment of Hal-1.16-VH (SEQ ID NO: 22) to human VH6-1. Figure 4K. Alignment of Hal-1.16-VL (SEQ ID NO: 23) to human B3. Figure 4L. Alignment of Hal-4.37-VH (SEQ ID NO: 28) to human VH4-31. Figure 4M. Alignment of Hal-4.37-VL (SEQ ID NO: 29) to human 02. Figure 5. Expression of PSCA protein in recombinant murine, rat and human cell lines. The indicated murine, rat and human cell lines were infected with retroviruses carrying the human PSCA cDNA and a neomycin resistance gene or control virus with only the neomycin resistance gene. Stable recombinant cell lines were selected in the presence of G418. PS CA expression was determined by FACS labeling with lG8 anti-PSCA MAb (5 µg/ml). Shown is the FACS profile of each cell line demonstrating a fluorescent shift only in the PSCA-infected line indicative of cell surface PSCA expression. These lines are useful in MAb development as immunogens, MAb screening reagents and in functional assays. Figure 6. Purification of PSCA protein from E.Coli. E. coli strain BL21-pLysS was transformed with pET-21b vector encoding amino acids 21-94 of the PSCA cDNA. PSCA protein was expressed by induction of log-phase cultures with IPTG and purified by affinity chromatography from either the soluble or insoluble fractions of the lysed bacteria. Shown are SDS-PAGE Coomasie blue-stained gels of the eluted fractions. This protein is useful as an MAb and pAb immunogen and as an antibody screening reagent. Figure 7. Purification of recombinant glycosylated PSCA protein expressed from 293T cells. 293T cells were transfected with the psecTag2 vector carrying a PSCA cDNA encoding amino acids 28-100. A stable recombinant PSCA-secreting cell line was generated by drug selection with hygromycin B. PSCA protein present in conditioned culture medium was purified by affinity chromatography using lG8-MAbet. Shown is a Coomasie blue-stained SDS-PAGE gel of the low-pH eluted fractions. The broad molecular weight smear ("the molecular weight smear") of the protein demonstrates glycosylation of recombinant PSCA protein seen in endogenously expressed PSCA. Figure 8. Purification of GST-PSCA protein from E.Coli. E. co//' strain BL21-DE3 was transformed with pGEX-2T encoding amino acids 18-98 of PSCA fused to glutathione-S-transferase (GST). GST-PSCA protein was induced with isopropyl-beta-D-thiogalactopyranoside (IPTG) from log-phase cultures and purified from lysed bacteria by glutathione-agarose matrix affinity chromatography. Shown is an SDS-PAGE Coomasie blue-stained gel of the glutathione-eluted fractions containing GST-PSCA. Indicated are the intact GST-PSCA fusion protein and a minor degradation product containing GST. This protein is useful as an MAb and pAb immunogen and as an Ab screening reagent. Figure 9. Screening for human PSCA antibodies by FACS. Antibody concentration of supernatants was determined by ELISA. 50ul/well (exact) was added to 96-well FACS plates and serially diluted. PSCA-expressing cells were added (endogenous or recombinant, 50,000 cells/well) and the mixture incubated at 4°C for two hours. After incubation, the cells were washed with FACS buffer and further incubated with 100ul of detection antibody (anti-hlgG-PE) for 45 minutes at 4°C. At the end of incubation, cells were washed with FACS buffer, fixed with formaldehyde and analyzed using FACScan. Data were analyzed using CellQuest Pro software. Solid histograms represent data from negative control cells and open histograms indicate data from PSCA-positive cells. Figure 10. Relative affinity ranking of PSCA-MAbs by FACS. 21 serial 1:2 dilutions of each PSCA antibody were incubated with SW780 cells (50,000 cells per well) overnight at 4°C (final MAb concentrations ranged from 40nM to 0.038pM). At the end of the incubation, cells were washed and incubated with anti-hlgG-PE detection antibody. After washing the unbound secondary antibodies, the cells were analyzed by FACS and the fluorescence intensity of each spot was averaged using CellQuest Pro software. Affinity was calculated with Graphpad Prisme software using a sigmoidal dose response (variable slope) equation. A representative FACS analysis showing the binding titration for PSCA-MAb-4.121 is presented in the figure. Figure 11. Expression of murine and cynomolgus monkey PSCA in 293T cells and recognition by antihuman PSCA MAbs. 293T cells were transiently transfected with pcDNA3.1 vector carrying the murine PSCA cDNA, the simian PSCA cDNA or with empty vector (neo). Two days following transfection, cells were harvested and labeled with either human anti-PSCA-MAb-Hal-4,117 or murine MAb-1G8 (5 µg/ml). Shown is the FACS profile demonstrating that MAb-Hal-4,117 raised against the human PSCA protein binds to both murine and simian PSCA protein expressed in 293T cells. The murine lG8-MAb binds to simian PSCA but not to murine PSCA. Such results demonstrate the ability to select MAbs that cross-react with antigen from other species. Cross-reacting MAbs will be useful for studying expression and toxicity in those species. Figure 12. Internalization of PSCA after incubation with MAb-4.121. PSCA-MAb-4,121 was incubated with PC3-PSCA cells at 4°C for 90 min to allow binding of the antibodies to the cell surface. The cells were then split into two aliquots and incubated at either 37°C (to allow antibody internalization) or 4°C (no internalization control). After incubation at 37°C or 4°C, the remaining PSCA-MAb-4,121 bound to the cell surface was removed with an acid wash. Subsequent permeabilization and incubation with a secondary detection antibody allowed detection of internalized PSCA-MAb-4,121. Cells were analyzed using FACS or observed under the fluorescence microscope. Approximately 30% of PSCA-MAb-4,121 was internalized after incubation at 37°C for two hours. Figure 13. Antibodies against PSCA-mediated saporin-dependent killing in PSCA-expressing cells. B300.19 cells (750 cells/well) were seeded in a 96-well plate on day 1. The following day, an equal volume of medium containing a 2X concentration of the indicated primary antibody together with a 2-fold excess of antihuman (Hum-Zap) or anti-goat (Goat-Zap) polyclonal antibody conjugated with saporin toxin (Advanced Targeting Systems, San Diego, CA) was added to each well. The cells were allowed to incubate for 5 days at 37 degrees C. At the end of the incubation period, MTS (Promega) was added to each well and incubation continued for an additional 4 hours. The OD at 450 nM was determined. The results in Figure 13(A) show that PSCA antibodies HA1-4.121 and HA1-4.117 mediated saporin-dependent cytotoxicity in B300.19-PSCA cells while a control, nonspecific IgGi antibody, had no effect. The results in Figure 13(B) show that the addition of a secondary saporin conjugate antibody that does not recognize human Fc fails to mediate cytotoxicity. Figure 14. Complement-mediated cytotoxicity of PSCA-Mabs. PSCA antibodies (0-50 µg/ml) were diluted with RHB buffer (RPMI1640, Gibco Life Technologies, 20 mM HEPES). B300.19-PSCA-expressing cells were washed in RHB buffer and resuspended at a density of 10 6 cells/ml. In a typical experiment, 50 µl of PSCA antibody, 50 µl of diluted complement rabbit serum (Cedarlane, Ontario, Can) and 50 µl of a cell suspension were added together in a flat-bottomed tissue culture 96-well plate. The mixture was incubated for 2 hours, at 37°C in a 5% CO 2 incubator to facilitate complement-mediated cell lysis. 50 µl of Alamar Blue (Biosource Intl. Camarillo, CA) was added to each well and incubation continued for an additional 4-5 hours at 37°C. The fluorescence in each well was read using a 96-well fluorometer with excitation at 530 nm and emission at 590 nm. The results show that PSCA antibodies having an IgGi-(HAl-4.121) or an IgG2 isotype (HAI-5.99.1) but not an IgG4 isotype (HAI-6.46) were able to mediate complement-dependent lysis of target cells . Figure 15. Formation of F(ab')2 fragments of MAb-Hal-4.121 by pepsin digestion. 20 mg of MAb-Hal-4,121 in 20 mM sodium acetate buffer pH 4.5 was incubated with and without immobilized pepsin (Pierce. Rockford IL) for the indicated times. Intact MAb and digested Fc fragments were removed by protein A chromatography. Shown is an SDS-PAGE Coomasie-stained gel of intact undigested unreduced MAb, unreduced aliquots of digested material taken at the indicated times, and a reduced sample of the final digested F(ab') 2 product. Figure 16. Binding of recombinant anti-PSCA human mAb to PSCA by flow cytometry. (16A) 293T cells were transfected with expression constructs encoding the heavy and light chains of anti-PSCA human mAb. Supernatant was collected after 48 hours and examined for binding to PSCA. (16B) Anti-PSCA human mAb was purified from hybridoma supernatant and used for PSCA binding assay. PSCA binding was tested as follows. PC3 parental or PC3-PSCA cells were incubated with the anti-PSCA human mAbs described above for 30 min on ice. The cells were washed and incubated with PE-conjugated antihuman Ig for 30 minutes on ice. The cells were washed and then examined by flow cytometry. Figure 17. Detection of PSCA protein by immunohistochemistry. Expression of PSCA protein in tumor samples from cancer patients was detected using the antibody HAI-4,117. Formalin-fixed, paraffin-embedded tissue was cut into 4 micron sections and placed on glass slides. The sections were awoken, rehydrated, and treated with antigen retrieval solution (Antigen Retrieval Citra Solution; BioGenex, 4600 Norris Canyon Road, San Ramon, CA, 94583) at high temperature. Sections were then incubated in fluorescein-conjugated human monoclonal anti-PSCA antibody, Hal-4,117, for 16 hours at 4°C. Slides were washed three times in buffer and further incubated with rabbit antifluorescein for 1 hour and, after washing in buffer, immersed in DAKO EnVision+™ peroxidase-conjugated goat antirabbit immunoglobulin secondary antibody (DAKO Corporation, Carpenteria, CA) for 30 minutes. The sections were then washed in buffer, developed using the DAB kit (Sigma Chemicals), counterstained using hematoxylin and analyzed by brightfield microscopy. The results show expression of PSCA in the tumor cells from prostate adenocarcinoma (panel A, panel B), bladder transitional carcinoma (panel C) and pancreatic ductal adenocarcinoma (panel D). These results indicate that PSCA is expressed in human cancer and that antibodies directed against this antigen are useful as diagnostic reagents. Figure 18. PSCA-MAb-Hal-4,120 inhibits the growth of subcutaneous prostate cancer xenografts. LAPC-9AI tumor cells (2.0 x 10 6 cells) were injected subcutaneously into male SCID mice. Mice were randomized into groups (n=10 in each group) and treatment initiated intraperitoneally (i.p.) on day 0 with HAI-4,120 or isotype MAb control as indicated. Animals were treated twice weekly with a total of 7 doses until study day 28. Tumor growth was monitored using caliper measurements every 3 to 4 days as indicated. The results show that human anti-PSCA monoclonal antibody Hal-4,120 significantly inhibited the growth of human prostate cancer xenografts implanted subcutaneously in SCID mice (p< 0.05). Figure 19. PSCA-MAb-Hal-5.99 inhibits the growth of established prostate cancer xenografts in SCID mice. LAPC-9AI tumor cells (2.0 x 10 6 cells) were injected subcutaneously into male SCID mice. When tumor volume reached 50 mm3, mice were randomized into groups (n=10 in each group) and treatment initiated intraperitoneally (i.p.) with HA1-5.99.1 or isotype MAb control as indicated. Animals were treated twice weekly with a total of 5 doses until study day 14. Tumor growth was monitored using caliper measurements every 3 to 4 days as indicated. The results show that fully human anti-PSCA monoclonal antibody Hal-5.99 significantly inhibited the growth of established androgen-independent human prostate cancer xenografts implanted subcutaneously in SCID mice (1X0.05). Figure 20. PSCA-MAb-HAl-4,121 inhibits the growth of established androgen-dependent human prostate cancer xenografts. LAPC-9AD tumor cells (2.5 x 10 6 cells) were injected subcutaneously into male SCID mice. When tumor volume reached 40 mm3, mice were randomized into groups (n=10 in each group) and treatment was initiated intraperitoneally (i.p.) with increasing concentrations of HA1-4,121 or isotype MAb control as indicated. Animals were treated twice weekly with a total of 7 doses until study day 21. Tumor growth was monitored using caliper measurements every 3 to 4 days as indicated. The results of this study demonstrated that HA1-4,121 inhibited the growth of established subcutaneous human androgen-dependent prostate xenografts in SCID mice. The results were statistically significant for the 300 ug dose group on days 14, 17 and 21 (p< 0.05, Kruskal-Wallis test, two-sided with a=0.05) and for the 700 ug dose group on days 10, 14, 17 and 21 (p< 0.05, Kruskal-Wallis test, two-sided with a=0.05). Figure 21. Patient-derived androgen-dependent LAPC-9AD tumor cells (2.0 x 10 6 cells) were injected into the dorsal lobes of prostates of male SCID mice. The tumors were allowed to grow for approximately 10 days at which time the mice were randomized into groups. Treatment with human 500 µg of HA1-4117, HA1-4121 or isotype control MAb was initiated 10 days after tumor implantation. Antibodies were delivered intraperitoneally twice weekly with a total of 7 doses. Four days after the last dose, animals were sacrificed and primary tumors were excised and weighed. The results show that human anti-PSCA monoclonal antibodies Hal-4,121 (p< 0.01) and Hal-4,117 (p< 0.05) significantly inhibited the growth of LAPC-9AD prostate cancer xenografts orthotopically implanted in SCID mice. Figure 22. PSCA-MAb-HAl-4,121 prolongs the survival of SCID mice with established orthotopic human androgen-dependent prostate tumors. Patient-derived, androgen-dependent LAPC-9AD tumor cells (2.0 x 10 6 cells) were injected into the dorsal lobes of prostates of male SCID mice. The tumors were allowed to grow for approximately 9 days at which time the mice were randomized into groups. The animals randomized into the survival groups include 11 mice in the isotype MAb control and 12 mice in the HAI-4,121-treated group. Animals were treated i.p. with 1000 µg of Hal-4,121 or 1000 µg of isotype control MAb twice weekly for a total of 9 doses. The results demonstrated that HAI-4,121 significantly (log-rank test: p<0.01) prolonged the survival of SCID mice bearing human androgen-dependent prostate tumors. Two mice in the HA1-4,121-treated group remained free of overt tumors 110 days after the last treatment. Figure 23. Improved inhibition of prostate tumor growth with HAI-4.21 and taxotere combination therapy. LAPC-9AI tumor cells (2 x 10 6 cells per animal) were injected subcutaneously into male SCID mice. When the tumor volume reached 65 mm3, animals were randomized and assigned to four different groups (n=10 in each group) as indicated. At the beginning of day 0, Hal-4,121 or isotype MAb control was administered i.p. twice weekly in a dose of 500 ug with a total of 6 doses. The last dose was given on day 17. Taxotere was given intravenously at a dose of 5 mg/kg on days 0, 3 and 7. Tumor growth was monitored every 3-4 days using caliper measurements. The results of this study demonstrate that HAI-4,121 as a single agent inhibited the growth of androgen-independent prostate xenografts in SCID mice by 45% when compared to control antibody treatment alone at day 28 (ANOVA/Tukey test: p<0.05). Administration of the isotype MAb control plus taxotere inhibited tumor growth by 28% when compared to control antibody treatment alone, which was not statistically significant. Administration of HA1-4,121 in combination with taxotere improved efficacy and resulted in a 69% inhibition of tumor growth when compared to control antibody alone (ANOVA/Tukey test: p<0.01). A statistically significant difference was also demonstrated when the HAl-4,121 plus taxotere combination group was compared to either the HAI-4,121 or isotype MAb control plus taxotere groups (ANOVA/Tukey test: p<0.05) . Figure 24. Human PSCA-MAbs inhibit the growth of pancreatic cancer xenografts in SCID mouse/human HPAC. Pancreatic cancer cells (2 x 106/mouse) were injected subcutaneously into immunodeficient ICR-SCID mice (Taconic Farm, Germantown, NY). The mice were randomized into groups (n= 10 animals/group) and treatment with the indicated human PSCA monoclonal antibody started on the same day. Antibodies (500 mg/mouse) were delivered intraperitoneally twice weekly for a total of 8 doses. The results demonstrated that human anti-PSCA monoclonal antibodies Hal-4.121, Hal-4.117 and Hal-1.16 significantly inhibited the growth of human pancreatic cancer xenografts subcutaneously implanted in SCID mice. Statistical analyzes were performed using a t-test (two-tailed, a=0.05). Figure 25. PSCA-MAb-HAl-4,121 inhibits the growth of pancreatic tumors implanted orthotopically in SCID mice. HPAC cells (3.0 x 10 6 cells) were implanted orthotopically into the pancreatic islet ("the pancreata") of SCID mice. Mice were randomly assigned to three groups (n=9 in each group) as indicated. Treatment with HAI-4,121 (250 µg or 1000 µg) or isotype MAb control (1000 µg) was initiated on the day of implantation. Antibodies were administered i.p. twice weekly with a total of 10 doses. Thirteen days after the last dose, animals were sacrificed and primary tumors were excised and weighed. The results of this study demonstrated that HAI-4,121 significantly inhibited the orthotopic growth of human pancreatic cancer xenografts in SCID mice at both dose levels examined. Treatment with 250 µg and 1000 µg AGS-PSCA inhibited tumor growth by 66% and 70% respectively (Kruskal-Wallis/Tukey test respectively: p<0.01 and p<0.01). Figure 26. PSCA-MAb-HAl-4.121 inhibits matatases. At autopsy, visible metastases in lymph nodes and scattered organs were observed in the control antibody-treated group. No visible metastases were observed in both HA1-4,121-treated groups. Lymph nodes, lungs and livers were removed from all animals and examined histologically for the presence of metastatic tumor. Sections from the lungs and lymph nodes removed from each animal were labeled for human cytokeratin and the number of metastases determined microscopically. The results of the histological analysis demonstrated a significant reduction in lymph node (LN) metastases in animals treated with HAI-4,121 (p= 0.0152 detected by Fisher's exact test). The incidence of metastasis and invasion was also significantly reduced in animals treated with both concentrations of HA 1-4,121 (p= 0.0152 detected by Fisher's exact test). The number of lung metastases was significantly reduced in mice treated with only the 1.0 mg dose of HAI-4,121 (p= 0.0498 detected by Fisher's exact test). Figure 27. Human PSCA-MAbs inhibit the growth of SW780 bladder tumors in SCID mice. Human SW780 bladder cancer cells (2 x 106/mouse) were injected subcutaneously into immunodeficient ICR-SCID mice (Taconic Farm, Germantown, NY). The mice were randomized into groups (n= 10 animals/group) and treatment with the specified human PSCA-MAb started on the same day. Antibodies (250 mg/mouse) were delivered intraperitoneally twice weekly with a total of 7 doses. The results demonstrated that HA1-4.117 (p= 0.014), HA1-4.37 (p= 0.0056), HA1-1.78 (p= 0.001), Hal-5.99 (p= 0.0002) and HA1-4.5 (p= 0 .0008) significantly inhibited the growth of SW780 bladder tumors implanted subcutaneously in SCID mice. Statistical analyzes were performed using a t-test (two-tailed, a=0.05).

DETAILED DESCRIPTION OF THE INVENTION

Section overview

I. ) Definitions

II. ) PSCA polynucleotides n.A.) Applications of PSCA polynucleotides

II. A. 1.) Monitoring of genetic abnormalities

II. A. 2.) Antisense Embodiments

II. A.3.) Primers and primer pairs

II. A.4.) Isolation of PSCA-encoding nucleic acid molecules

II.A.5.) Recombinant nucleic acid molecules and host vector systems

in.) PSCA-related proteins

III.A.) Motif-bearing protein embodiments

UIB.) Expression of PSCA-related proteins

m.C.) Modifications of PSCA-related proteins

ULD.) Applications of PSCA-related proteins

IV. ) PSCA antibodies

V. ) PSCA cellular immune responses

WE. ) PSCA transgenic animals

VII. ) Procedures for detection of PSCA

VIII. ) Procedures for monitoring the status of PSCA-related genes and their products IX. ) Identification of molecules that interact with PSCA

X. ) Therapeutic methods and compositions

X.A.) Anticancer vaccines

X.B.) PSCA as a target for antibody-based therapy

X.C.) PSCA as a measure of cellular immune responses

X. C. 1. Minigene vaccines

X.C.2. Combinations of CTL peptides with helper peptides

X. C. 3. Combinations of CTL peptides with T-cell priming agents

X.C.4. Vaccine compositions comprising DC pulsed with CTL and/or HTL peptides

X.D.) Adoptive immunotherapy

X.E.) Administration of vaccines for therapeutic or prophylactic purposes

XE.) Diagnostic and prognostic embodiments of PSCA

XII. ) Inhibition of PS CA protein function

XE. A.) Inhibition of PSCA by intracellular antibodies

XII.B.) Inhibition of PSCA with recombinant proteins

Xn.C.)Inhibition of PSCA transcription or translation

XH.D.) General considerations for therapeutic strategies

XIII. ) Identification, characterization and application of modulators of PSCA

XIV. ) RN Ai and therapeutic application of small interfering RNAs (siRNAs)

XV. ) Putty/articles for manufacture

I.) Definitions:

Many of the techniques and procedures described or referenced herein are well understood and commonly employed using conventional methodology by those skilled in the art, such as, for example, the widely used molecular cloning procedures described in Sambrook et al, Molecular Cloning: A Laboratory Manual 2nd. edition (1989) Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. As appropriate, procedures involving the use of commercially available putties and reagents are generally performed according to manufacturer-defined protocols and/or parameters unless otherwise stated.

The terms "advanced prostate cancer", "locally advanced prostate cancer", "advanced disease" and "locally advanced disease" mean prostate cancer that has spread through the prostatic capsule and shall include American Urological Association stage C disease ") (AUA) system, stage Cl - C2 disease by the Whitmore-Jewett system and stage T3 - T4 and N+ disease by the TNM (tumour, node, metastasis) system. In general, surgery is not recommended for patients with locally advanced disease and these patients mainly have less favorable outcomes compared to patients who have clinically localized (organ-confined) prostate cancer. Locally advanced disease is clinically identified by obvious evidence of invasion beyond the lateral border of the prostate or asymmetry or invasion above the prostate base. Locally advanced prostate cancer is currently diagnosed pathologically after radical prostatectomy if the tumor invades or penetrates the prostatic capsule, extends into the surgical margin, or invades the seminal vesicles.

"Altering the native glycosylation pattern" is intended for purposes herein to mean the removal of one or more carbohydrate moieties found in native sequence PSCA (either by removal of the underlying glycosylation site or by removal of the glycosylation by chemical and/or enzymatic means) and/or addition of one or more glycosylation sites not present in the natural sequence of PSCA. In addition, the term includes qualitative changes in the glycosylation of the natural proteins, involving a change in the nature and proportions of the various carbohydrate groups present.

The term "analog" denotes a molecule that is structurally similar or shares similar or equivalent characteristics with another molecule (eg, a PSCA-related protein). For example, an analog of a PSCA protein may be specifically bound by an antibody or T cell that specifically binds to PSCA.

The term "antibody" is used in the broadest sense unless clearly indicated otherwise. Therefore, an "antibody may be naturally occurring or man-made such as monoclonal antibodies produced by conventional hybridoma technology. Anti-PSCA antibodies include monoclonal and polyclonal antibodies as well as fragments containing the antigen binding domain and/or one or more complementarity determining regions of these antibodies. As used herein, the term "antibody" refers to any form of antibody or fragment thereof that specifically binds PSCA and/or exhibits the desired biological activity and specifically covers monoclonal antibodies (including full-length monoclonal antibodies), polyclonal antibodies, multispecific antibodies (e.g. bispecific antibodies) and antibody fragments as long as they specifically bind PSCA and/or exhibit the desired biological activity. Any specific antibody may be used with the methods and compositions provided herein. Thus, in one embodiment, the term "antibody" includes a molecule comprising at least one variable region from a light chain immunoglobulin molecule and at least one variable region from a heavy chain molecule which in combination forms a specific binding site for the target antigen. In one embodiment, the antibody is an IgG antibody. For example, the antibody is an IgG1, IgG2, IgG3 or IgG4 antibody. The antibodies useful in the preliminary methods and compositions can be formed in cell culture, in subjects or in various animals, including but not limited to cows, rabbits, goats, mice, rats, hamsters, guinea pigs, sheep, dogs, cats, monkeys, chimpanzees. Therefore, in one embodiment, an antibody of the present invention is a mammalian antibody. Professional techniques can be used to isolate an initial antibody or to generate variants with altered specificity or voracious characteristics. Such techniques are routine and well known in the field. In one embodiment, the antibody is produced by recombinant means known in the art. For example, a recombinant antibody can be produced by transfecting a host cell with a vector comprising a DNA sequence encoding the antibody. One or more vectors can be used to transfect the DNA sequence expressing at least one VL and one VH region into the host cell. Examples of descriptions of recombinant means for antibody formation and production include Delves, ANTIBODY PRODUCTION: ESSENTIAL TECHNIQUES (Wiley, 1997); Shepherd,

et al, MONOCLONAL ANTIBODIES (Oxford University Press, 2000); Goding, MONOCLONAL ANTIBODIES: PRINCIPLES AND PRACTICE (Academic Press, 1993); CURRENT PROTOCOLS IN IMMUNOLOGY (John Wiley & Sons, latest edition). An antibody according to the present invention can be modified with recombinant means to increase greater efficiency of the antibody in mediating the desired function. Typically, the substitutions will be conservative substitutions. For example, at least one amino acid in the constant region of the antibody can be replaced with a different residue. See e.g. US Patent No. 5,624,821, US Patent No. 6,194,551, Application No. WO 9958572; and Angal, et al., Mol. Immunol. 30:105-08 (1993). The modification in amino acids includes deletions, additions, substitutions of amino acids. In some cases, such changes are made to reduce unwanted activities, e.g. complement-dependent cytotoxicity. Often the antibodies are labeled by coupling, either covalently or non-covalently, to a substance that provides a detectable signal. A number of brands and conjugation techniques are known and are widely disclosed in both the scientific and patent literature. These antibodies can be screened for binding to normal or defective PSCA. See e.g. ANTIBODY ENGINEERING: A PRACTICAL APPROACH (Oxford University Press, 1996). Suitable antibodies with the desired biological activities can be identified by the following in vitro assays including but not limited to: proliferation, migration, adhesion, soft agar growth, angiogenesis, cell-cell communication, apoptosis, transport, signal transduction and

the following in vzvo trials such as inhibition of tumor growth. The antibodies provided herein may also be useful in diagnostic applications. As capture or non-neutralizing antibodies, they can be screened for the ability to bind to the specific antigen without inhibiting the receptor binding or biological activity of the antigen. As neutralizing antibodies, the antibodies can be useful in competitive binding experiments. They can also be used to quantify PSCA or its receptor.

An "antibody fragment" is defined as at least part of the variable region of the immunoglobulin molecule that binds to its target, i.e. the antigen-binding region. In one embodiment, it specifically covers single anti-PSCA antibodies and clones thereof (including agonist, antagonist and neutralizing antibodies) and anti-PSCA antibody compositions with polyepitopic specificity. The antibody of the present methods and compositions may be monoclonal or polyclonal. An antibody can be in the form of an antigen-binding antibody fragment comprising a Fab fragment, F(ab')2 fragment, a single chain variable region and the like. Fragments of intact molecules can be generated using methods well known in the art and include enzymatic digestion and recombinant agents.

As used herein, any form of the "antigen" can be used to generate an antibody specific for PSCA. Thus, the induced antigen can be a single epitope, multiple epitopes or the entire protein alone or in combination with one or more immunogenicity enhancing agents known in the field. The induced antigen can be an isolated full-length protein, a cell surface protein (eg, immunization with cells transfected with at least part of the antigen), or a soluble protein (eg, immunization with only the extracellular domain portion of the protein). The antigen can be produced in a genetically modified cell. The DNA encoding the antigen can be genomic or non-genomic (eg cDNA) and encodes at least part of the extracellular domain. As used herein, the term "part" denotes the minimal number of amino acids or nucleic acids, as appropriate, to constitute an immunogenic epitope of the antigen of interest. Any of the genetic vectors suitable for transformation of the cells of interest may be used, including but not limited to adenoviral vectors, plasmids, and non-viral vectors such as cationic lipids. In one embodiment, the antibody of the methods and compositions herein specifically binds at least a portion of the extracellular domain of the PSCA of interest.

The antibodies or antigen-binding fragments thereof provided herein may be conjugated to a "bioactive agent." As used herein, the term "bioactive agent" refers to any synthetic or naturally occurring compound that binds the antigen and/or enhances or mediates a desired biological effect to enhance cytotoxic toxins.

In one embodiment, the binding fragments applicable in the present invention are biologically active fragments. As used herein, the term "biologically active" denotes an antibody or antibody fragment capable of binding the desired antigenic epitope and directly or indirectly exerting a biological effect. Direct effects include, but are not limited to the modulation, stimulation and/or inhibition of a growth signal, the modulation, stimulation and/or inhibition of an antiapoptotic signal, the modulation, stimulation and/or inhibition of an apoptotic or necrotic signal, modulation, stimulation and / or inhibiting the ADCC cascade and modulating, stimulating and / or inhibiting the CDC cascade.

"Bispecific" antibodies are also useful in the present methods and compositions. As used herein, the term "bispecific antibody" refers to an antibody, typically a monoclonal antibody, that has binding specificities for at least two different antigen epitopes. In one embodiment, the epitopes are from the same antigen. In another embodiment, the epitopes are from two different antigens. Methods of Making Bispecific antibodies are known in the art. For example, bispecific antibodies can be produced recombinantly using the coexpression of two immunoglobulin heavy chain/light chain pairs. See, e.g., Milstein et al, Nature 305:537-39 (1983). Alternatively, bispecific antibodies can be produced by application of chemical binding. See, e.g., Brennan, et al, Science 229:81 (1985). Bispecific antibodies include bispecific antibody fragments. See, e.g., Hollinger, et al, Proe. Nati. Acad. Sei. U.S. A. 90: 6444-48 (1993), Gruber, et al, J. Immunol. 152:5368 (1994).

The monoclonal antibodies here specifically include "chimeric" antibodies where part of the heavy and/or light chain is identical to or homologous to corresponding sequences in antibodies derived from a particular species or belonging to a particular antibody class or subclass, while the rest of the chain(s) ) is identical to or homologous to corresponding sequences in antibodies derived from another species or belonging to another antibody class or subclass, as well as fragments of such antibodies, as long as they specifically bind the target antigen and/or display the desired biological activity (US- Pat. No. 4,816,567; and Morrison et al, Proe. Nat. Acad. Sci. USA 81: 6851-6855

(1984)).

The term "chemotherapeutic agent" refers to all chemical compounds that are effective in inhibiting tumor growth. Non-limiting examples of chemotherapeutic agents include alkylating agents; for example, nitrogen mustards, ethylene imine compounds and alkyl sulfonates; antimetabolites; for example folic acid, purine or pyrimidine antagonists; mitotic inhibitors; for example, vinca alkaloids and derivatives of podophyllotoxin, cytotoxic antibiotics, compounds that damage or interfere with DNA expression and growth factor receptor antagonists. In addition, chemotherapeutic agents include cytotoxic agents (as defined herein), antibodies, biological molecules and small molecules.

The term "codon-optimized sequences" refers to nucleotide sequences that are optimized for a particular host species by repositioning any codons that have a usage frequency of less than about 20%. Nucleotide sequences that are optimized for expression in a given host species by elimination of spurious polyadenylation sequences, elimination of exon/intron splicing signals, elimination of transposon-like repeats, and/or optimization of GC content in addition to codon optimization are referred to herein as "expression-enhanced sequences." "

A "combinatorial library" is a collection of different chemical compounds formed by either chemical synthesis or biological synthesis by combining several chemical "building blocks" such as reagents. For example, a linear combinatorial chemical library, such as a polypeptide (e.g., mutein) library, is formed by combining a set of chemical building blocks designated amino acids in every possible way for a given bond length (i.e., the number of amino acids in a polypeptide compound ). A variety of chemical compounds are synthesized through such combinatorial mixing of chemical building blocks (Gallop et al., J. Med. Chem. 37(9): 1233-1251 (1994)).

Preparation and screening of combinatorial libraries is well known to those skilled in the art. Such combinatorial chemical libraries include, but are not limited to, peptide libraries (see, e.g., US Patent No. 5,010,175, Furka, Pept. Prot. Res. 37:487-493 (1991), Houghton et al, Nature , 354:84-88 (1991)), peptoids (PCT Publication No. WO 91/19735), encoded peptides (PCT Publication WO 93/20242), random bio-oligomers (PCT Publication WO 92/00091), benzodiazepines (US Pat. No. 5,288,514), "diversomers" such as hydantoins, benzodiazepines and dipeptides (Hobbs et al, Proe. Nat. Acad. Sei. USA 90:6909-6913 (1993)), vinylogous polypeptides ( Hagihara et al, J. Amer. Chem. Soc. 114:6568 (1992)), non-peptidic peptidomimetics with a beta-D-glucose scaffold ("scaffolding") (Hirschmann et al, J. Amer. Chem. Soc. 114 :9217-9218 (1992)), analogous organic syntheses of small compound libraries (Chen et al, J. Amer. Chem. Soc. 116:2661 (1994)), oligocarbarnates (Cho, et al, Science 261:1303

(1993)) and/or peptidyl phosphonates (Campbell et al, J. Org. Chem. 59:658 (1994)). See generally Gordon et al, J. Med. Chem. 37:1385 (1994), nucleic acid libraries (see, e.g., Stratagene, Corp.), peptide-nucleic acid libraries (see, e.g., US Patent 5,539,083), antibody libraries (see, e.g., Vaughn et al, Nature Biotechnology 14(3): 309-314 (1996) and PCT/US96/10287), carbohydrate libraries (see, e.g., Liang et al, Science 274:1520-1522 (1996) and US Patent No. 5,593,853) and small organic molecule libraries (see, e.g., benzodiazepines, Baum, C&EN, Jan 18, page 33 (1993); isoprenoids, US Patent No. 5,569,588; thiazolidinones and metathiazanones, US Patent No. 5,549,974; pyrrolidines, US Patent Nos. 5,525,735 and 5,519,134; morpholino compounds, US Patent No. 5,506, 337; benzodiazepines, US Patent No. 5,288,514; and the like).

Devices for making combinatorial libraries are commercially available (see, e.g., 357 NIPS, 390 NIPS, Advanced Chem Tech, Louisville KY; Symphony, Rainin, Woburn, MA; 433A, Applied Biosystems, Foster City, CA; 9050, Pluss, Millipore, Bedford, NIA). Several well-known robotic systems have also been developed for solution phase chemistry. These systems include automated workstations such as the automated synthesizer developed by Takeda Chemical Industries, LTD. (Osaka, Japan) and many robotic systems using robotic arms (Zymate H, Zymark Corporation, Hopkinto, Mass.; Orca, Hewlett-Packard, Palo Alto, Calif), which mimic the manual synthetic operations performed by a chemist. Any of the above devices are suitable for use with the present invention. The nature and implementation of modifications to these devices (if any) to enable them to operate as described herein will be apparent to those skilled in the relevant art. In addition, a number of combinatorial libraries are themselves commercially available (see, e.g., ComGenex, Princeton, NJ; Asinex, Moscow, RU; Tripos, Inc., St. Louis, MO; ChemStar, Ltd, Moscow, RU; 3D Pharmaceuticals, Exton, PA; Martek Biosciences, Columbia, MD; etc).

As used herein, the term "conservative substitution" denotes substitutions of amino acids that are known to those skilled in the art and can generally be made without altering the biological activity of the resulting molecule. Those skilled in the art generally recognize that simple amino acid substitutions in non-essential regions of a polypeptide do not substantially alter biological activity (see, e.g., Watson, et al, MOLECULAR BIOLOGY OF THE GENE, The Benjamin/Cummings Pub. Co., p. 224 (4th Edition 1987)). Such examples of substitutions are preferably made according to those indicated in table(s) Ul(a-b). For example, such changes include the substitution of any of isoleucine (I), valine (V) and leucine (L) for any other of these hydrophobic amino acids; aspartic acid (D) for glutamic acid (E) and vice versa; glutamine (Q) for asparagine (N) and vice versa; and serine (S) for threonine (T) and vice versa. Other substitutions may also be considered conservative, depending on the environment of the particular amino acid and its role in the three-dimensional structure of the protein. For example, glycine (G) and alanine (A) can often be interchanged, as can alanine (A) and valine (V). Methionine (M), which is relatively hydrophobic, can often be exchanged with leucine and isoleucine and sometimes with valine. Lysine (K) and arginine (R) are often interchanged in locations where the significant feature of the amino acid residue is its charge and the different pKs of these two amino acid residues are not significant. Still other changes may be considered "conservative" in particular settings (see e.g. table ni(a) here; pages 13-15 "Biochemistry" 2nd ED. Lubert Stryer ed (Stanford University); Henikoff et al, PNAS 1992 Vol 89 10915-10919; Lei et al, J Biol Chem 1995 May 19; 270(20): 11882-6). Other substitutions are also permitted and can be determined empirically or in accordance with known conservative substitutions.

The term "cytotoxic agent" denotes a substance that inhibits or prevents expression activity of cells, function of cells and/or causes destruction of cells. The term shall include radioactive isotopes, chemotherapeutic agents and toxins such as small molecule toxins or enzymatically active toxins of bacterial, fungal, plant or animal origin, including fragments and/or variants thereof. Examples of cytotoxic agents include, but are not limited to auristatins, auristatin E, auromycins, maytansinoids, yttrium, bismuth, ricin, ricin A chain, combrestatin, duocarmyeins, dolostatins, doxorubicin, daunorubicin, taxol, cisplatin, ccl065, ethidium bromide, mitomycin , etoposide, tenoposide, vincristine, vinblastine, colchicine, dihydroxyanthraquinedione, actinomycin, diphtheriatoxin, Pseudomonas exotoxin (PE) A, PE40, abrin, abrin A chain, modeccin A chain, alpha-sarcin, gelonin, mitogellin, retstrictocin, phenomycin, enomycin, curicin, crotin, calicheamicin, Sapaonaria officinalis inhibitor and glucocorticoid and other chemotherapeutic agents, as well as radioisotopes such as At211,11<31>,I12<5>,Y9<0>, Re<186>, Re<188 >, Sm<153>, Bi<212>or213, P<32>and radioactive isotopes of Lu including Lu177 Antibodies may also be conjugated to an anticancer prodrug activating enzyme capable of converting the prodrug into its active form.

As used herein, the term "diameric antibody fragments" refers to small antibody fragments with two antigen-binding sites, which fragments comprise a heavy chain variable domain (Vh) linked to a light chain variable domain (Vl) in the same polypeptide chain (Vh-Vl). By using a linker that is too short to allow pairing between the two domains on the same chain, the domains are forced to pair with the complementary domains of another chain, creating two antigen-binding sites. Dimeric antibody fragments are described more fully in, e.g. EP 404,097; WO 93/11161; and Hollinger et al, Proe. Nati. Acad. Pollock. USA 90:6444-48 (1993). The "gene product" is used herein to indicate a peptide/protein or mRNA. For example, a "gene product of the invention" is sometimes referred to herein as a "cancer amino acid sequence", "cancer protein", "protein from a cancer listed in Table I", a "cancer mRNA", "mRNA from a cancer listed up in Table I", etc. In one embodiment, the cancer protein is encoded for by a nucleic acid from figure 1. The cancer protein can be a fragment or, alternatively, be the full-length protein encoded for by nucleic acids from figure 1. In one embodiment, a cancer amino acid sequence is used to determine sequence identity or similarity. In another embodiment, the sequences are naturally occurring allelic variants of a protein coded for by a nucleic acid from figure 1.1 another embodiment, the sequences are sequence variants as further described here. "Heteroconjugate" antibodies are useful in the present method and composition. As used herein, the term "heteroconjugate antibody" refers to two covalently linked antibodies. Such antibodies can be prepared using known methods in synthetic protein chemistry, including the use of cross-linking agents. See, e.g., US Patent No. 4,676,980. "High throughput screening" assays for the presence, absence, quantification or other properties of particular nucleic acids or protein products are well known to those skilled in the art. Correspondingly, binding experiments and reporter gene experiments are similarly well known. Thus describes e.g. US Patent No. 5,559,410 High Throughput Screening Methods for Proteins; US Patent No. 5,585,639 describes high-throughput screening methods for nucleic acid binding (ie, in arrays); while US Patent Nos. 5,576,220 and 5,541,061 describe high-throughput methods for screening for ligand/antibody binding.

In addition, high-throughput screening systems are commercially available (see, e.g., Amersham Biosciences, Piscataway, NJ; Zymark Corp., Hopkinnton, MA; Air Tehnical Industries, Mentor, OH; Beckman Instruments, Inc. Fullerton, CA; Precision Systems, Inc. ., Natick, MA; etc.). These systems typically automate entire procedures, including all sample and reagent pipetting, liquid dispensing, time incubations, and final readings with the microplate in detector(s) appropriate for the experiment. These configurable systems provide high throughput and fast startup as well as a high degree of flexibility and user customization. The manufacturers of such systems provide detailed protocols for various high-throughput systems. Thus provides e.g. Zymark Corp. technical reports describing screening systems for detecting the modulation of gene transcription, ligand binding and the like.

The term "homologous" denotes a molecule that exhibits homology with another molecule, for example by having sequences of chemical residues that are the same or similar at corresponding positions.

In one embodiment, the antibody provided herein is a "human antibody." As used herein, the term "human antibody" denotes an antibody in which substantially all of the light chain and heavy chain sequences, including the complementarity-determining regions (CDRs), are from human genes. In one embodiment, human monoclonal antibodies are prepared by the trioma technique, the human B-cell technique (see, e.g., Kozbor, et al, Immunol. Today 4: 72 (1983) ), the EBV transformation technique (see, e.g., Cole et al. MONOCLONAL ANTIBODIES AND CANCER THERAPY 77-96 (1985 )) or by using phage display (see, e.g., Marks et al, J. Mol. Biol. 222:581 (1991)). In a specific embodiment, the human antibody is generated in a transgenic mouse. Techniques for producing such partial to complete human antibodies are known in the art and any of such techniques may be used.According to one particularly preferred embodiment, complete human antibody sequences are made in a transgenic mouse engineered to express human heavy and light chain an tistogenes. An example of a description of the production of transgenic mice that produce human antibodies can be found in application no. WO 02/43478 and US Patent 6,657,103 (Abgenix) and its progeny. B cells from transgenic mice that produce the desired antibody can then be condensed to produce hybridoma cell lines for continuous production of the antibody. See e.g. US Patent No. 5,569,825; 5,625,126; 5,633,425; 5,661,016; and 5,545,806; and Jakobovits, Adv. Medication Del. Fox. 31:33-42 (1998); Green,

et al, J. Exp. Med. 188:483-95 (1998).

"Human leukocyte antigen" or "HLA" is a human class I or class II major histocompatibility complex (MHC) protein (see, e.g., Stites, et al, IMMUNOLOGY, 8TH ED., Lange Publishing, Los Altos, CA ( 1994).

As used herein, the term "humanized antibody" denotes forms of antibodies that contain sequences from non-human (eg, murine) antibodies as well as human antibodies. Such antibodies are chimeric antibodies that contain minimal sequence derived from non-human immunoglobulin. General the humanized antibody will comprise substantially all of at least one, and typically two, variable domains, wherein all or substantially all of the hypervariable loops correspond to those of a non-human immunoglobulin and all or substantially all of the FR regions are those of a human immunoglobulin sequence. will optionally also include at least a portion of an immunoglobulin constant region (Fe), typically that of a human immunoglobulin See, eg, Cabilly US Patent No. 4,816,567; Queen et al. (1989) Proe. Nat'1 Acad. Sci. USA 86:10029-10033; and ANTIBODY ENERGY ENERGY: A PRACTICAL APPROACH (Oxford University Press 1996).

The terms "hybridize", "hybridize", "hybridize" and the like, used in the context of polynucleotides, are meant to refer to conventional hybridization conditions, preferably such as hybridization in 50% formamide/6XSSC/0.1% SDS/100 µg/ml ssDNA , where temperatures for hybridization are above 37 degrees C and temperatures for washing in 0.1XSSC/0.1% SDS are above 55 degrees C.

The phrases "isolated" or "biologically pure" refer to material that is substantially or essentially free of components that normally accompany the material as it is found in its natural state. Thus, isolated peptides according to the invention preferably do not contain materials normally associated with the peptides in their in situ environment. For example, a polynucleotide is said to be "isolated" when it is substantially separated from contaminant polynucleotides that correspond to or are complementary to genes other than the PSCA genes or that encode polypeptides other than the PSCA gene product or fragments thereof. One skilled in the art can readily use nucleic acid isolation procedures to obtain an isolated PSCA polynucleotide. A protein is said to be "isolated," for example, when physical, mechanical, or chemical methods are used to remove the PSCA proteins from cellular constituents normally associated with the protein. One skilled in the art can readily use standard purification procedures to obtain an isolated PSCA protein. Alternatively, an isolated protein can be prepared by chemical means.

Suitable "labels" include radionuclides, enzymes, substrates, cofactors, inhibitors, fluorescent groups, chemiluminescent groups, magnetic particles and the like. Patents teaching the use of such marks include US Patent No. 3,817,837; 3,850,752; 3,939,350; 3,996,345; 4,277,437; 4,275,149; and 4,366,241.1 in addition, the antibodies provided herein may be useful as the antigen-binding component of fluorobodies. See e.g. Zeytun et al., Nat. Biotechnol. 21:1473-79 (2003).

The term "mammal" denotes any organism classified as a mammal including mice, rats, rabbits, dogs, cats, cows, horses and humans.

The terms "metastatic prostate cancer" and "metastatic disease" mean prostate cancer that has spread to regional lymph nodes or to scattered sites and shall include stage D disease in the AUA system and stage TxNxM+ in the TNM system. As is the case with locally advanced prostate cancer, surgery is generally not indicated for patients with metastatic disease and hormonal (androgen ablation) therapy is a preferred treatment modality. Patients with metastatic prostate cancer eventually develop an androgen-resistant state within 12 to 18 months of initiation of treatment. Approximately half of these androgen-resistant patients die within 6 months of developing that status. The most common site for prostate cancer metastasis is bone. Prostate cancer bone metastases are often osteoblastic rather than osteolytic (ie resulting in net bone formation). Bone metastases are most often found in the spine, followed by the femur, pelvis, ribs, cranium and humerus. Other common sites for metastasis include the lymph nodes, lung, liver and brain. Metastatic prostate cancer is typically diagnosed by open or laparoscopic pelvic lymphadenectomy, whole-body radionuclide scanning, skeletal radiography and/or bone lesion biopsy.

The term "modulator" or "test compound" or "drug candidate" or grammatical equivalents as used herein describes any molecule, e.g. protein, oligopeptide, small organic molecule, polysaccharide, polynucleotide, etc., to be tested for the capacity to directly or indirectly alter the cancer phenotype or the expression of a cancer sequence, e.g. a nucleic acid or protein sequences or effects of cancer sequences (eg signaling, gene expression, protein interaction, etc.) In one aspect, a modulator will neutralize the effect of a cancer protein according to the invention. By "neutralize" is meant that an activity of a protein is inhibited or blocked, together with the subsequent effect on the cell. In another aspect, a modulator will neutralize the effect of a gene, and its corresponding protein, according to the invention by neutralizing levels of said protein. In preferred embodiments, modulators alter expression profiles or expression profile nucleic acids or proteins provided herein or downstream effector reaction pathways. In one embodiment, the modulator suppresses a cancer phenotype,/e£s. to a normal tissue fingerprint. In another embodiment, a modulator induces a cancer phenotype. In general, a number of test mixtures are run in parallel with different agent concentrations to achieve a differential response to the different concentrations. Typically, one of these concentrations serves as a negative control, i.e. at zero concentration or below the detection level.

Modulators, drug candidates or test compounds encompass a number of chemical classes, although they are typically organic molecules, preferably small organic compounds having a molecular weight of greater than 100 and less than about 2,500 daltons. Preferred small molecules are less than 2000 or less than 1500 or less than 1000 or less than 500 D. Candidate agents include functional groups necessary for structural interaction with proteins, particularly hydrogen bonding and typically include at least one amine, carbonyl, hydroxyl or carboxyl group, preferably at least two of the functional chemical groups. The candidate agents often comprise cyclic carbon or heterocyclic structures and/or aromatic or polyaromatic structures substituted with one or more of the above functional groups. Modulators also include biomolecules such as peptides, saccharides, fatty acids, steroids, purines, pyrimidines, derivatives, structural analogues or combinations thereof. Particularly preferred are peptides. One class of modulators are peptides, for example from about five to about 35 amino acids, with from approximately five to approx. 20 amino acids that are preferred and from approx. 7 to approx. 15 which is particularly preferred. Preferably, the cancer modulating protein is soluble, comprises a non-transmembrane region and/or has an N-terminal Cys to aid solubility. In one embodiment, the C-terminus of the fragment is kept as a free acid and the N-terminus is a free amine to aid in coupling, ie to cysteine. In one embodiment, a cancer protein according to the invention is conjugated to an immunogenic agent as described herein. In one embodiment, the cancer protein is conjugated to BSA. The peptides according to the invention, e.g. of preferred lengths, may be linked to each other or to other amino acids to create a longer peptide/protein. The modulating peptides can be digested by naturally occurring proteins described above, random peptides or "one-sided" random peptides. In a preferred embodiment, peptide/protein-based modulators are antibodies and fragments thereof, as defined herein.

Cancer modulators can also be nucleic acids. Nucleic acid modulating agents can be naturally occurring nucleic acids, random nucleic acids or "one-sided" random nucleic acids. For example, digestion of prokaryotic or eukaryotic genomes can be used in an approach analogous to that described above for proteins.

The term "monoclonal antibody," as used herein, denotes an antibody obtained from a population of substantially homogeneous antibodies, i.e., the individual antibodies comprised in the population are identical except for possible naturally occurring mutations that may be present in minor amounts. Monoclonal antibodies are highly specific that are directed against a single antigen epitope. In contrast, conventional (polyclonal) antibody preparations typically comprise a plurality of antibodies directed against (or specific for) different epitopes. In one embodiment, the polyclonal antibody contains a variety of monoclonal antibodies with different epitope specificities, affinities or avidities within a single antigen containing multiple antigenic epitopes. The modifier "monoclonal" indicates the nature of the antibody which is obtained from a substantially homogeneous population of antibodies and has not been engineered to require production of the antibody by any special procedure. For example, the monoclonal antibodies to be used according to the present invention can be prepared by the hybridoma method first described by Kohler et al., Nature 256: 495 (1975) or can be prepared by recombinant DNA methods (see e.g. US Pat. no . 4,816,567). The "monoclonal antibodies" can also be isolated from phage antibody libraries using the techniques described, for example, in Clackson et al, Nature 352: 624-628 (1991) and Marks et al, J. Mol. Biol. 222: 581-597 (1991). These monoclonal antibodies will usually bind with at least a Kd of approx. 1 uM, more usually at least approx. 300 nM, typically at least approx. 30 nM, preferably at least approx.

10 nM, more preferably at least approx. 3 nM or better, usually determined by ELISA.

A "motif", as in biological motif of a PSCA-related protein, denotes any pattern of amino acids forming part of the primary sequence of a protein that is associated with a particular function (eg, protein-protein interaction, protein-DNA -interaction, etc) or modification ( e.g. phosphorylated, glycosylated or amidated) or localization ( e.g. secretory sequence, nuclear localization sequence, etc.) or a sequence consistent with becoming immunogenic, either humorally or cellular. A motif may be either consecutive or capable of being arranged in certain positions which are generally consistent with a certain function or property. In the context of HLA motifs, "motif" denotes the pattern of residues in a peptide of defined length, usually a peptide of from approx. 8 to approx. 13 amino acids for a class I-HLA motif and from approx. 6 to approx. 25 amino acids for a class II HLA motif, which is recognized by a special HLA molecule. Peptide motifs for HLA binding are typically different for each protein encoded by each human HLA allele and differ in the pattern of the primary and secondary anchor residues. Frequently occurring motifs are indicated in table V.

A "pharmaceutical additive" includes a material such as an adjuvant, a carrier, pH regulating and buffering agents, tonicity regulating agents, wetting agents, preservatives and the like.

"Pharmaceutically acceptable" denotes a non-toxic, inert and/or composition that is physiologically compatible with humans or other mammals.

The term "polynucleotide" means a polymer formed from nucleotides of at least 10 bases or base pairs in length, either ribonucleotides or deoxynucleotides or a modified form of any type of nucleotide and shall include single- and double-stranded forms of DNA and/or RNA. In the field, this term is often used interchangeably with "oligonucleotide". A polynucleotide can comprise a nucleotide sequence described here where thymidine (T), as shown for example in figure 1, can also be uracil (U); this definition relates to the differences between the chemical structures of DNA and RNA, in particular the observation that one of the four main bases in RNA is uracil (U) instead of thymidine (T).

The term "polypeptide" means a polymer of at least approx. 4, 5, 6, 7 or 8 amino acids. Throughout the specification, standard three-letter or single-letter designations for amino acids are used. In the field, this term is often used interchangeably with "peptide" or "protein".

An HLA "primary anchor residue" is an amino acid at a specific position along a peptide sequence that is understood to provide a point of contact between the immunogenic peptide and the HLA molecule. One to three, usually two, primary anchor residues within a peptide of defined length generally define a "motif" for an immunogenic peptide. These residues are understood to fit in close contact with the peptide binding pit of an HLA molecule, with their side chains recessed into specific pockets of the binding pit. In one embodiment, for example, the primary anchor residues for an HL A class I molecule are located in position 2 (from the amino-terminal position) and at the carboxyl-terminal position of an 8, 9, 10, 11 or 12 residues peptide epitope according to the invention . Alternatively, in another embodiment, the primary anchor residues of a peptide bind an HLA class II molecule distributed relative to each other, rather than to the terminus of a peptide, where the peptide is generally at least 9 amino acids in length. The primary anchor positions for each motif and supermotif are listed in Table IV(a). For example, analogous peptides can be formed by altering the presence or absence of particular residues in the primary and/or secondary anchor positions shown in Table IV. Such analogs are used to modulate the binding affinity and/or population coverage of a peptide comprising a particular HLA motif or supermotif.

"Radioisotopes" include, but are not limited to the following (non-limiting examples of applications are also set forth in Table IV(I)).

By "randomized" or grammatical equivalents as here applied to nucleic acids and proteins is meant that each nucleic acid and peptide consists of essentially random nucleotides and amino acids respectively. These random peptides (or nucleic acids, described herein) may incorporate any nucleotide or amino acid at any position. The synthetic process can be designed to generate randomized proteins or nucleic acids, to allow the generation of all or most of the possible combinations over the length of the sequence, thus generating a library of randomized candidate bioactive proteinaceous agents.

In one embodiment, a library is "fully randomized," with no sequence preferences or constants at any position. In another embodiment, the library is a "one-sided random" library. That is, some positions within the sequence are either held constant or are selected from a limited number of possibilities. For example, the nucleotides or amino acid residues are randomized within a defined class, e.g. of hydrophobic amino acids, hydrophilic residues, sterically one-sided (either small or large) residues, against the formation of nucleic acid binding domains, the formation of cysteines, for cross-linking, prolines in SH-3 domains, serines, threonines, tyrosines or histidines in phosphorylation sites, etc. or to purines, etc.

A "recombinant" DNA or RNA molecule is a DNA or RNA molecule that is subjected to molecular manipulation in vitro.

As used herein, the term "single chain Fv" or "scFv" or "single chain" antibody refers to antibody fragments comprising the Vh and VL domains of antibody, where these domains are present in a single polypeptide chain. In general, the Fv polypeptide further comprises a polypeptide linker between the Vh and VL domains which enables the sFv to form the desired structure for antigen binding. For an overview of sFv see Pluckthun, THE PHARMACOLOGY OF MONOCLONAL ANTIBODIES, vol. 113, Rosenburg and Moore eds. Springer-Verlag, New York, pp. 269-315

(1994).

Non-limiting examples of "small molecules" include compounds that bind or interact with PSCA, ligands including hormones, neuropeptides, chemokines, odorants, phospholipids, and functional equivalents thereof that bind and preferentially inhibit PSCA protein function. Such non-limiting small molecules preferably have a molecular weight of less than approx. 10 kDa, more preferably below approx. 9, approx. 8, approx. 7, approx. 6, approx. 5 or approx. 4 kDa. In certain embodiments, small molecules are physically associated with or bind PSCA protein; not found in naturally occurring metabolic pathways; and/or is more soluble in aqueous than non-aqueous solutions.

As used herein, the term "specific" denotes the selective binding of the antibody to the target antigen epitope. Antibodies can be tested for specificity of binding by comparing binding to the appropriate antigen against binding to an irrelevant antigen or antigen mixture under a given set of conditions. If the antibody binds to the appropriate antigen at least 2, 5, 7 and preferably 10-fold more than to the irrelevant antigen or antigen mixture then it is considered to be specific. In one embodiment, a specific antibody is one that only binds the PSCA antigen, but does not bind to the irrelevant antigen. In another embodiment, a specific antibody is one that binds human PSCA antigen but does not bind a non-human PSCA antigen by 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or greater amino acid homology with the PSCA antigen. In another embodiment, a specific antibody is one that binds human PSCA antigen and binds murine PSCA antigen, but with a higher degree of binding to the human antigen. In another embodiment, a specific antibody is one that binds human PSCA antigen and binds primate PSCA antigen, but with a higher degree of binding to the human antigen. In another embodiment, the specific antibody binds to human PSCA antigen and any non-human PSCA antigen, but with a higher degree of binding to the human antigen or any combination thereof.

The "stringency" of hybridization reactions is easily determinable by a person skilled in the art and is generally an empirical calculation dependent on probe length, washing temperature and salt concentration. In general, longer probes require higher temperatures for correct annealing, while shorter probes need lower temperatures. Hybridization generally depends on the ability of denatured nucleic acid sequences to reanneal when complementary strands are present in an environment below their melting temperature. The higher the degree of desired homology between the probe and hybridizable sequence, the higher the relative temperature that can be used. As a result, it follows that higher relative temperatures will tend to make the reaction conditions more stringent, while lower temperatures less so. For further details and explanation of hybridization stringency reactions, see Ausubel et al, Current Protocols in Molecular Biology, Wiley Interscience Publishers, (1995).

"Stringent conditions" or "high stringency conditions" as defined herein are identified by, but not limited to, those that: (1) use low ionic strength and high temperature in washing, such as 0.015 M sodium chloride/0.0015 M sodium citrate/0 .1% sodium dodecyl sulfate at 50°C; (2) apply during hybridization a denaturing agent, such as formamide, for example 50%

(v/v) formamide with 0.1% bovine serum albumin/0.1% Ficoll/0.1% polyvinylpyrrolidone/50 mM sodium phosphate buffer at pH 6.5 with 750 mM sodium chloride, 75 mM sodium citrate at 42°C; or (3) using 50% formamide, 5 x SSC (0.75 M NaCl, 0.075 M sodium citrate), 50 mM sodium phosphate (pH 6.8), 0.1% sodium pyrophosphate, 5 x Denhardt's solution, sonicated salmon sperm DNA (50 ug/ml), 0.1% SDS and 10% dextran sulfate at 42°C, with washes at 42°C in 0.2 x SSC (sodium chloride/sodium citrate) and 50% formamide at 55°C, followed by a high stringency wash consisting of 0.1 x SSC containing EDTA at 55°C. "Moderately stringent conditions" are described by, but not limited to, those in Sambrook et al, Molecular Cloning: A Laboratory Manual, New York: Cold Spring Harbor Press, 1989 and include the use of washing solution and hybridization conditions (e.g., temperature, ionic strength and %SDS) less stringent than those described above. An example of moderately stringent conditions is overnight

incubation at 65°C in a solution comprising: 1% bovine serum albumin, 0.5M sodium phosphate pH7.5, 1.25mM EDTA and 7% SDS 5 x SSC (150mM NaCl, 15mM trisodium citrate), followed by washing the filters in 2 x SSC/1% SDS at 50°C and 0.2 x SSC/0.1% SDS at 50°C. Those skilled in the art will understand how to regulate the temperature, ionic strength, etc. necessary to adjust factors such as probe length and the like.

An HLA "supermotif" is a peptide binding specificity shared by HLA molecules encoded by two or more HLA alleles. Total phenotypic frequencies of HLA supertypes in different ethnic populations are indicated in Table IV (f). The non-limiting constituents of various supertypes are as follows: A2: A<*>0201, A<*>0202, A<*>0203, A<*>0204, A<*>0205, A<*>0206, A <*>6802, A<*>6901, A<*>0207 A3: A3, All, A31, A<*>3301, A<*>6801, A<*>0301, A<*>1101, A< *>3101

B7: B7, B<*>3501-03, B<*>51, B<*>5301, B<*>5401, B<*>5501, B<*>5502, B<*>5601, B< *>6701, B<*>7801, B*0702, B*5101,B*5602

B44: B<*>3701, B<*>4402, B<*>4403, B<*>60 (B<*>4001), B61 (B<*>4006)

Al: A<*>0102, A<*>2604, A<*>3601, A<*>4301, A<*>8001

A24: A<*>24, A<*>30, A<*>2403, A<*>2404, A<*>3002, A<*>3003

B27: B<*>1401-02, B<*>1503, B<*>1509, B<*>1510, B<*>1518, B<*>3801-02, B<*>3901, B< *>3902, B<*>3903-04, B<*>4801-02, B<*>7301, B<*>2701-08

B58: B<*>1516, B<*>1517, B<*>5701, B<*>5702, B58

B62: B<*>4601, B52, B<*>1501 (B62), B<*>1502 (B75), B<*>1513 (B77)

Estimated population coverage provided by different HLA supertype combinations is given in Table IV(g).

As used herein, "treating" or "therapeutic" and grammatically related terms refer to any improvement of any consequence of disease, such as prolonged survival, less disease and/or a reduction in side effects that are the byproducts of an alternative therapeutic modality; which is easily recognized in the field, full eradication of disease is a preferred outcome although not a requirement for a treatment to work.

A "transgenic animal" (eg, a mouse or rat) is an animal that has cells containing a transgene, which transgene was introduced into the animal or an ancestor of the animal prenatally. an embryonic stage. A "transgene" is a DNA integrated into the genome of a cell from which a transgenic animal is developed.

As used herein, an HLA or cellular immune response "vaccine" is a composition containing or encoding one or more peptides of the invention. There are a number of embodiments of such vaccines, such as a cocktail of one or more individual peptides; one or more peptides according to the invention comprised of a polyepitopic peptide; or nucleic acids encoding such individual peptides or polypeptides,/e£s. a minigene encoding a polyepitopic peptide. The "one or more peptides" may comprise any whole integer unit from 1-150 or more e.g. at least 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26 , 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 55 , 60, 65, 70, 75, 80, 85, 90, 95, 100, 105, 110, 115, 120, 125, 130, 135, 140, 145 or 150 or more peptides according to the invention. The peptides or polypeptides may optionally be modified, such as by lipidation, addition of targeted or other sequences. HLA class I peptides according to the invention can be mixed with or bound to HLA class U peptides, to facilitate activation of both cytotoxic T lymphocytes and helper T lymphocytes. HLA vaccines may also comprise peptide-pulsed antigen-presenting cells, e.g. dendritic cells.

The term "variant" denotes a molecule that exhibits a variation from a described type or norm, such as a protein that has one or more different amino acid residues in the corresponding position(s) of a specifically described protein (eg, the PSCA protein shown in Figure 1. An analog is an example of a variant protein Splice isoforms and single nucleotide polymorphisms (SNPs) are further examples of variants.

The "PSCA-related proteins" according to the invention include those specifically identified here, as well as allelic variants, conservative substitution variants, analogues and homologues which can be isolated/formed and characterized without unnecessary experimentation by following the methods described here or readily available in the field. Fusion proteins combining parts of different PSCA proteins or fragments thereof, as well as fusion proteins of a PSCA protein and a heterologous polypeptide are also encompassed. Such PSCA proteins are collectively referred to as the PSCA-related proteins, the proteins of the invention or PSCA. The term "PSCA-related protein" denotes a polypeptide fragment or a PSCA protein sequence of 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25 or more than 25 amino acids; or at least 30, 35, 40, 45, 50, 55, 60, 65, 70, 80, 85, 90, 95, 100, 105, 110, 115, 120, 125, 130, 135, 140, 145, 150, 155, 160, 165, 170, 175, 180, 185, 190, 195, 200, 225, 250, 275, 300, 325, 330, 335, 339 or more amino acids.

II.) PSCA polynucleotides

Polynucleotides corresponding or complementary to all or parts of a PSCA gene, mRNA and/or coding sequence have been described, preferably in isolated form, comprising polynucleotides that code for a PSCA-related protein and fragments thereof, DNA, RNA, DNA/RNA -hybrid and related molecules, polynucleotides or oligonucleotides complementary to a PSCA gene or mRNA sequence or part thereof and polynucleotides or oligonucleotides that hybridize to a PSCA gene, mRNA or to a PSCA encoding polynucleotide (collectively "PSCA polynucleotides "). In all cases when referred to in this paragraph, T can also be U in Figure 1.

Embodiments of a PSCA polynucleotide include: a PSCA polynucleotide having the sequence shown in Figure 1, the nucleotide sequence of PSCA as shown in Figure 1 where T is U; at least 10 consecutive nucleotides of a polynucleotide having the sequence shown in Figure 1; or at least 10 consecutive nucleotides of a polynucleotide having the sequence shown in Figure 1 where T is U.

Polynucleotides which code for relatively long parts of a PSCA protein are also within the scope of the invention. For example, polynucleotides encoding from about amino acid 1 (or 20 or 30 or 40 etc.) to about amino acid 20, (or 30 or 40 or 50 etc.) of the PSCA protein "or variant" shown in Figure 1 or Figure 3 is formed by a number of techniques well known in the field. These polynucleotide fragments can comprise any part of the PSCA sequence as shown in Figure 1.

n.A.) Applications of PSCA Polynucleotides

ILA. 1. Monitoring of genetic abnormalities

The polynucleotides of the preceding paragraphs have several different specific applications. The human PSCA gene map to the chromosomal location indicated in the example designated "Chromosomal Mapping of PSCA." For example, because the PSCA gene maps to this chromosome, polynucleotides encoding different regions of the PSCA proteins are used to characterize cytogenetic abnormalities of this chromosomal locus, such as abnormalities identified to be associated with various cancers. In certain genes, a number of chromosomal abnormalities including rearrangements have been identified as frequent cytogenetic abnormalities in several different cancers (see, e.g., Krajinovic et al, Mutat. Res. 382(3-4): 81-83 (1998); Johansson et al , Blood 86(10): 3905-3914 (1995) and Finger et al, P. N. A. S. 85(23): 9158-9162 (1988)). Thus, polynucleotides encoding specific regions of the PSCA proteins provide new tools that can be used to delineate, with greater precision than previously possible, cytogenetic abnormalities in the chromosomal region encoding PSCA that may contribute to the malignant phenotype. In this context, these polynucleotides satisfy a need in the field to expand the sensitivity of chromosomal screening to identify more subtle and less common chromosomal abnormalities (see, e.g., Evans et al, Am. J. Obstet. Gynecol 171(4): 1055 -1057

(1994)).

Furthermore, since PSCA has been shown to be highly expressed in prostate and other cancers, PSCA polynucleotides are used in methods to determine the status of PSCA gene products in normal versus cancerous tissue. Typically, polynucleotides encoding specific regions of the PSCA proteins are used to assess the presence of perturbations (such as deletions, insertions, point mutations or changes resulting in a loss of an antigen, etc.) in specific regions of the PSCA gene, such as regions containing one or more motifs. Exemplary assays include both RT-PCR assays as well as single-strand conformation polymorphism (SSCP) analysis (see, e.g., Marrogi et al, J. Cutan. Pathol. 26(8): 369-378 (1999), both of which use polynucleotides that code for specific regions of a protein to examine these regions within the protein.

II.A.2. Antisense embodiments

Other specifically anticipated nucleic acid-related embodiments of the invention described herein are genomic DNA, cDNAs, ribozymes and antisense molecules, as well as nucleic acid molecules based on an alternative backbone or comprising alternative bases, whether derived from natural sources or synthesized, and include molecules capable of inhibit the RNA or protein expression of PSCA. For example, antisense molecules can be RNAs or other molecules, including peptide nucleic acids (PNAs) or non-nucleic acid molecules such as phosphorothioate derivatives that specifically bind DNA or RNA in a base pair-dependent manner. One skilled in the art can readily obtain these classes of nucleic acid molecules using the PSCA polynucleotides and polynucleotide sequences described herein.

Antisense technology involves the administration of exogenous oligonucleotides that bind to a target polynucleotide located in the cells. The term "antisense" denotes the fact that such oligonucleotides are complementary to their intracellular targets e.g. PSCA. See, for example, Jack Cohen, Oligodeoxynucleotides, Antisense Inhibitors of Gene Expression, CRC Press, 1989; and Synthesis 1:1-5 (1988). The PSCA antisense oligonucleotides according to the present invention comprise derivatives such as S-oligonucleotides (phosphorothioate derivatives or S-oligos, see Jack Cohen, supra), which show improved cancer cell growth inhibitory action. S oligos

(nucleoside phosphorothioates) are isoelectronic analogues of an oligonucleotide (O-oligo) where a non-bridging oxygen atom of the phosphate group is replaced by a sulfur atom. The S-oligos according to the present invention can be prepared by treating the corresponding O-oligos with 3H-1,2-benzodithiol-3-one-1,1-dioxide, which is a sulfur transfer reagent. See e.g. Lyer, R.P. et al, J. Org. Chem. 55:4693-4698 (1990); and Lyer, R.P. et al, J. Am. Chem. Soc. 112:1253-1254 (1990). Additional PSCA antisense oligonucleotides according to the present invention include morpholino antisense oligonucleotides known in the art (see, e.g., Partridge et al, 1996, Antisense & Nucleic Acid Drug Development 6: 169-175).

The PSCA antisense oligonucleotides described may typically be RNA or DNA that is complementary to and stably hybridizes to the first 100 5' codons or the last 100 3' codons of a PSCA genomic sequence or the corresponding mRNA. Absolute complementarity is not required, although high degrees of complementarity are preferred. Use of an oligonucleotide complementary to this region allows the selective hybridization to PSCA mRNA and not to mRNA specifying other regulatory subunits of protein kinase. In one embodiment, PSCA antisense oligonucleotides of the present invention are 15 to 30-mer fragments of the antisense DNA molecule having a sequence that hybridizes to PSCA mRNA. Optionally, the PSCA antisense oligonucleotide is a 30-mer oligonucleotide that is complementary to a region in the first 10 5' codons or last 10 3' codons of PSCA. Alternatively, the antisense molecules are modified to use ribozymes in the inhibition of PSCA expression, see e.g. L.A. Couture & D.T. Stinchcomb; Trends Genet 12: 510-515 (1996).

II. A. 3. Primers and primer pairs

Furthermore, specific embodiments of these nucleotides include primers and primer pairs, which allow the specific amplification of polynucleotides according to the invention or of any specific parts thereof and probes that selectively or specifically hybridize to nucleic acid molecules or to any part thereof. Probes may be labeled with a detectable label, such as, for example, a radioisotope, fluorescent compound, bioluminescent compound, chemiluminescent compound, metal chelator or enzyme. Such probes and primers are used to detect the presence of a PSCA polynucleotide in a sample and as a means of detecting a cell expressing a PSCA protein.

Examples of such probes include polypeptides comprising all or part of the human PSCA cDNA sequence shown in Figure 1. Examples of primer pairs capable of specifically amplifying PSCA mRNAs are also described in the examples. As will be appreciated by those skilled in the art, a large number of different primers and probes can be prepared based on the sequences provided herein and used effectively to amplify and/or detect a PSCA mRNA.

The PSCA polynucleotides herein are useful for a variety of purposes, including but not limited to their use as probes and primers for the amplification and/or detection of the PSCA gene(s), mRNA(s) or fragments thereof; as reagents for the diagnosis and/or prognosis of prostate cancer and other cancers; as coding sequences capable of directing the expression of PSCA polypeptides; as tools to modulate or inhibit the expression of the PSCA gene(s) and/or the translation of the PSCA transcript(s); and as therapeutic agents. II. A.4. Isolation of PSCA Encoding Nucleic Acid Molecules The PSCA cDNA sequences described herein enable the isolation of other polynucleotides encoding PSCA gene product(s), as well as the isolation of polynucleotides encoding PSCA gene product homologues, alternatively spliced isoforms, allelic variants and mutant forms of a PSCA gene product as well as polynucleotides encoding analogs of PSCA-related proteins. Various molecular cloning procedures that can be used to isolate full-length cDNAs encoding a PSCA gene are well known (see, for example, Sambrook, J. et al, Molecular Cloning: A Laboratory Manual, 2d edition, Cold Spring Harbor Press, New York, 1989 ; Current Protocols in Molecular Biology. Ausubel et al, Eds., Wiley and Sons, 1995). For example, lambda phage cloning procedures can be conveniently employed using commercially available cloning systems (eg, Lambda ZAP Express, Stratagene). Phage clones containing PSCA gene cDNAs can be identified by probing with a labeled PSCA cDNA or a fragment thereof. For example, in one embodiment, a PSCA cDNA (eg Figure 1) or part thereof can be synthesized and used as a probe to recover overlapping and full-length cDNAs corresponding to a PSCA gene. A PSCA gene itself can be isolated by screening genomic DNA libraries, bacterial artificial chromosome libraries (BACs), yeast artificial chromosome libraries (YACs), and the like, with PSCA DNA probes or primers.

II. A. 5. Recombinant nucleic acid molecules and host vector systems

Further described are recombinant DNA or RNA molecules containing a PSCA polynucleotide, a fragment, analog or homolog thereof, including but not limited to phages, plasmids, phagemids, cosmids, YACs, BACs, as well as various viral and non- viral vectors well known in the art and cells transformed or transfected with such recombinant DNA or RNA molecules. Methods for the formation of such molecules are well known (see, for example, Sambrook et al, 1989, supra).

A host vector system comprising a recombinant DNA molecule containing a PSCA polynucleotide, fragment, analog or homologue thereof within a suitable prokaryotic or eukaryotic host cell is further described. Examples of suitable eukaryotic host cells include a yeast cell, a plant cell, or an animal cell, such as a mammalian cell or an insect cell (eg, a baculovirus infectable cell such as an Sf9 or HighFive cell). Examples of suitable mammalian cells include various prostate cancer cell lines such as DU 145 and TsuPrl, other transfectable or transducible prostate cancer cell lines, primary cells (PrEC), as well as several mammalian cells routinely used for expression of recombinant proteins (e.g. COS-, CHO- , 293, 293T cells). More particularly, a polynucleotide comprising the coding sequence of PSCA or a fragment, analog or homolog thereof can be used to generate PSCA proteins or fragments thereof using any number of host vector systems routinely used and widely known in the art.

A wide range of host vector systems suitable for expression of PSCA proteins or fragments thereof are available, see for example Sambrook et al, 1989, supra; Current Protocols in Molecular Biology, 1995, supra). Preferred vectors for mammalian expression include but are not limited to pcDNA 3.1-myc-His-tag (Invitrogen) and the retroviral vector pSRDtkneo (Muller et al, 1991, MCB 11:1785). Using these expression vectors, PSCA can be expressed in many prostate cancer and non-prostate cancer cell lines, including, for example, 293, 293T, rat-1, NTH 3T3 and TsuPrl. The host vector systems according to the invention are applicable for the production of a PSCA protein or fragment thereof. Such host vector systems can be used to investigate the functional properties of PSCA and PSCA mutations or analogs.

Recombinant human PSCA protein or an analog or homologue or fragment thereof can be produced by mammalian cells transfected with a construct encoding a PSCA-related nucleotide. For example, 293 T cells can be transfected with an expression plasmid encoding PSCA or fragment, analog or homolog thereof, a PSCA-related protein is expressed in the 293T cells and the recombinant PSCA protein is isolated using standard purification procedures (e.g. .affinity purification using anti-PSCA antibodies). In another embodiment, a PSCA coding sequence is subcloned into the retroviral vector pSRaMSVtkneo and used to infect various mammalian cell lines, such as NLH 3T3, TsuPrl, 293 and rat-1 to establish PSCA expressing cell lines. Various other expression systems well known in the art can also be used. Expression constructs encoding a leader peptide linked in frame to a PSCA coding sequence can be used for the generation of a secreted form of recombinant PSCA protein.

As described herein, redundancy in the genetic code allows variation in PSCA gene sequences. In particular, it is known in the field that specific host species often have specific codon preferences and thus one can adapt the described sequence as preferred for a desired host. For example, preferred analog codon sequences typically have rare codons (ie codons that have a usage frequency of less than about 20% in known sequences of the desired host) replaced with higher frequency codons. Codon preferences for a specific species are calculated, for example, using codon usage tables available on the INTERNET such as at URL dna. Affr. go.j p/~nakamura/codon. html.

Additional sequence modifications are known to enhance protein expression in a cellular host. These include the elimination of sequences encoding spurious polyadenylation signals, exon/intron splice site signals, transposon-like repeats, and/or other such well-characterized sequences that are detrimental to gene expression. The GC content of the sequence is normalized to average levels for a given cellular host, as calculated with reference to known genes expressed in the host cell. Where possible, the sequence is modified to avoid predicted secondary mRNA hairpin structures. Other useful modifications include the addition of a translation initiation consensus sequence at the start of the open reading frame, as described in Kozak, Mol. Cell Biol., 9:5073-5080 (1989). Those skilled in the art will appreciate that the general rule that eukaryotic ribosomes initiate translation exclusively at the 5'-proximal AUG codon is abrogated only in rare circumstances (see, e.g., Kozak PNAS 92(7): 2662-2666, (1995) and Kozak NAR 15(20): 8125-8148 (1987)).

III.) PSCA-related proteins

There are also PSCA-related proteins. Specific embodiments of PSCA proteins comprise a polypeptide having all or part of the amino acid sequence of human PSCA as shown in Figure 1, preferably Figure IA. Alternatively, embodiments of PSCA proteins comprise variant, homologous or analogous polypeptides having changes in the amino acid sequence of PSCA shown in Figure 1.

Embodiments of a PSCA polypeptide include: a PSCA polypeptide having a sequence shown in Figure 1, a peptide sequence of a PSCA as shown in Figure 1 where T is U; at least 10 consecutive nucleotides of a polypeptide having the sequence shown in Figure 1; or at least 10 consecutive peptides of a polypeptide having the sequence shown in Figure 1 where T is U.

Amino acid abbreviations are given in Table LT. Conservative amino acid substitutions can often be made in a protein without changing either the conformation or the function of the protein. Proteins according to the invention may comprise 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 conservative substitutions.

It includes a number of species-accepted variants or analogues of PSCA proteins such as polypeptides that have amino acid insertions, deletions and substitutions. PSCA variants can be prepared using methods known in the art such as site-directed mutagenesis, alanine scanning and PCR mutagenesis. Site-directed mutagenesis (Carter et al., Nucl. Acids Res., 13:4331 (1986); Zoller et al, Nucl. Acids Res., 10:6487 (1987)), cassette mutagenesis (Wells et al, Gene, 34:315 (1985)), restriction selection mutagenesis (Wells et al, Philos. Trans. R. Soc. London SerA, 317:415 (1986)) or other known techniques that can be performed on the cloned DNA to produce the PSCA variant DNA.

Scanning amino acid analysis can also be used to identify one or more amino acids along a contiguous sequence that are involved in a specific biological activity such as a protein-protein interaction. Among the preferred scanning amino acids are relatively small, neutral amino acids. Such amino acids include alanine, glycine, serine and cysteine. Alanine is typically a preferred scanning amino acid among this group because it eliminates the side chain beyond the beta carbon and is less likely to alter the main chain conformation of the variant. Alanine is also typically preferred because it is the most common amino acid. Furthermore, it is often found in both submerged and exposed positions (Creighton, The Proteins, (W.H. Freeman & Co., N.Y.); Chothia, J. Mol. Biol., 150:1 (1976)). If alanine substitution does not yield sufficient amounts of variant, an isosteric amino acid can be used.

As defined herein, PSCA variants, analogs, or homologues have the distinct attributes of having at least one epitope that is "cross-reactive" with a PSCA protein having an amino acid sequence from Figure 1. As used in this sentence, "cross-reactive" means that an antibody or T cell that specifically binds to a PSCA variant also specifically binds to a PSCA protein having an amino acid sequence shown in Figure 1. A polypeptide ceases to be a variant of a protein shown in Figure 1 when it no longer contains which any epitope capable of being recognized by an antibody or T cell that specifically binds to the initiating PSCA protein. Those skilled in the art understand that antibodies that recognize proteins bind to epitopes of varying size and a grouping on the order of about four or five amino acids, consecutive or not, is considered a typical number of amino acids in a minimal epitope. See e.g. Nair et al, J. Immunol 2000 165(12): 6949-6955; Hebbes et al, Mol Immunol (1989) 26(9):865-73; Schwartz et al, J Immunol (1985) 135(4):2598-608.

Other classes of PSCA-related protein variants share 70%, 75%, 80%, 85%, or 90% or more similarity with an amino acid sequence from Figure 1 or a fragment thereof. Another specific class of PSCA protein variants or analogs comprises one or more of the PSCA biological motifs described herein or currently known in the art. Thus, according to the present invention, analogues of PSCA fragments (nucleic acid or amino acid) which have changed functional (e.g. immunogenic) properties in relation to the starting fragment are covered. It must be recognized that motifs now or which become part of the area must be applied to the nucleic acid or amino acid sequences of Figure 1.

Polypeptides containing less than the full amino acid sequence of a PSCA protein shown in Figure 1 have been described. For example, representative embodiments of the invention include peptides/proteins having any 4, 5, 6, 7, 8, 9, 10, 11 , 12, 13, 14, 15 or more consecutive amino acids from a PSCA protein shown in Figure 1.

PSCA-related proteins are formed using standard peptide synthesis technology or using chemical cleavage procedures well known in the art. Alternatively, recombinant methods can be used to generate nucleic acid molecules encoding a PSCA-related protein. In one embodiment, nucleic acid molecules provide a means to form defined fragments of a PSCA protein (or variants, homologues or analogs thereof).

LTI.A.) Motif-bearing protein embodiments

Additional illustrative embodiments are described herein that comprise PSCA polypeptides comprising the amino acid residues of one or more of the biological motifs contained within a PSCA polypeptide sequence set forth in Figure 1. Various motifs are known in the art and a protein can be evaluated for the presence of such motifs by several publicly available Internet sites (see e.g. URL addresses: pfam.wustl.edu/; searchlauncher.bcm.tmc.edu/seq-search/struc-predicthtml; psort.ims.u-tokyo.ac.jp/; cbs .dtu.dk/; ebi.ac.uk/interpro/scan.html; expasy.ch/tools/scnpsitl.html; Epimatrix™ and Epimer™, Brown University, brown.edu/Research/TB-HIV_Lab/epimatrix/epimatrix .html; and BIMAS, bimas.dcrt.nih.gov/.).

Motif-bearing subsequences of all PSCA variant proteins are listed and identified in Tables V-xvm and XXII-LI.

Table IV(h) indicates many frequently occurring motifs based on pfam searches (see URL address pfam.wustl.edu/). The columns in Table IV(h) list (1) motif name abbreviation, (2) percent identity found among the various members of the motif family, (3) motif name or description, and (4) most common function; localization information is included if the subject is relevant for localization.

Polypeptides comprising one or more of the PSCA motifs described above are useful in elucidating the specific characteristics of a malignant phenotype in view of the observation that the PSCA motifs described above are associated with growth dysregulation and because PSCA is overexpressed in certain cancers (see e.g. .table I). Casein kinase II, cAMP and cAMP-dependent protein kinase and Protein kinase C, for example, are enzymes known to be associated with the development of the malignant phenotype (see, e.g., Chen et al, Lab Invest, 78(2): 165- 174 (1998); Gaiddon et al, Endocrinology 136(10): 4331-4338 (1995); Hall et al, Nucleic Acids Research 24(6): 1119-1126 (1996); Peterziel et al, Oncogene 18(46): 6322-6329 (1999) and 0'Brian, Oncol. Rep. 5(2): 305-309 (1998)). Furthermore, both glycosylation and myristoylation are protein modifications also associated with cancer and cancer progression (see e.g. Dennis et al., Biochem. Biophys. Acta 1473(l):21-34 (1999); Raju et al., Exp. Cell Res. 235(1 ): 145-154 (1997)). Amidation is another protein modification also associated with cancer and cancer progression (see, e.g., Treston et al., J. Nat. Cancer Inst. Monogr. (13): 169-175 (1992)).

In another embodiment, proteins herein comprise one or more of the immunoreactive epitopes identified according to species-accepted methods, such as the peptides set forth in Tables V-XVLTI and XXII-LI. CTL epitopes can be determined using specific algorithms to identify peptides within a PSCA protein that can optimally bind to specified HLA alleles (eg, Table IV; Epimatrix™ and Epimer™, Brown University, URL brown.edu/ Research/TB-HEV Lab/epimatrix/epimatrix.html; and BLMAS, URL bimas.dcrt.nih.gov/.) Furthermore, processes for identifying peptides that have sufficient binding affinity for HLA molecules and that are consistent with being immunogenic epitopes well known in the art and are performed without unnecessary experimentation. In addition, processes for identifying peptides that are immunogenic epitopes are well known in the art and are performed without unnecessary experimentation either in vitro or in vivo.

Also known in the art are principles for generating analogs of such epitopes to modulate immunogenicity. For example, one begins with an epitope bearing a CTL or HTL motif (see, e.g., the HLA class I and HLA class II motifs/supermotifs in Table IV). The epitope is analogized by substituting an amino acid in one of the specified positions and repositioning it with another amino acid specified for that position. For example, on the basis of residues defined in Table IV, one can substitute out a deleterious residue in favor of any other residue, such as a preferred residue; substituting a less preferred residue with a preferred residue; or substituting an originally occurring preferred residue with another preferred residue. Substitutions can occur at primary anchor positions or at other positions in a peptide; see e.g. table IV.

A number of references reflect the field regarding the identification and generation of epitopes in a protein of interest as well as analogs thereof. See, for example, WO 97/33602 to Chesnut et al; Sette, Immunogenetics 1999 50(3-4): 201-212; Sette etal, J. Immunol. 2001 166(2): 1389-1397; Sidney et al., Hum. Immunol. 1997 58(1): 12-20; Kondo et al, Immunogenetics 1997 45(4): 249-258; Sidney et al., J. Immunol. 1996 157(8): 3480-90; and Falk et al., Nature 351: 290-6 (1991); Hunt et al., Science 255:1261-3 (1992); Parker et al., J. Immunol. 149:3580-7 (1992); Parker et al., J. Immunol. 152:163-75 (1994)); Caste-a/., 1994 152(8): 3904-12; Borras-Cuesta et al., Hum. Immunol. 2000 61(3): 266-278; Alexander et al., J. Immunol. 2000 164(3); 164(3): 1625-1633; Alexander etal, PMID: 7895164, Ul: 95202582; O'Sullivan et al., J. Immunol. 1991 147(8): 2663-2669; Alexander et al., Immunity 1994 1(9): 751-761 and Alexander et al., Immunol. Res. 1998 18(2): 79-92.

It is further described polypeptides comprising combinations of the various motifs indicated in table(s) IV(a), IV(b), IV(c), IV(d) and IV(h) and/or, one or more of the the putative CTL epitopes in Tables V-XVffl and XXLI-LI and/or, one or more of the putative HTL epitopes in Tables XLVLII-LI and/or, one or more of the T-cell binding motifs known in the art. Preferred embodiments contain no insertions, deletions or substitutions either within the motifs or within the intervening sequences of the polypeptides. In addition, embodiments that include several of either N-terminal and/or C-terminal amino acid residues on any side of these motifs may be desirable (for example to include a larger part of the polypeptide architecture where the motif is located). Typically, the number of N-terminal and/or C-terminal amino acid residues on any side of a motif is between approx. 1 to approx. 100 amino acid residues, preferably 5 to approx. 50 amino acid residues.

PSCA-related proteins are concretized in many forms, preferably in isolated form. A purified PSCA protein molecule will be substantially free of other proteins or molecules that impair the binding of PSCA to antibody, T cell or other ligand. The nature and degree of isolation and purification will depend on the intended application. Embodiments of PSCA-related proteins include purified PSCA-related proteins and functional, soluble PSCA-related proteins. In one embodiment, a functional, soluble PSCA protein or fragment thereof retains the ability to be bound to antibody, T cell or other ligand.

Further described are PSCA proteins comprising biologically active fragments of a PSCA amino acid sequence shown in Figure 1. Such proteins exhibit properties of the nascent PSCA protein, such as the ability to elicit the formation of antibodies that specifically bind an epitope associated with the nascent the PSCA protein; to be bound by such antibodies; to induce activation of HTL or CTL; and/or to be recognized by HTL or CTL that also specifically bind to the starting protein.

PSCA-related polypeptides containing structures of particular interest may be predicted and/or identified using various analytical techniques well known in the art, including, for example, the methods of Chou-Fasman, Garnier-Robson, Kyte-Doolittle, Eisenberg, Karplus-Schultz or Jameson -Wolf assay or based on immunogenicity. Fragments containing such structures are particularly useful in the generation of subunit-specific anti-PSCA antibodies or T cells or in the identification of cellular factors that bind to PSCA. For example, hydrophilicity profiles can be generated and immunogenic peptide fragments identified using the method of Hopp, T.P. and Woods, K.R., 1981, Proe. Nati. Acad. Pollock. U.S.A. 78:3824-3828. Hydrophaticity profiles can be generated and immunogenic peptide fragments identified using the method of Kyte, J. and Doolittle, R.F., 1982, J. Mol. Biol. 157:105-132. Percent (%) available residue profiles can be generated and immunogenic peptide fragments identified using the method of Janin J., 1979, Nature 277:491-492. Average flexibility profiles can be generated and immunogenic peptide fragments identified using the method of Bhaskaran R., Ponnuswamy P.K., 1988, Int. J. Pept. Protein Res. 32:242-255. Beta turn profiles can be generated and immunogenic peptide fragments identified using the method of Deleage, G, Roux B., 1987, Protein Engineering 1:289-294.

CTL epitopes can be determined using specific algorithms to identify peptides within a PSCA protein that can optimally bind to specified HLA alleles { e.g. using the SYFPEITHI site at the "world wide web" URL syfpeithi.bmi-heidelberg.com/; the listings in Table IV(A)-(E); Epimatrix™ and Epimer™, Brown University, URL (brown.edu/Research/TB-HIV_Lab/epimatrix/epimatrix.html); and BLMAS, URL bimas.dcrt.nih.gov/). To illustrate this, peptide epitopes from PSCA as presented in the context of human MHC class I molecules are shown. were HLA-Al, A2, A3, Al 1, A24, B7 and B35 predicted (see e.g. Tables V-XVUI, XXLT-LI). Specifically, the complete amino acid sequence of the PSCA protein and relevant parts of other variants, i.e. for HLA class I predictions 9 flanking residues on any side of a point mutation or exon splicing and for HLA class Ll predictions 14 flanking residues on any side of a point mutation or exon splicing corresponding to that variant entered into the HLA peptide motif search algorithm found in the "Bioinformatics and Molecular Analysis Section"

(BLMAS) website listed above; in addition to the site SYFPEITHI at URL syfpeithi.bmi-heidelberg.com/.

The HLA peptide motif search algorithm was developed by Dr. Ken Parker based on binding of specific peptide sequences in the pit of HLA class I molecules, particularly HLA-A2 (see, e.g., Falk et al, Nature 351: 290-6 (1991); Hunter et al., Science 255:1261-3 (1992); Parker et al., J. Immunol. 149:3580-7

(1992); Parker et al., J. Immunol. 152:163-75 (1994)). This algorithm allows the localization and ranking of 8-mer, 9-mer and 10-mer peptides from a complete protein sequence for predicted binding to HLA-A2 as well as a variety of other HLA class I molecules. Many HLA class I binding peptides are 8-, 9-, 10-, or 11-mers. For example, for class I-HLA-A2, the epitopes preferably contain a leucine (L) or methionine (M) at position 2 and a valine (V) or leucine (L) at the C-terminus (see, e.g., Parker et al, J. Immunol. 149:3580-7 (1992)). Selected results from PSCA predicted binding peptides are shown in Tables V-XVJJI and XXII-LI herein. In Tables V-XVILT and XXII-XLVin, selected 9-mer and 10-mer candidates for each family member are shown along with their location, amino acid sequence of each specific peptide and calculated binding score. In Tables XLVLII-LI, selected 15-mer candidates for each family member are shown along with their localization, the amino acid sequence of each specific peptide and calculated binding score. The binding score corresponds to the calculated half-life for dissociation of complexes containing the peptide at 37°C at pH 6.5. Peptides with the highest binding score are predicted to be most tightly bound to HLA class I on the cell surface for the greatest period of time and thus represent the best immunogenic target for T cell recognition.

Actual binding of peptides to an HLA allele can be evaluated by stabilization of HLA expression on the antigen-processing-defective cell line T2 (see, e.g., Xue et al., Prostate 30:73-8

(1997) and Peshwa et al, Prostate 36:129-38 (1998)). Immunogenicity of specific peptides can be evaluated in vitro by stimulation of CD8+ cytotoxic T lymphocytes (CTL) in the presence of antigen presenting cells such as dendritic cells.

It shall be recognized that each epitope predicted at the BIMAS site, Epimer™ and Epimatrix™ sites or the specified HLA class I or class LT motifs available at the site or becoming part of the site as indicated in Table IV (or determined using the "world wide web" site URL syfpeithi.bmi-heidelberg.com/ or BIMAS, bimas.dcrt.nih.gov/) shall be "used" for a PSCA protein according to the invention. As used in this context, "used" means that a PSCA protein has been evaluated, e.g. visually or by computer-based pattern matching methods, as recognized by those skilled in the relevant field. Each subsequence of a PSCA protein of 8, 9, 10 or 11 amino acid residues carrying an HLA class I motif or a subsequence of 9 or more amino acid residues carrying an HLA class LT motif is within the scope of the invention.

LTI.B.) Expression of PSCA-related proteins

In an embodiment described in the examples that follow, PSCA may be conveniently expressed in cells (such as 293T cells) transfected with a commercially available expression vector such as a CMV-driven expression vector encoding PSCA with a C-terminal 6XHis and MYC tag (pcDNA3.1/mycHIS, Invitrogen or Tag5, GenHunter Corporation, Nashville TN). The Tag5 vector provides an IgGK secretion signal that can be used to facilitate the production of a secreted PSCA protein in transfected cells. The secreted HES-labeled PSCA in the culture media can be purified, e.g. using a nickel column using standard techniques.

LTI.C.) Modifications of PSCA-related proteins

Modifications of PSCA-related proteins such as covalent modifications are included within the scope of the present invention. One type of covalent modification involves reacting the targeted amino acid residues of a PSCA polypeptide with an organic derivatizing agent that can be reacted with selected side chains or the N- or C-terminal residues of a PSCA protein. Another type of covalent modification of a PSCA polypeptide encompassed within the scope of the present invention comprises changing the natural glycosylation pattern of a protein according to the invention. Another type of covalent modification of PSCA involves linking a PSCA polypeptide to one of a number of non-proteinaceous polymers, e.g. polyethylene glycol (PEG), polypropylene glycol or polyoxyalkylenes, as disclosed in US Patent No. 4,640,835; 4,496,689; 4,301,144; 4,670,417; 4,791,192 or 4,179,337.

The PSCA-related proteins can also be modified to form a chimeric molecule comprising PSCA fused to another, heterologous polypeptide or amino acid sequence. Such a chimeric molecule can be synthesized chemically or recombinantly. A chimeric molecule can have a protein according to the invention condensed to another tumor-associated antigen or fragment thereof. Alternatively, a protein may comprise a fusion of fragments of a PSCA (amino or nucleic acid) sequence so that a molecule is formed which is not, throughout its length, directly homologous to the amino or nucleic acid sequences shown in Figure 1. Such a chimeric molecule may include multiples to the same subsequence of PSCA. A chimeric molecule may comprise a fusion of a PSCA-related protein with a polyhistidine epitope tag, which provides an epitope to which immobilized nickel can bind selectively with cytokines or with growth factors. The epitope tag is generally located at the amino or carboxyl terminus of a PSCA protein. In an alternative embodiment, the chimeric molecule may comprise a fusion of a PSCA-related protein with an immunoglobulin or a special region of an immunoglobulin. For a bivalent form of the chimeric molecule (also referred to as an "immunoadhesin"), such a fusion could be to the Fc region of an IgG molecule. The Ig fusions preferably comprise the substitution of a soluble (transmembrane domain deleted or - inactivated) form of a PSCA polypeptide instead of at least one variable region within an Ig molecule. In a preferred embodiment, the immunoglobulin fusion comprises the hinged CH2 and CH3 or hinged CHI, CH2 and CH3 regions of an IgGI molecule. For the production of immunoglobulin fusions see e.g. US Patent No. 5,428,130 published June 27, 1995.

LTI.D.) Applications of PSCA-related proteins

The proteins herein have several different specific uses. As PSCA is highly expressed in prostate and other cancers, PSCA-related proteins are used in procedures that determine the status of PSCA gene products in normal versus cancerous tissue, thereby elucidating the malignant phenotype. Typically, polypeptides from specific regions of a PSCA protein are used to assess the presence of perturbations (such as deletions, insertions, point mutations, etc.) in those regions (such as regions containing one or more motifs). Examples of experiments using antibodies or T cells targeting PSCA-related proteins comprising the amino acid residues of one or more of the biological motifs contained within a PSCA polypeptide sequence to evaluate the characteristics of this region in normal versus cancerous tissue or to elicit an immune response to the epitope. Alternatively, PSCA-related proteins containing the amino acid residues of one or more of the biological motifs in a PSCA protein are used to screen for factors that interact with that region of PSCA.

PSCA protein fragments/subsequences are particularly useful in the generation and characterization of domain-specific antibodies (eg, antibodies that recognize an extracellular or intracellular epitope of a PSCA protein), to identify agents or cellular factors that bind to PSCA or a particular structural domain thereof and in various therapeutic and diagnostic contexts, including but not limited to diagnostic tests, cancer vaccines and methods for making such vaccines.

Proteins encoded by the PSCA genes or by analogs, homologues or fragments thereof have a variety of applications, including but not limited to the generation of antibodies and in methods of identifying ligands and other agents and cellular constituents that bind to a PSCA gene product. Antibodies induced against a PSCA protein or fragment thereof are useful in diagnostic and prognostic assays and imaging methods in the treatment of human cancer characterized by expression of PSCA protein, such as those listed in Table I. Such antibodies can be expressed intracellularly and used in methods of treatment of patients with such cancer. PSCA-related nucleic acids or proteins have also been used in generating HTL or CTL responses.

Various immunological assays applicable to the detection of PSCA proteins are used, including but not limited to various types of radioimmunoassays, enzyme-linked immunosorbent assay (ELISA), enzyme-linked immunofluorescent assay (ELIFA), immunocytochemical methods and the like. Antibodies can be labeled and used as immunological imaging reagents capable of detecting PSCA-expressing cells (eg, in radioscintigraphic imaging procedures). PSCA proteins are also particularly useful in the formation of cancer vaccines, as further described herein.

IV.) PSCA antibodies

Antibodies that bind to PSCA-related proteins have also been described. Preferred antibodies bind specifically to a PSCA-related protein and do not bind (or bind weakly) to peptides or proteins that are not PSCA-related proteins under physiological conditions. In this context, examples of physiological conditions include: 1) phosphate buffered saline; 2) Tris-buffered saline containing 25 mM Tris and 150 mM NaCl; or normal saline (0.9% NaCl); 4) animal serum such as human serum; or 5) a combination of any of 1) through 4); these reactions preferably take place at pH 7.5, alternatively in a range of pH 7.0 to 8.0 or alternatively in a range of pH 6.5 to 8.5; these reactions also take place at a temperature between 4°C and 37°C. For example, antibodies that bind PSCA may bind PSCA-related proteins such as their homologues or analogs.

PSCA antibodies are particularly useful in cancer (see, eg, Table I) diagnostic and prognostic assays and imaging procedures. Similarly, such antibodies are useful in the treatment, diagnosis and/or prognosis of prostate and other cancers, to the extent that PSCA is also expressed or overexpressed in these other cancers. Furthermore, intracellularly expressed antibodies (eg, single chain antibodies) are therapeutically useful for the treatment of cancers where the expression of PSCA is involved, such as advanced or metastatic prostate cancer or other advanced or metastatic cancers.

Various immunological assays useful for the detection and quantification of PSCA and mutant PSCA-related proteins have been described. Such assays may include one or more PSCA antibodies capable of recognizing and binding a PSCA-related protein, as appropriate. These tests are performed within various immunological test formats well known in the field, including but not limited to different types of radioimmunoassays, enzyme-linked immunosorbent assay (ELISA), enzyme-linked immunofluorescent assay (ELIFA) and the like.

Non-antibody immunological assays also include T-cell immunogenicity assays (inhibitory or stimulatory) as well as major histocompatibility complex (MHC) binding assays.

In addition, immunological imaging methods capable of detecting prostate cancer and other cancers expressing PSCA are also described, including but not limited to radioscintigraphic imaging methods using labeled PSCA antibodies. Such assays are clinically applicable in the detection, monitoring and prognosis of PSCA expressing cancers such as prostate cancer.

PSCA antibodies have also been used in methods for purification of a PSCA-related protein and for isolating PSCA homologues and related molecules. For example, a method for purifying a PSCA-related protein comprises incubating a PSCA antibody, which is coupled to a solid matrix, with a lysate or other solution containing a PSCA-related protein under conditions that allow the PSCA antibody to bind to the PSCA-related protein; washing the solid matrix to remove impurities; and elution of the PSCA-related protein from the coupled antibody. Other uses of PSCA antibodies according to the invention include the generation of anti-idiotypic antibodies that mimic a PSCA protein.

Various methods for the production of antibodies are well known in the field. For example, antibodies can be produced by immunization of a suitable mammalian host using a PSCA-related protein, peptide or fragment, in isolated or immunoconjugated form (Antibodies: A Laboratory Manual, CSH Press, Eds., Harlow and Lane (1988); Harlow , Antibodies, Cold Spring Harbor Press, NY (1989)). In addition, fusion proteins of PSCA can also be used, such as a PSCA-GST fusion protein. In a particular embodiment, a GST fusion protein comprising all or most of the amino acid sequence of Figure 1 is produced, then used as an immunogen to generate appropriate antibodies. In another embodiment, a PSCA-related protein is synthesized and used as an immunogen.

In addition, naked DNA immunization techniques known in the art are used (with or without purified PSCA-related protein or PSCA-expressing cells) to generate an immune response to the encoded immunogen (for review, see Donnelly et al., 1997, Ann. Rev. Immunol. 15: 617-648).

The amino acid sequence of a PSCA protein as shown in Figure 1 can be analyzed to select specific regions of the PSCA protein for generation of antibodies. For example, hydrophobicity and hydrophilicity analyzes of a PSCA amino acid sequence are used to identify hydrophilic regions of the PSCA structure. Regions of a PSCA protein that exhibit immunogenic structure, as well as other regions and domains, can be readily identified using various other methods known in the art, such as Chou-Fasman, Garnier-Robson, Kyte-Doolittle, Eisenberg -, Karplus-Schultz or Jameson-Wolf analysis. Hydrophilicity profiles can be formed using the method of Hopp, T.P. and Woods, K.R., 1981, Proe. Nati. Acad. Pollock. U.S.A. 78:3824-3828. Hydrophaticity profiles can be generated using the method of Kyte, J. and Doolittle, R.F., 1982, J. Mol. Biol. 157:105-132. Percent (%) available residue profiles can be generated using the method of Janin J., 1979, Nature 277:491-492. Average flexibility profiles can be generated using the method of BhaskaranR., Ponnuswamy P.K., 1988, Int. J. Pept. Protein Res. 32:242-255. Beta turn profiles can be generated using the method of Deleage, G, Roux B., 1987, Protein Engineering 1:289-294. Thus, each region identified by any of these programs or methods is within the scope of the present invention. Preferred methods for generating PSCA antibodies are further illustrated by means of the examples provided herein. Methods for preparing a protein or polypeptide for use as an immunogen are well known in the art. Also well known in the field are methods for producing immunogenic conjugates of a protein with a carrier, such as BSA, KLH or other carrier protein. In some cases, direct conjugation using, for example, carbodiimide reagents is employed; in other cases, linking reagents such as those supplied by Pierce Chemical Co., Rockford, IL are effective. Administration of a PSCA immunogen is often accomplished by injection over a suitable period of time and with the use of a suitable adjuvant, as is understood in the art. During the course of the immunization program, antibody titers may be taken to determine the adequacy of antibody formation.

PSCA monoclonal antibodies can be produced by various agents well known in the art. For example, immortalized cell lines secreting a desired monoclonal antibody are prepared using the standard hybridoma technology of Kohler and Milstein or modifications such as immortalization of antibody-producing B cells, which are generally known. Immortalized cell lines secreting the desired antibodies are screened by immunoassay where the antigen is a PSCA-related protein. Once the appropriate immortalized cell culture is identified, the cells can be expanded and antibodies produced either from in v/Yro cultures or from ascites fluid.

The antibodies or fragments according to the invention can also be produced by recombinant means. Regions that specifically bind to the desired regions of a PSCA protein can also be produced in the context of chimeric or complementarity-determining region (CDR)-transplanted antibodies of multiple species origin. Humanized or human PSCA antibodies can also be produced and are preferred for use in therapeutic contexts. Methods for humanizing murine and other non-human antibodies by substituting one or more of the non-human antibody CDRs for corresponding human antibody sequences are well known (see, for example, Jones et al., 1986, Nature 321: 522-525; Riechmann et al., 1988, Nature 332: 323-327; Verhoeyen et al, 1988, Science 239: 1534-1536). See also, Carter et al., 1993, Proe. Nati. Acad. Pollock. USA 89: 4285 and Sims et al., 1993, J. Immunol. 151: 2296.

Methods for producing fully human monoclonal antibodies include phage display and transgenic methods (for review see Vaughan et al, 1998, Nature Biotechnology 16: 535-539). Complete human PSCA monoclonal antibodies can be generated using cloning technologies using large human Ig gene combinatorial libraries (ie phage display) (Griffiths and Hoogenboom, Building an in vitro immune system: human antibodies from phage display libraries. In: Protein Engineering of Antibody Molecules for Prophylactic and Therapeutic Applications in Man, Clark, M. (Ed.), Nottingham Academic, pp 45-64 (1993); Burton and Barbas, Human Antibodies from combinatorial libraries. Id, pp 65-82). Fully human PSCA monoclonal antibodies can also be produced using transgenic mice engineered to contain human immunoglobulin gene loci as described in PCT Patent Application WO98/24893, Kucherlapati and Jakobovits et al, published December 3, 1997 (see also Jakobovits, 1998, Exp. Opin. Invest. Drugs 7(4): 607-614; US Patents 6,162,963 published December 19, 2000; 6,150,584 published November 12, 2000; and 6,114,598 published September 5, 2000). This method avoids the in v/7/*o manipulation required with phage display technology and efficiently provides high affinity authentic human antibodies.

Reactivity of PSCA antibodies to a PSCA-related protein can be established by several well-known means, including Western blot, immunoprecipitation, ELISA and FACS assays using, as appropriate, PSCA-related proteins, PSCA-expressing cells or extracts thereof. A PSCA antibody or fragment thereof may be labeled with a detectable marker or conjugated to another molecule. Suitable detectable labels include, but are not limited to, a radioisotope, a fluorescent compound, a bioluminescent compound, chemiluminescent compound, a metal chelator, or an enzyme. Furthermore, bispecific antibodies specific for two or more PSCA epitopes are generated using methods generally known in the art. Homodimeric antibodies can also be formed by cross-linking techniques known in the art (eg, Wolff et al, Cancer Res. 53: 2560-2565).

Described are monoclonal antibodies identified as Hal-1.16, Hal-5.99, Hal-4.117, Hal-4.20, Hal-4.121, Hal-4.37 which were sent (via Federal Express) to the "American Type Culture Collection" (ATCC), P.O. Box 1549, Manassas, VA 20108 04-May-2005 and designated accession numbers PTA-6698 and PTA-6703 and PTA-6699 and PTA-6700 and PTA-6701 and PTA-6702 respectively.

V.) PSCA cellular immune responses

The mechanism by which T cells recognize antigens is outlined. Effective

peptide epitope vaccine compositions elicit therapeutic or prophylactic immune responses in very broad segments of the world population. For an understanding of the value and effectiveness of compositions of the invention that elicit cellular immune responses, a brief overview of immunology-related technology is provided.

A complex of an HLA molecule and a peptidic antigen acts as the ligand recognized by HLA-restricted T cells (Buus, S. et al., Cell 47:1071, 1986; Babbitt, B. P. et al., Nature 317:359, 1985; Townsend, A. andBodmer, H., Annu. Rev. Immunol. 7:601, 1989; Germain, R.N., Annu. Rev. Immunol. 11:403, 1993). Through examination of simple amino acid substituted antigen analogs and the sequencing of endogenously bound, naturally processed peptides, critical residues corresponding to motifs required for specific binding to HLA antigen molecules have been identified and are listed in Table IV (see also, e.g., Southwood, et al, J .Immunol. 160:3363,1998; Rammensee, etal, Immunogenetics 41:178, 1995; Rammensee etal, SYFPEITHI, accessed via "world wide web" at URL (134.2.96.221/scripts.hlaserver.dll/home.htm); Sette, A. and Sidney, J. Curr. Opin. Immunol. 10:478, 1998; Engelhard, V. H., Curr. Opin. Immunol. 6:13, 1994; Sette, A. and Grey, H. M., Curr. Opin. Immunol. 4:79, 1992; Sinigaglia, F. and Hammer, J. Curr. Biol. 6:52, 1994; RuppertÉrta/., Cell 74:929-937, 1993; Kondo et al, J. Immunol. 155:4307 -4312, 1995; Sidney et al, J. Immunol. 157:3480-3490, 1996; Sidney et al, Human Immunol. 45:79-93, 1996; Sette, A. and Sidney, J. Immunogenetics 1999 Nov; 50(3 -4):201-12, Review).

Furthermore, X-ray crystallographic analyzes of HLA-peptide complexes have revealed pockets within the peptide-binding cleft/pit of HLA molecules that fit, in an allele-specific mode, residues originating in peptide ligands; these residues in turn determine the HLA binding capacity of the peptides where they are present, (see, e.g., Madden, D.R. Annu. Rev. Immunol. 13:587, 1995; Smith, et al., Immunity 4:203, 1996; Fremontef a/., Immunity 8:305, 1998; Stern etal, Structure 2:245, 1994; Jones, E. Y. Curr. Opin. Immunol. 9:75, 1997; Brown, J. H. etal, Nature 364:33, 1993; Guo, H. C. etal, Proe. Nati. Acad. Sei. USA 90:8053, 1993; Guo, H. C. etal, Nature 360:364, 1992; Silver, M. L. etal, Nature 360:367, 1992; Matsumura, M. etal, Science 257 :927, 1992; Madden et al, Cell 70:1035, 1992; Fremont, D. H. et al, Science 257:919, 1992; Såper, M. A., Bjorkman, P. J. and Wiley, D. C, J. Mol. Biol. 219:277 , 1991.)

Accordingly, the definition of class I and class II allele-specific HLA binding motifs or class I or class II supermotifs allows the identification of regions within a protein that are consistent with binding to particular HLA antigen(s).

Thus, by a method for HLA motif identification candidates for epitope-based vaccines are identified; such candidates may further be evaluated by HLA peptide binding assays to determine binding affinity and/or the time period for association of the epitope and its corresponding HLA molecule. Further confirmatory work can be performed to select, among these vaccine candidates, epitopes with preferred characteristics in terms of population coverage and/or immunogenicity.

Different strategies can be used to assess cellular immunogenicity including:

1) Evaluation of primary T cell cultures from normal individuals (see, e.g., Wentworth, P. A. et al., Mol. Immunol. 32:603, 1995; Celis, E. et al., Proe. Nati. Acad. Sei. USA 91:2105, 1994; Tsai, V. et al., J. Immunol. 158:1796, 1997; Kawashima, I. et al., Human Immunol. 59:1, 1998). This procedure involves stimulation of peripheral blood lymphocytes (PBL) from normal patients with a test peptide in the presence of antigen-presenting cells in vitro over a period of many weeks. T cells specific for the peptide are activated during this time and are detected using e.g. a lymphokine or 51Cr release assay involving peptide sensitized target cells. 2) Immunization of HLA transgenic mice (see, e.g., Wentworth, P. A. et al., J. Immunol. 26:97, 1996; Wentworth, P. A. et al., Int. Immunol. 8:651, 1996; Alexander, J. et al, J. Immunol. 159:4753,1997). For example, in such methods, peptides in incomplete Freund's adjuvant are administered subcutaneously to HLA transgenic mice. Several weeks after immunization, spleen cells are removed and cultured in vitro in the presence of test peptide for approximately one week. Peptide-specific T cells are detected using e.g. a 51Cr release assay involving peptide-sensitized target cells and target cells expressing endogenously formed antigen. 3) Demonstration of revived T-cell responses from immune individuals either effectively vaccinated and/or from chronically ill patients (see, e.g., Rehermann, B. et al., J. Exp. Med. 181:1047, 1995; Doolan , D. L. et al., Immunity 7:97,1997; Bertoni, R. et al., J. Clin. Invest. 100:503, 1997; Threlkeld, S. C. et al., J. Immunol. 159:1648,1997; Diepolder, H.M. et al., J. Virol. 71:6011, 1997). Accordingly, reawakened responses are detected by culturing PBL from patients who have been exposed to the antigen due to disease and thus have formed an immune response "naturally" or from patients who were vaccinated against the antigen. PBL from patients are cultured in vitro for 1-2 weeks in the presence of test peptide plus antigen presenting cells (APC) to allow activation of "memory" T cells, compared to "natural" T cells. At the end of the culture period, T cell activity is detected using assays including 51 Cr release involving peptide sensitized targets, T cell proliferation or lymphokine release.

VI.) PSCA transgenic animals

Nucleic acids encoding a PSCA-related protein can also be used to generate either transgenic animals or "knock out" animals which in turn are useful in the development and screening of therapeutically useful reagents. According to established techniques, cDNA encoding PSCA can be used to clone genomic DNA encoding PSCA. The cloned genomic sequences can then be used to generate transgenic animals containing cells expressing DNA encoding PSCA. Methods for generating transgenic animals, particularly animals such as mice or rats, have become conventional in the art and are described, for example, in US Patent Nos. 4,736,866 published April 12, 1988 and 4,870,009 published September 26, 1989. Typically, particular cells be targeted for PSCA transgene introduction with tissue-specific enhancers.

Transgenic animals comprising a copy of a transgene that codes for PSCA can be used to investigate the effect of increased expression of DNA that codes for PSCA. Such animals can be used as test animals for reagents believed to confer protection from, for example, pathological disorders associated with its overexpression. According to this aspect, an animal is treated with a reagent, and a reduced incidence of a pathological condition, compared to untreated animals carrying the transgene, would indicate a potential for therapeutic intervention of the pathological condition.

Alternatively, non-human homologs of PSCA can be used to construct a PSCA "knock out" animal that has a defective or altered gene encoding PSCA as a result of homologous recombination between the endogenous gene encoding PSCA and altered genomic DNA that codes for PSCA introduced into an embryonic cell of the animal. For example, cDNA encoding PSCA can be used to clone genomic DNA encoding PSCA according to established techniques. A portion of the genomic DNA encoding PSCA can be deleted or replaced with another gene, such as a gene encoding a selectable marker that can be used to monitor integration. Typically, many kilobases of unchanged flanking DNA (at both the 5' and 3' ends) are included in the vector (see, e.g., Thomas and Capecchi, Cell, 51:503 (1987) for a description of homologous recombination vectors). The vector is introduced into an embryonic stem cell line (e.g. by electroporation) and cells where the introduced DNA has homologously recombined with the endogenous DNA are selected (see e.g. Li et al., Cell, 69:915 (1992)). The selected cells are then injected into a blastocyst of an animal (eg, a mouse or rat) to form aggregation chimeras (see, eg, Bradley, in Teratocarcinomas and Embryonic Stem Cells: A Practical Approach, E. J. Robertson, ed. (LRL, Oxford, 1987), pp. 113-152). A chimeric embryo can then be implanted into a suitable pseudopregnant female fetus and the embryo brought to term to create a "knock out" animal. Offspring carrying the homologously recombined DNA in their germ cells can be identified by standard techniques and used to breed animals where all cells of the animal contain the homologously recombined DNA. "Knock out" animals can be characterized, for example, for their ability to defend against certain pathological conditions or for their development of pathological conditions due to the absence of a PSCA polypeptide.

VII.) Procedures for the detection of PSCA

Methods for detecting PSCA polynucleotides and PSCA-related proteins are described, as well as methods for identifying a cell expressing PSCA. The expression profile of PSCA makes it a diagnostic marker for metastatic disease. Accordingly, the status of PSCA gene products provides information useful in predicting a number of factors including susceptibility to advanced stage disease, rate of progression and/or tumor aggressiveness. As described in detail herein, the status of PSCA gene products in patient samples can be analyzed by a variety of protocols well known in the art including immunohistochemical analysis, the variety of Northern blotting techniques including in situ hybridization, RT-PCR analysis (eg on laser-captured microdissected samples), Western blot analysis and tissue array analysis.

Experiments have been described for the detection of PSCA polynucleotides in a biological sample, such as serum, bone, prostate and other tissues, urine, seminal fluid, cell preparations and the like. Detectable PSCA polynucleotides include, for example, a PSCA gene or fragment thereof, PSCA mRNA, alternative splice variant of PSCA mRNAs and recombinant DNA or RNA molecules containing a PSCA polynucleotide. Several methods for amplifying and/or detecting the presence of PSCA polynucleotides are well known in the art and can be used in the practice of this aspect of the invention.

In one embodiment, a method for detecting a PSCA mRNA in a biological sample comprises producing cDNA from the sample by reverse transcription using at least one primer; amplifying the cDNA so produced using PSCA polynucleotides as sense and antisense primers to amplify PSCA cDNAs therein; and detecting the presence of the amplified PSCA cDNA. Optionally, the sequence of the amplified PSCA cDNA can be determined.

In another embodiment, a method for detecting a PSCA gene in a biological sample first comprises isolating genomic DNA from the sample; amplifying the isolated genomic DNA using PSCA polynucleotides as sense and antisense primers; and detecting the presence of the amplified PSCA gene. Any number of suitable sense and antisense probe combinations can be designed from a PSCA nucleotide sequence (see, eg, Figure 1) and used for this purpose.

It is also possible to detect the presence of a PSCA protein in a tissue or other biological sample such as serum, seminal fluid, bone, prostate, urine, cell preparations and the like. Methods for detecting a PSCA-related protein are also well known and include, for example, immunoprecipitation, immunohistochemical analysis, Western blot analysis, molecular binding assays, ELISA, ELLFA and the like. For example, a method for detecting the presence of a PSCA-related protein in a biological sample comprises first contacting the sample with a PSCA antibody, a PSCA-reactive fragment thereof or a recombinant protein containing an antigen-binding region of a PSCA antibody ; and then detecting the binding of PSCA-related protein in the sample.

Methods of identifying a cell expressing PSCA are also possible. In one embodiment, an attempt to identify a cell expressing a PSCA gene comprises detecting the presence of PSCA mRNA in the cell. Methods for the detection of particular mRNAs in cells are well known and include, for example, hybridization experiments using complementary DNA probes (such as in situ hybridization using labeled PSCA riboprobes, Northern blots and related techniques) and various nucleic acid amplification experiments (such as RT-PCR using complementary primers specific for PSCA and other amplification type detection methods, such as for example branched DNA, SISBA, TMA and the like). Alternatively, an attempt to identify a cell expressing a PSCA gene comprises detecting the presence of PSCA-related protein in the cell or secreted from the cell. Various methods for the detection of proteins are well known in the art and are used for the detection of PSCA-related proteins and cells expressing PSCA-related proteins.

PSCA expression analysis is also useful as a tool to identify and evaluate agents that modulate PSCA gene expression. For example, PSCA expression is significantly upregulated in prostate cancer and is expressed in cancers in the tissues listed in Table I. Identification of a molecule or biological agent that inhibits PSCA expression or overexpression in cancer cells is of therapeutic value. For example, such an agent can be identified using a screen that quantifies PSCA expression by RT-PCR, nucleic acid hybridization or antibody binding.

VIII.) Procedures for monitoring the status of PSCA-related genes and their products

Oncogenesis is known to be a multistep process where cellular growth is progressively dysregulated and cells develop from a normal physiological state to precancerous and then cancerous states (see e.g. Ålers etal, Lab Invest. 77(5): 437-438 ( 1997) and Isaacs et al, Cancer Surv. 23: 19-32 (1995)). In this context, examination of a biological sample for evidence of dysregulated cell growth (such as aberrant PSCA expression in cancer) allows early detection of such aberrant physiology, before a pathological condition such as cancer has progressed to a stage where therapeutic options are more limited and or the prognosis is worse. In such investigations, the status of PSCA in an interesting biological sample can be compared, for example, with the status of PSCA in a corresponding normal sample (e.g. a sample from that individual or alternatively another individual not affected by a pathology). A change in the status of PSCA in the biological sample (compared to the normal sample) provides evidence of dysregulated cellular growth. In addition to using a biological sample that is not affected by a pathology as a normal sample, one can also use a predetermined normative value such as a predetermined normal level of mRNA expression (see, e.g., Grever et al, J. Comp. Neurol .1996 Dec 9;376(2):306-14 and US Patent No. 5,837,501) to compare PSCA status in a sample.

The term "status" in this context is used according to its field-accepted meaning and denotes the relationship or condition of a gene and its products. Typically, those skilled in the art use several parameters to judge the relationship or condition of a gene and its products. These include, but are not limited to, the location of expressed gene products (including the location of PSCA-expressing cells) as well as the level and biological activity of expressed gene products (such as PSCA mRNA, polynucleotides and polypeptides). Typically, a change in the status of PSCA includes a change in the localization of PSCA and/or PSCA-expressing cells and/or an increase in PSCA mRNA and/or protein expression.

PSCA status in a sample can be analyzed by several means well known in the art, including without limitation immunohistochemical analysis, in situ hybridization, RT-PCR analysis on laser-captured microdissected samples, Western blot analysis, and tissue array analysis. Typically, protocols for evaluating the status of a PSCA gene and gene products are found, for example, in Ausubel et al. eds., 1995, Current Protocols in Molecular Biology, Units 2 (Northern blotting), 4 (Southern blotting), 15 (immunoblotting) and 18 (PCR analysis). Thus, the status of PSCA in a biological sample is evaluated by various methods used by those skilled in the art including, but not limited to, genomic Southern analysis (to examine, for example, disturbances in a PSCA gene), Northern analysis and/or PCR analysis of PSCA mRNA (to examine, for example, changes in the polynucleotide sequences or expression levels of PSCA mRNAs) and Western and/or immunohistochemical analysis (to examine, for example, changes in polypeptide sequences, changes in polypeptide localization within a sample, changes in expression levels of PSCA proteins and/or associations of PSCA proteins with polypeptide binding partners). Detectable PSCA polynucleotides include, for example, a PSCA gene or fragment thereof, PSCA mRNA, alternative splice variants, PSCA mRNAs and recombinant DNA or RNA molecules containing a PSCA polynucleotide.

The expression profile of PSCA makes it a diagnostic marker for local and/or metastatic disease and provides information on the growth or oncogenic potential of a biological sample. In particular, the status of PSCA provides information useful in predicting susceptibility to particular disease stages, progression and/or tumor aggressiveness. The invention provides methods and assays for determining PSCA status and diagnosing cancers that express PSCA, such as cancers in the tissues listed in Table I. For example, because PSCA mRNA is so highly expressed in prostate and other cancers relative to normal prostate tissue assays evaluating the levels of PSCA mRNA transcripts or proteins in a biological sample can be used to diagnose a disease associated with PSCA dysregulation and can provide prognostic information useful in defining appropriate therapeutic options.

The term status of PSCA provides information including the presence, stage and localization of dysplastic, precancerous and cancerous cells, to predict susceptibility to different stages of disease and/or to measure tumor aggressiveness. Furthermore, its expression profile makes it useful as an imaging reagent for metastatic disease. Accordingly, one aspect of the invention relates to the various molecular prognostic and diagnostic methods for investigating the status of PSCA in biological samples such as those from individuals suffering from or suspected of suffering from a pathology characterized by dysregulated cellular growth, such as cancer.

As described above, the status of PSCA in a biological sample can be investigated by several well-known procedures in the field. For example, the status of PSCA in a biological sample taken from a specific location in the body can be examined by evaluating the sample for the presence or absence of PSCA-expressing cells (eg, those expressing PSCA mRNAs or proteins). This examination can provide evidence of dysregulated cellular growth, for example when PSCA-expressing cells are found in a biological sample that does not normally contain such cells (such as a lymph node), because such changes in the status of PSCA in a biological sample are often associated with dysregulated cellular growth. Specifically, one indicator of dysregulated cellular growth is the metastases of cancer cells from an organ of origin (such as the prostate) to a different area of the body (such as a lymph node). In this context, evidence of dysregulated cellular growth is important, for example because occult lymph node metastases can be detected in a significant proportion of patients with prostate cancer and such metastases are associated with known predictors of disease progression (see e.g. Murphy et al, Prostate 42(4): 315-317 (2000); Su et al, Semin. Surg. Oncol. 18(1): 17-28 (2000) and Freeman et al, J Uroi 1995 Aug 154(2 Pt l):474-8).

Methods are described for monitoring PSCA gene products by determining the status of PSCA gene products expressed by cells from an individual suspected of having a disease associated with dysregulated cell growth (such as hyperplasia or cancer) and then comparing the status so determined against the status to PSCA gene products in a corresponding normal sample. The presence of aberrant PSCA gene products in the test sample relative to the normal sample provides an indication of the presence of dysregulated cell growth in the cells of the subject.

Tests useful for determining the presence of cancer in an individual are described, comprising detecting a significant increase in PSCA mRNA or protein expression in a test cell or tissue sample relative to expression levels in the corresponding normal cell or tissue. For example, the presence of PSCA mRNA can be evaluated in tissues including but not limited to those listed in Table I. The presence of significant PSCA expression in any of these tissues is useful to indicate the occurrence, presence and/or severity of a cancer, since the corresponding normal tissues do not express PSCA mRNA or express it at lower levels.

In a related embodiment, PSCA status is determined at the protein level rather than at the nucleic acid level. For example, such a method comprises determining the level of PSCA protein expressed by cells in a test tissue sample and comparing the level so determined to the level of PSCA expressed in a corresponding normal sample. In one embodiment, the presence of PSCA protein is evaluated, for example, using immunohistochemical methods. PSCA antibodies or binding partners capable of detecting PSCA protein expression are used in a variety of assay formats well known in the art for this purpose.

In a further embodiment, one can evaluate the status of PSCA nucleotide and amino acid sequences in a biological sample to identify perturbations in the structure of these molecules. These disturbances may include insertions, deletions, substitutions and the like. Such evaluations are useful because disturbances in the nucleotide and amino acid sequences have been observed in a large number of proteins associated with a growth dysregulated phenotype (see, e.g., Marrogi et al., 1999, J. Cutan. Pathol. 26(8):369-378). For example, a mutation in the sequence of PSCA may be indicative of the presence or progression of a tumor. Such tests therefore have diagnostic and predictive value where a mutation in PSCA indicates a potential loss of function or increase in tumor growth.

A number of tests for observing disturbances in nucleotide and amino acid sequences are well known in the art. For example, the size and structure of nucleic acid or amino acid sequences of PSCA gene products observed by Northern, Southern, Western, PCR and DNA sequencing protocols described herein. In addition, other methods for observing disturbances in nucleotide and amino acid sequences such as single-strand conformation polymorphism analysis are well known in the field (see, for example, US patent no. 5,382,510 published September 7, 1999 and 5,952,170 published January 17, 1995).

In addition, one can examine the methylation status of a PSCA gene in a biological sample. Aberrant demethylation and/or hypermethylation of CpG islands in gene 5' regulatory regions frequently occurs in immortalized and transformed cells and can result in altered expression of various genes. For example, promoter hypermethylation of the pi class of glutathione-S-transferase (a protein expressed in normal prostate but not expressed in >90% of prostate carcinomas) appears to permanently silence transcription of this gene and is the most frequently detected genomic alteration in prostate carcinomas ( De Marzo et al, Am. J. Pathol. 155(6): 1985-1992 (1999)). In addition, this change is present in at least 70% of cases of high-grade prostatic intraepithelial neoplasia (PIN)

(Brooks et al, Cancer Epidemiol. Biomarkers Prev., 1998, 7:531-536). In another example, expression of the LAGE-I tumor-specific gene (which is not expressed in normal prostate but is expressed in 25-50% of prostate cancer) is induced by deoxyazacytidine in lymphoblastoid cells, indicating that tumoral expression is due to demethylation (Lethe et al, Int. J. Cancer 76(6): 903-908

(1998)). A number of attempts to investigate the methylation status of a gene are well known in the art. For example, one can use, in Southern hybridization approaches, methylation-sensitive restriction enzymes that cannot cleave sequences containing methylated CpG sites to assess the methylation status of CpG islands. In addition, MSP (methylation-specific PCR) can rapidly profile the methylation status of all CpG sites present in a CpG island for a given gene. This procedure involves initial modification of DNA with sodium bisulfite (which will convert all unmethylated cytosines to uracil) followed by amplification using primers specific for methylated versus unmethylated DNA. Protocols involving methylation interference can also be found, for example, in Current Protocols in Molecular Biology, Unit 12, Frederick M. Ausubel et al eds., 1995.

Gene amplification is an additional method for determining the status of PSCA. Gene amplification is measured in a sample directly, for example by conventional Southern blotting or Northern blotting to quantify the transcription of mRNA (Thomas, 1980, Proe. Nati. Acad. Sei. USA, 77:5201 5205), dot blotting (DNA- assay) or in situ hybridization, using an appropriately labeled probe, based on the sequences provided herein. Alternatively, antibodies are used that recognize specific duplexes, including DNA duplexes, RNA duplexes and DNA-RNA hybrid duplexes or DNA-protein duplexes. The antibodies, on the other hand, are labeled and the experiment carried out where the duplex is bound to a surface, so that when the duplex is formed on the surface, the presence of antibody bound to the duplex can be detected.

Biopsy tissue or peripheral blood can be conveniently examined for the presence of cancer cells using, for example, Northern, dot blot or RT-PCR analysis to detect PSCA expression. The presence of RT-PCR amplifiable PSCA mRNA provides an indication of the presence of cancer. RT-PCR assays are well known in the art. RT-PCR detection assays for peripheral blood tumor cells are currently being evaluated for use in the diagnosis and treatment of several human solid tumors. In the prostate cancer field, these include RT-PCR assay detection of cells expressing PSA and PSM (Verkaik et al, 1997, Uroi. Res. 25:373-384; Ghossein et al, 1995, J. Clin. Oncol. 13:1195-2000 ; Heston et al., 1995, Clin. Chem. 41:1687-1688).

An assessment of an individual's susceptibility to developing cancer is described. In one embodiment, a method of predicting susceptibility to cancer comprises detecting PSCA mRNA or PSCA protein in a tissue sample, its presence indicative of susceptibility to cancer, wherein the degree of PSCA mRNA expression correlates with the degree of susceptibility. In a specific embodiment, the presence of PSCA in the prostate or other tissues is assayed, with the presence of PSCA in the sample providing an indication of prostate cancer susceptibility (or the appearance or existence of a prostate tumor). Similarly, one can evaluate the integrity of PSCA nucleotide and amino acid sequences in a biological sample to identify disturbances in the structure of these molecules such as insertions, deletions, substitutions and the like. The presence of one or more disturbances in PSCA gene products in the sample is indicative of cancer susceptibility (or the appearance or existence of a tumor).

It is possible to measure tumor aggressiveness. In one embodiment, a method of measuring aggressiveness of a tumor comprises determining the level of PSCA mRNA or PSCA protein expressed by tumor cells, comparing the level so determined to the level of PSCA mRNA or PSCA protein expressed in a corresponding normal tissue taken from the same individual or a normal tissue reference sample, where the degree of PSCA mRNA or PSCA protein expression in the tumor sample relative to the normal sample indicates the degree of aggressiveness. In a specific embodiment, aggressiveness of a tumor is evaluated by determining the extent to which PSCA is expressed in the tumor cells, with higher expression levels indicating more aggressive tumors. Another embodiment is the evaluation of the integrity of PSCA nucleotide and amino acid sequences in a biological sample, to identify disturbances in the structure of these molecules such as insertions, deletions, substitutions and the like. The presence of one or more disturbances indicates more aggressive tumors.

It is possible to observe the progression of a malignant disease in an individual over time. In one embodiment, methods of observing the progression of a malignant disease in an individual over time comprise determining the level of PSCA mRNA or PSCA protein expressed by cells in a sample of the tumor, comparing the level so determined to the level of PSCA mRNA or PSCA protein expressed in an equivalent tissue sample taken from the same individual at a different time, where the degree of PSCA mRNA or PSCA protein expression in the tumor sample over time provides information about cancer progression. In a specific embodiment, the cancer progression is evaluated by determining PSCA expression in the tumor cells over time, where increased expression over time indicates a progression of the cancer. One can also evaluate the integrity of PSCA nucleotide and amino acid sequences in a biological sample to identify disturbances in the structure of these molecules such as insertions, deletions, substitutions and the like, where the presence of one or more disturbances indicates a progression of the cancer.

The above diagnostic approaches can be combined with any one of a number of prognostic and diagnostic protocols known in the art. For example, another embodiment of the invention relates to methods for observing a coincidence between the expression of PSCA gene and PSCA gene products (or disturbances in PSCA gene and PSCA gene products) and a factor associated with malignant disease, as a means of diagnosing and prognosticating the status of a tissue sample. A number of factors associated with malignancy can be used, such as gene expression associated with malignancy (e.g. PSA, PSCA and PSM expression for prostate cancer etc.) as well as gross cytological observations (see e.g. Bocking et al. , 1984, Anal. Quant. Cytol. 6(2):74-88; Epstein, 1995, Hum. Pathol. 26(2):223-9; Thorson et al, 1998, Mod. Pathol. 11(6): 543-51; Baisden et al., 1999, Am. J. Surg. Pathol. 23(8):918-24). Methods for observing a coincidence between expression of PSCA gene and PSCA gene products (or perturbations in PSCA gene and PSCA gene products) and another factor associated with malignant disease are useful, for example, because the presence of a set of specific factors that coincide with disease that provide important information for diagnosing and prognosticating the status of a tissue sample.

In one embodiment, methods for observing a coincidence between the expression of PSCA gene and PSCA gene products (or perturbations in PSCA gene and PSCA gene products) and another factor associated with malignant disease involve detecting the overexpression of PSCA mRNA or protein in a tissue sample, detecting the overexpression of PSA mRNA or protein in a tissue sample (or PSCA or PSM expression) and observing a coincidence of PSCA mRNA or protein and PSA mRNA or protein overexpression (or PSCA or PSM expression) . In a specific embodiment, the expression of PSCA and PSA mRNA in prostate tissue is examined, where the coincidence of PSCA and PSA mRNA overexpression in the sample indicates the existence of prostate cancer, prostate cancer susceptibility, or the occurrence or status of a prostate tumor.

Methods for detecting and quantifying the expression of PSCA mRNA or protein are described herein and standard nucleic acid and protein detection and quantification technologies are well known in the art. Standard procedures for the detection and quantification of PSCA mRNA include in situ hybridization using labeled PSCA riboprobes, Northern blots and related techniques using PSCA polynucleotide probes, RT-PCR analysis using primers specific for PSCA and other types of amplification detection methods, such as for example branched DNA, SISBA, TMA and the like. In a specific embodiment, semiquantitative RT-PCR is used to detect and quantify PSCA mRNA expression. Any number of primers capable of amplifying PSCA can be used for this purpose, including but not limited to the various primer sets specifically described herein. In a specific embodiment, polyclonal or monoclonal antibodies specifically reactive with the wild-type PSCA protein can be used in an immunohistochemical assay of biopsy tissue.

IX.) Identification of molecules that interact with PSCA

The PSCA protein and nucleic acid sequences described herein allow one skilled in the art to identify proteins, small molecules, and other agents that interact with PSCA, as well as pathways activated by PSCA via any one of a variety of field-accepted protocols. For example, one can use one of the so-called interaction precipitation systems (also referred to as the "two-hybrid experiment"). In such systems, molecules interact and reconstitute a transcription factor that directs expression of a reporter gene, after which the expression of the reporter gene is examined. Other systems identify protein-protein interactions in vivo through reconstitution of a eukaryotic transcriptional activator, see e.g. US Patent Nos. 5,955,280 published September 21, 1999, 5,925,523 published July 20, 1999, 5,846,722 published December 8, 1998 and 6,004,746 published December 21, 1999. Algorithms are also available in the field of genome-based prediction of protein function (see eg Marcotte, et al., Nature 402: 4 November 1999, 83-86).

Alternatively, one can screen peptide libraries to identify molecules that interact with PSCA protein sequences. In such methods, peptides that bind to PSCA are identified from screening libraries that encode a random or controlled collection of amino acids. Peptides encoded by the libraries are expressed as fusion proteins to bacteriophage coat proteins, the bacteriophage particles are then screened against the PSCA protein(s).

Consequently, peptides that have a variety of applications, such as therapeutic, prognostic or diagnostic reagents, are thus identified without any prior information about the structure of the expected ligand or receptor molecule. Typical peptide libraries and

screening methods that can be used to identify molecules that interact with PSCA protein sequences are described, for example, in US Patent Nos. 5,723,286 published March 3, 1998 and 5,733,731 published March 31, 1998.

Alternatively, cell lines expressing PSCA are used to identify protein-protein interactions mediated by PSCA. Such interactions can be examined using immunoprecipitation techniques (see, e.g., Hamilton B.J., et al. Biochem. Biophys. Res. Commun. 1999, 261:646-51). PSCA protein can be immunoprecipitated from PSCA-expressing cell lines using anti-PSCA antibodies. Alternatively, antibodies against His-tag can be used in a cell line engineered to express fusions of PSCA and a His-tag (vectors mentioned above). The immunoprecipitated complex can be examined for protein association by procedures such as Western blotting, <35>S-methionine labeling of proteins, protein microsequencing, silver labeling, and two-dimensional gel electrophoresis.

Small molecules and ligands that interact with PSCA can be identified through related embodiments of such screening experiments. For example, small molecules can be identified that interfere with protein function, including molecules that interfere with PSCA's ability to mediate phosphorylation and dephosphorylation, interaction with DNA or RNA molecules as an indication of cell cycle regulation, "second messenger" signaling or tumorigenesis. Similarly, small molecules that modulate PSCA-related ion channel, protein pump, or cell communication functions have been identified and used to treat patients who have cancers that express PSCA (see, e.g., Hille, B., Ionic Channels of Excitable Membranes 2nd Ed., Sinauer Assoc., Sunderland, MA, 1992). Furthermore, ligands that regulate PSCA function can be identified based on their ability to bind PSCA and activate a reporter construct. Typical methods are described, for example, in US patent no. 5,928,868 published 27 July 1999 and include methods for forming hybrid ligands where at least one ligand is a small molecule. In an illustrative embodiment, cells engineered to express a fusion protein of PSCA and a DNA binding protein are used to co-express a hybrid ligand/small molecule fusion protein and a cDNA library transcriptional activator protein. The cells further contain a reporter gene, the expression of which is conditioned in the proximity of the first and second fusion proteins to each other, an event that occurs only if the hybrid ligand binds to target sites on both hybrid proteins. The cells expressing the reporter gene are selected and the unknown small molecule or ligand is identified. This method provides a means for the identification of modulators that activate or inhibit PSCA.

It is possible to screen for a molecule that interacts with a PSCA amino acid sequence shown in Figure 1, comprising the steps of contacting a population of molecules with a PSCA amino acid sequence, allowing the population of molecules and the PSCA amino acid sequence to interact under conditions that facilitates an interaction, determining the presence of a molecule that interacts with the PSCA amino acid sequence and then separating molecules that do not interact with the PSCA amino acid sequence from molecules that do. In a specific embodiment, the method further comprises purification, characterization and identification of a molecule that interacts with the PSCA amino acid sequence. The identified molecule can be used to modulate a function performed by PSCA. In a preferred embodiment, the PSCA amino acid sequence is contacted with a library of peptides.

X.) Therapeutic methods and compositions

Identification of PSCA as a protein that is normally expressed in a limited set of tissues, but is also expressed in cancers such as those listed in Table I, opens several therapeutic avenues for treating such cancers.

Of note, targeted antitumor therapies are applicable even when the targeted protein is expressed in normal tissue, even in vital normal organ tissue. A vital organ is one that is necessary to sustain life, such as the heart or colon. A non-vital organ is one that can be removed after which the individual is still able to survive. Examples of non-vital organs are the ovary, breast and prostate.

For example, Herceptin® is an FDA-approved pharmaceutical that consists of an antibody that is immunoreactive with the protein variously known as HER2, HER2/neu, and erb-b-2. It is marketed by Genentech and is a commercially successful antitumor agent. Herceptin® sales reached nearly $400 million in 2002. Herceptin® is a treatment for HER2-positive metastatic breast cancer. However, the expression of HER2 is not limited to such tumors. The same protein is expressed in several normal tissues. In particular, it is known that HER2/neu is present in normal kidney and heart, thus these tissues are present in all human recipients of Herceptin. The presence of HER2/neu in normal kidney has also been confirmed by Latif, Z., etal, B.J.U. International (2002) 89:5-9. As shown in this article (which evaluated whether renal cell carcinoma should be a preferred indication for anti-HER2 antibodies such as Herceptin) both protein and mRNA are produced in benign kidney tissue. In particular, HER2/neu protein was highly overexpressed in benign kidney tissue.

Despite the fact that HER2/neu is expressed in such vital tissues as the heart and kidney, Herceptin is a very useful, FDA-approved and commercially successful drug. The effect of Herceptin on heart tissue, i.e. "cardiotoxicity," has only been a side effect of treatment. When patients were treated with Herceptin alone, significant cardiotoxicity occurred in a very low percentage of patients. To minimize karyotoxicity, there is a more stringent entry requirement for the treatment with HER2/neu. Factors such as predisposition to heart conditions are evaluated before treatment can occur.

Note in particular that although kidney tissue is indicated to show normal expression, possibly even higher expression than heart tissue, kidney has no noticeable Herceptin side effect whatsoever. Furthermore, from the different array from normal tissue where HER2 is expressed, there is very little incidence of any side effect. Only heart tissue has manifested any noticeable side effect at all. A tissue such as kidney, where HER2/neu expression is particularly notable, is not the basis for any side effect.

Furthermore, beneficial therapeutic effects have been found for antitumor therapies that target the epidermal growth factor receptor (EGFR); Erbitux (ImClone). EGFR is also expressed in a variety of normal tissues. There are very limited side effects in normal tissues after the use of anti-EGFR therapeutic agents. A common side effect that occurs with the EGFR treatment is a severe skin rash observed in 100% of patients undergoing treatment.

Thus, expression of a target protein in normal tissue, even vital normal tissue, does not overcome the utility of targeting the protein as a therapeutic agent for certain tumors where the protein is also overexpressed. For example, expression in vital organs is not in and of itself harmful. In addition, organs considered dispensable, such as the prostate and ovary, can be removed without affecting mortality. Finally, some vital organs are not affected by normal organ expression due to an immunoprivilege. Immunoprivileged organs are organs that are protected from blood by a blood-organ barrier and are thus not available for immunotherapy. Examples of immunoprivileged organs are the brain and the testicle.

Accordingly, therapeutic approaches that inhibit the activity of a PSCA protein are useful for patients suffering from cancers that express PSCA. These therapeutic approaches generally fall into three classes. The first class modulates PSCA function as it relates to tumor cell growth leading to the inhibition or retardation of tumor cell growth or inducing its killing. The second class comprises various methods for inhibiting the binding or association of a PSCA protein with its binding partner or with other proteins. The third class comprises a variety of methods for inhibiting the transcription of a PSCA gene or translation of PSCA mRNA.

X.A.) Anticancer vaccines

Cancer vaccines comprising a PSCA-related protein or PSCA-related nucleic acid can be created. Considering expression of PSCA, cancer vaccines prevent and/or treat PSCA-expressing cancers with minimal or no effects on non-target tissues. The use of a tumor antigen in a vaccine that generates cell-mediated humoral immune responses as anticancer therapy is well known in the art and has been used in prostate cancer using human PSMA and rodent PAP immunogens (Hodge et al, 1995, Int. J. Cancer 63:231- 237; Fong et al., 1997, J. Immunol. 159:3113-3117).

Such methods can be readily practiced using a PSCA-related protein or a PSCA-encoding nucleic acid molecule and recombinant vectors capable of expressing and presenting the PSCA immunogen (typically comprising multiple T-cell epitopes or antibody). Those skilled in the art will appreciate that a variety of vaccine systems for delivery of immunoreactive epitopes are known in the art (see, e.g., Heryln et al, Ann Med 1999 Feb 31(l):66-78; Maruyama et al, Cancer Immunol Immunother 2000 Jun 49(3 ): 123-32) Briefly, such methods for generating an immune response (e.g. cell-mediated and/or humoral) in a mammal include the steps: exposure of the mammal's immune system to an immunoreactive epitope (e.g. an epitope present in a PSCA protein shown in Figure 1 or analog or homolog thereof) such that the mammal generates an immune response specific to that epitope (eg, generates antibodies that specifically recognize that epitope).

The entire PSCA protein, immunogenic regions or epitopes thereof, can be combined and delivered by various means. Such vaccine compositions may include, for example, lipopeptides (eg, Vitiello, A. et al., J. Clin. Invest. 95:341, 1995), peptide compositions encapsulated in poly(DL-lactide-coglycolide) ("PLG") microspheres (see, e.g., Eldridge, etal, Molec. Immunol. 28:287-294, 1991: Alonso etal, Vaccine 12:299-306, 1994; Jones etal, Vaccine 13:675-681, 1995), peptide compositions contained in immunostimulatory complexes (ISCOMS) (see, e.g., Takahashi et al, Nature 344:873-875, 1990; Hu et al, Clin Exp Immunol. 113:235-243, 1998), multiple antigen peptide systems (MAPs) (see, e.g. eg Tam, J.P., Proe. Nati. Acad. Sei. U.S.A. 85:5409-5413, 1988; Tam, J.P., J. Immunol. Methods 196:17-32, 1996), peptides formulated as multivalent peptides; peptides for use in ballistic delivery systems, typically crystallized peptides, viral delivery vectors (Perkus, M. E. etal, I: Concepts in vaccine development, Kaufmann, S. H. E., ed., p. 379, 1996; Chakrabarti, S. etal, Nature 320:535, 1986; Hu, S. L. et al., Nature 320:537, 1986; Kieny, M.-P. et al., AIDS Bio/Technology 4:790, 1986; Top, F. H. et al., J. Infect. Dis. 124:148, 1971; Chanda, P. K. etal, Virology 175:535, 1990), particles of viral or synthetic origin (eg, Kofler, N. etal, J. Immunol. Methods. 192:25, 1996; Eldridge, J. H. etal, Sem. Hematol . 30:16, 1993; Falo, L. D., Jr. et al., Nature Med. 7:649, 1995), adjuvantia (Warren, H. S., Vogel, F. R. and Chedid, L. A. Annu. Rev. Immunol. 4:369, 1986; Gupta, R. K. etal, Vaccine 11:293, 1993), liposomes (Reddy, R. etal, J. Immunol. 148:1585, 1992; Rock, K. L., Immunol. Today 17:131, 1996), or naked or particle-absorbed cDNA (Ulmer, J.B. et al, Science 259:1745, 1993; Robinson, H.L., Hunt, L.A. and Webs ter, R.G., Vaccine 11:957,1993; Shiver, J.W. et al., In: Concepts in vaccine development, Kaufmann, S.H.E., ed., p. 423, 1996; Cease, K.B. and Berzofsky, J.A., Annu. Fox. Immunol. 12:923, 1994 and Eldridge, J.H. et al., Sem. Hematol. 30:16,1993). Toxin-targeted delivery technologies also known as receptor-mediated targeting, such as those from Avant Immunotherapeutics, Inc. (Needham, Massachusetts) may also be used.

In patients with PSCA-associated cancer, the vaccine compositions according to the invention can also be used together with other treatments used for cancer, e.g. surgery, chemotherapy, drug therapies, radiation therapies, etc. extensive use in combination with immune adjuvants such as IL-2, IL-12, GM-CSF and the like.

Cellular vaccines:

CTL epitopes can be determined using specific algorithms to identify peptides within PSCA protein that bind corresponding HLA alleles (see, e.g., Table IV; Ep imer™ and Epimatrix™, Brown University (URL brown.edu/Research/ TB-fflV_Lab/epimatrix/epimatrix.html); and BIMAS (URL bimas.dcrt.nih.gov/; SYFPEITHI at URL syfpeithi.bmi-heidelberg.com/). In a preferred embodiment, a PSCA immunogen contains one or more amino acid sequences identified using techniques well known in the art, such as the sequences shown in Tables V-XVEH and XXTI-LI or a peptide of 8, 9, 10 or 11 amino acids specified by an HLA class I motif/supermotif (e.g. table IV (A), table IV (D) or table IV (E)) and/or a peptide of at least 9 amino acids comprising an HLA class LT motif/supermotif (e.g. table IV (B) or table IV (C)).As is recognized in the art, the HLA class I binding cleft is essentially closed terminated so that peptides of only a particular larger reading region can fit into the pit and be bound, generally HLA class I epitopes are 8, 9, 10 or 11 amino acids long. In contrast, the HLA class LT binding cleft is essentially open ended; therefore, a peptide of approx. 9 or more amino acids be bound by an HLA class II molecule. Because of the binding pit differences between HLA class I and II, HLA class I motifs are length specific, i.e., position two of a class I motif is the second amino acid in an amino to carboxyl direction of the peptide. Amino acid positions in a class II motif are relative only to each other, not the total peptide, i.e. additional amino acids may be bound to the amino and/or carboxyl terminus of a motif-bearing sequence. HLA class II epitopes are often 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24 or 25 amino acids long or longer than 25 amino acids.

A number of methods for generating an immune response in a mammal are known in the art (for example, as the first step in the generation of hybridomas). Methods of generating an immune response in a mammal comprise exposing the mammalian immune system to an immunogenic epitope on a protein (eg, a PSCA protein) such that an immune response is generated. A typical embodiment comprises a method of generating an immune response to PSCA in a host by contacting the host with a sufficient amount of at least one PSCA B cell or cytotoxic T cell epitope or analog thereof; and at at least one periodic interval thereafter again contacting the host with the PSCA-B cell or cytotoxic T cell epitope or analog thereof. A specific embodiment comprises a method of generating an immune response against a PSCA-related protein or a man-made multiepitopic peptide comprising: administering a PSCA immunogen ( e.g. , a PSCA protein or a peptide fragment thereof, a PSCA fusion protein or analog etc.) in a vaccine preparation for a human or another mammal. Typically, such vaccine formulations further contain a suitable adjuvant (see, e.g., US Patent No. 6,146,635) or a universal helper epitope such as a PADRETM peptide (Epimmune Inc., San Diego, CA; see, e.g., Alexander et al. , J. Immunol. 2000 164(3); 164(3): 1625-1633; Alexander et al., Immunity 1994 1(9): 751-761 and Alexander et al., Immunol. Res. 1998 18(2): 79-92 ). An alternative method comprises generating an immune response in a subject against a PSCA immunogen by: administering in vivo to muscle or skin of the subject's body a DNA molecule comprising a DNA sequence encoding a PSCA immunogen, the DNA sequence operative bound to regulatory sequences that control the expression of the DNA sequence; where the DNA molecule is taken up by cells, the DNA sequence is expressed in the cells and an immune response is formed against the immunogen (see e.g. US patent no. 5,962,428). Optionally, a genetic vaccine facilitator such as anionic lipids; saponins; lectins; estrogenic compounds; hydroxylated lower alkyls; dimethyl sulfoxide; and urea also administered. Additionally, an anti-idiotypic antibody that mimics PSCA can be administered to generate a response to the target antigen.

Nucleic acid vaccines:

Vaccine compositions include nucleic acid-mediated modalities. DNA or RNA encoding protein(s) herein may be administered to a patient. Genetic

immunization procedures can be used to generate prophylactic or therapeutic humoral and cellular immune responses directed against PSCA-expressing cancer cells. Constructs comprising DNA encoding a PSCA-related protein/immunogen and appropriate regulatory sequences can be injected directly into the muscle or skin of an individual so that the cells of the muscle or skin take up the construct and express the encoded PSCA protein/immunogen. Alternatively, a vaccine comprises a PSCA-related protein. Expression of the PSCA-related protein immunogen results in the generation of prophylactic or therapeutic humoral and cellular immunity against cells bearing a PSCA protein. Various prophylactic and therapeutic genetic immunization techniques known in the field can be used (for overview see information and references published at the internet address genweb.com). Nucleic acid-based delivery is described, for example, in Wolff et. al., Science 247:1465 (1990) as well as US Patent No. 5,580,859; 5,589,466; 5,804,566; 5,739,118; 5,736,524; 5,679,647; WO 98/04720. Examples of DNA-based delivery technologies include "naked DNA", facilitated (bupivicaine, polymers, peptide-mediated) delivery, cationic lipid complexes, and particle-mediated ("gene gun") or pressure-mediated delivery (see, e.g., US Patent No. 5,922,687).

For therapeutic or prophylactic immunization purposes, proteins according to the invention can be expressed via viral or bacterial vectors. Various viral gene delivery systems that can be used in the practice of the invention include, but are not limited to, vaccinia, fowlpox, canarypox, adenovirus, influenza, poliovirus, adeno-associated virus, lentivirus and sindbisvirus (see e.g. Restifo, 1996, Curr. Opin. Immunol. 8:658-663; Tsang et al. J. Nat. Cancer Inst. 87:982-990

(1995)). Non-viral delivery systems can also be used by introducing naked DNA encoding a PSCA-related protein into the patient (eg, intramuscularly or intradermally) to elicit an antitumor response.

Vaccinia virus is used, for example, as a vector to express nucleotide sequences that code for the peptides according to the invention. When the recombinant vaccinia virus is introduced into a host, the protein immunogenic peptide is expressed and a host immune response is thereby elicited. Vaccinia vectors and methods useful in immunization protocols are described in e.g. US Patent No. 4,722,848. Another vector is BCG (Bacille Calmette Guerin). BCG vectors are described in Stover et al, Nature 351:456-460 (1991). A number of other vectors useful for therapeutic administration or immunization of the peptides according to the invention, e.g. adeno and adeno-associated virus vectors, retroviral vectors, Salmonella typhi vectors, detoxified anthrax toxin vectors, and the like, will be readily apparent to those skilled in the art from the description herein.

Thus, gene delivery systems are used to deliver a PSCA-related nucleic acid molecule. In one embodiment, the full-length human PSCA cDNA is used. In another embodiment, PSCA nucleic acid molecules encoding specific cytotoxic T lymphocyte (CTL) and/or antibody epitopes are used.

Ex vivo vaccines

Different ex vzvo strategies can also be used to generate an immune response. One approach involves the use of antigen presenting cells (APCs) such as dendritic cells (DC) to present PSCA antigen to a patient's immune system. Dendritic cells express MHC class I and LT molecules, B7 co-stimulator and LL-12 and are thus highly specialized antigen-presenting cells. In prostate cancer, autologous dendritic cells are pulsed with peptides from prostate-specific membrane antigen (PSMA) which are used in a phase I clinical trial to stimulate prostate cancer patients' immune systems (Tjoa et al, 1996, Prostate 28:65-69; Murphy et al, 1996, Prostate 29:371 -380). Thus, dendritic cells can be used to present PSCA peptides to T cells in the context of MHC class I or U molecules. In one embodiment, autologous dendritic cells pulsed with PSCA peptides are capable of binding to MHC class I and/or class II molecules. In another embodiment, dendritic cells are pulsed with the full-length PSCA protein. Yet another embodiment involves engineering the overexpression of a PSCA gene in dendritic cells using various delivery vectors known in the art, such as adenovirus (Arthur et al, 1997, Cancer Gen Ther. 4:17-25), retrovirus (Henderson et al, 1996, Cancer Res. 56:3763-3770), lentivirus, adeno-associated virus, DNA transfection (Ribas et al, 1997, Cancer Res. 57:2865-2869) or tumor-derived RNA transfection (Ashley et al, 1997, J. Exp. Med. 186:1177-1182). Cells expressing PSCA can also be engineered to express immune modulators, such as GM-CSF, and used as immunizing agents.

X.B.) PSCA as a target for antibody-based therapy

PSCA is an attractive target for antibody-based therapeutic strategies. Several antibody strategies are known in the art for targeting both extracellular and intracellular molecules (see e.g. complement- and ADCC-mediated killing as well as the use of intrabodies). Because PSCA is expressed by cancer cells of different lineages relative to corresponding normal cells, systemic administration of PSCA-immunoreactive compositions is produced that exhibits excellent sensitivity without toxic, non-specific and/or off-target effects caused by binding of the immunoreactive composition to non-target organs and loom. Antibodies specifically reactive with domains of PSCA are useful for treating PSCA-expressing cancers systemically, either as conjugates with a toxin or therapeutic agent or as naked antibodies capable of inhibiting cell proliferation or function.

PSCA antibodies can be introduced into a patient such that the antibody binds to the PSCA and modulates a function, such as an interaction with a binding partner, thereby mediating destruction of the tumor cells and/or inhibiting the growth of the tumor cells. Mechanisms by which such antibodies exert a therapeutic effect may include complement-mediated cytolysis, antibody-dependent cellular cytotoxicity, modulation of the physiological function of PSCA, inhibition of ligand-binding or signal transduction pathways, modulation of tumor cell differentiation, alteration of tumor angiogenesis factor profiles and/or apoptosis. Examples include Rituxan® for Non-Hodgkin's Lymphoma, Herceptin® for metastatic breast cancer and Erbitux® for colorectal cancer.

Those skilled in the art understand that antibodies can be used to specifically measure and bind immunogenic molecules such as an immunogenic region to a PSCA sequence shown in Figure 1.1. In addition, those skilled in the art understand that it is routine to conjugate antibodies to cytotoxic agents (see, e.g., Slever's et al Blood 93:11 3678-3684 (June 1, 1999)). When cytotoxic and/or therapeutic agents are delivered directly to cells, such as by conjugating them to antibodies specific for a molecule expressed by that cell (eg, PSCA), the cytotoxic agent will exert its known biological effect (ie, cytotoxicity ) on those cells.

A variety of compositions and methods for using antibody cytotoxic agent conjugates to kill cells are known in the art. In the context of cancer, typical methods involve administering to an animal having a tumor a biologically effective amount of a conjugate comprising a selected cytotoxic and/or therapeutic agent bound to a targeting agent (eg, an anti-PSCA antibody) that binds to a marker (eg PSCA) expressed, available for binding or located on cell surfaces. A typical embodiment is a method for delivering a cytotoxic and/or therapeutic agent to a cell expressing PSCA, comprising conjugating the cytotoxic agent to an antibody that immunospecifically binds to a PSCA epitope and exposing the cell to the antibody-agent conjugate . Another illustrative embodiment is a method of treating an individual suspected of suffering from metastatic cancer, comprising a step of parenterally administering to said individual a pharmaceutical composition comprising a therapeutically effective amount of an antibody conjugated to a cytotoxic and/or therapeutic agent.

Cancer immunotherapy using anti-PSCA antibodies can be performed according to various approaches that have been successfully used in the treatment of other types of cancer, including but not limited to colon cancer (Arien et al., 1998, Crit. Rev. Immunol. 18:133- 138), multiple myeloma (Ozaki et al, 1997, Blood 90:3179-3186, Tsunenari et al, 1997, Blood 90:2437-2444), gastric cancer (Kasprzyk et al, 1992, Cancer Res. 52:2771-2776), B -cell lymphoma (Funakoshi et al, 1996, J. Immunother. Emphasis Tumor Immunol. 19:93-101), leukemia (Zhong et al, 1996, Leuk. Res. 20:581-589), colorectal cancer (Moun et al, 1994, Cancer Res. 54:6160-6166; Veldersefa/., 1995, Cancer Res. 55:4398-4403) and breast cancer (Shepard et al, 1991, J. Clin. Immunol. 11:117-127). Some therapeutic approaches involve the conjugation of naked antibody to a toxin or radioisotope, such as the conjugation of Y<91> or I<131>, respectively, to anti-CD20 antibodies (eg, ZevalinTM, LDEC Pharmaceuticals Corp. or Bexxar™, Coulter Pharmaceuticals), while others involve the co-administration of antibodies and other therapeutic agents, such as HerceptinTM (trastuzuMAb) with paclitaxel (Genentech, Inc.). The antibodies may be conjugated to a therapeutic agent. To treat prostate cancer, for example, PSCA antibodies can be administered together with radiation, chemotherapy or hormone ablation. Antibodies may also be conjugated to a toxin such as calicheamicin (eg, Mylotarg™, Wyeth-Ayerst, Madison, NJ, a recombinant humanized IgG4kappa antibody conjugated to the antitumor antibiotic calicheamicin) or a maytansinoid (eg, taxane-based tumor-activated prodrug , TAP, platform, ImmunoGen, Cambridge, MA see also, e.g., US Patent 5,416,064) or Auristatin E (Nat Biotechnol. 2003 Jul; 21(7):778-84. (Seattle Genetics)).

Although PSCA antibody therapy is applicable to all stages of cancer, antibody therapy may be particularly appropriate in advanced or metastatic cancer. Treatment with the antibody therapy according to the invention is indicated for patients who have received one or more rounds of chemotherapy. Alternatively, antibody therapy according to the invention is combined with a chemotherapeutic agent or radiation regimen for patients who have not received chemotherapeutic treatment. In addition, antibody therapy may enable the use of reduced doses of concomitant chemotherapy, especially for patients who do not tolerate the toxicity of the chemotherapeutic agent very well. Fan et al (Cancer Res. 53:4637-4642, 1993), Prewettefa/. (International J. of Onco. 9:217-224, 1996) and Hancock et al. (Cancer Res. 51:4575-4580,1991) describes the use of various antibodies in conjunction with chemotherapeutic agents.

Although PSCA antibody therapy is applicable to all stages of cancer, antibody therapy may be particularly appropriate in advanced or metastatic cancer. Treatment with the antibody therapy according to the invention is indicated for patients who have received one or more rounds of chemotherapy. Alternatively, antibody therapy according to the invention is combined with a chemotherapeutic agent or radiation regimen for patients who have not received chemotherapeutic treatment. In addition, antibody therapy may enable the use of reduced doses of concomitant chemotherapy, especially for patients who do not tolerate the toxicity of the chemotherapeutic agent very well.

Cancer patients can be evaluated for the presence and level of PSCA expression, preferably using immunohistochemical assessments of tumor tissue, quantitative PSCA imaging, or other techniques that reliably indicate the presence and level of PSCA expression. Immunohistochemical analysis of tumor biopsies or surgical specimens is preferred for this purpose. Methods for immunohistochemical analysis of tumor tissue are well known in the field.

Anti-PSCA monoclonal antibodies that treat prostate and other cancers include those that initiate a strong immune response against the tumor or those that are directly cytotoxic. In this regard, anti-PSCA monoclonal antibodies (MAbs) can induce tumor cell lysis by either complement-mediated or antibody-dependent cellular cytotoxicity (ADCC) mechanisms, both of which require an intact Fc portion of the immunoglobulin molecule for interaction with effector cell Fc receptor sites on complementary proteins. In addition, anti-PSCA MAbs that exert a direct biological effect on tumor growth are useful for treating PSCA-expressing cancers. Mechanisms by which direct cytotoxic MAbs act include: inhibition of cell growth, modulation of cellular differentiation, modulation of tumor angiogenesis factor profiles and induction of apoptosis. The mechanism(s) by which a particular anti-PSCA-MAb exerts an antitumor effect is evaluated using any number of in v/Yro assays that evaluate cell death such as ADCC, ADMMC, complement-mediated cell lysis, and so on, which are generally known in the area.

In some patients, the use of murine or other non-human monoclonal antibodies or human/mouse chimeric MAbs may elicit moderate to strong immune responses against the non-human antibody. This can result in excretion of the antibody from circulation and reduced effectiveness. In the most severe cases, such an immune response can lead to the extensive formation of immune complexes that can potentially cause kidney failure. Accordingly, preferred monoclonal antibodies used in the therapeutic methods of the invention are those which are either fully human or humanized and which bind specifically to the target PSCA antigen with high affinity but show low or no antigenicity in the patient.

Therapeutic methods of the invention contemplate the administration of single anti-PSCA MAbs as well as combinations or cocktails of different MAbs. Such MAb cocktails may have certain advantages as they contain MAbs targeting different epitopes, exploit different effector mechanisms or combine directly cytotoxic MAbs with MAbs that rely on immune effector functionality. Such MAbs in combination may show synergistic therapeutic effects. In addition, anti-PSCA MAbs can be co-administered with other therapeutic modalities, including but not limited to various chemotherapeutic agents, androgen blockers, immune modulators (eg, IL-2, GM-CSF), surgery, or radiation. The anti-PSCA MAs are administered in their "naked" or unconjugated form or may have a therapeutic agent(s) conjugated to them.

Anti-PSCA antibody formulations are administered via any route capable of delivering the antibodies to a tumor cell. Methods of administration include, but are not limited to, intravenous, intraperitoneal, intramuscular, intratumor, intradermal, and the like. Treatment generally involves repeated administration of the anti-PSCA antibody preparation, via an acceptable route of administration such as intravenous (IV) injection, at a typical dose in the range of approx. 0,1, ,2, ,3, ,4, ,5, ,6, ,7, ,8, ,9, 1, 2, 3, 4, 5, 6, 7, 8, 9,10, 15, 20 or 25 mg/kg body weight. In general, doses are in the range of 10-1000 mg MAb per week effective and well tolerated.

Based on clinical experience with Herceptin™-MAb in the treatment of metastatic breast cancer, an initial loading dose of approximately 4 mg/kg patient body weight IV, followed by weekly doses of approximately 2 mg/kg IV of the anti-PSCA MAb preparation an acceptable dosing regimen. Preferably, the initial "loading dose" is administered as a 90-minute or longer infusion. The periodic maintenance dose is administered as a 30-minute or longer infusion, provided the initial dose was well tolerated. As recognized by those skilled in the art, various factors may influence the ideal dosage regimen in a particular case. Such factors include, for example, the binding affinity and half-life of Ab or MAbs used, the degree of PSCA expression in the patient, the extent of circulating field PSCA antigen, the desired equilibrium antibody concentration level, treatment frequency and the influence of chemotherapeutic or other agents used in combination with the treatment method according to the invention as well as the health status of a particular patient.

Optionally, patients should be evaluated for the levels of PSCA in a given sample (eg, the levels of circulating PSCA antigen and/or PSCA-expressing cells) to assist in determining the most effective dosing regimen, etc. Such evaluations have also been used for monitoring purposes throughout therapy and are applicable to measure therapeutic success in combination with the evaluation of other parameters (for example urine cytology and/or ImmunoCyt levels in bladder cancer therapy or, analogously, serum PSA levels in prostate cancer therapy).

Anti-idiotypic anti-PSCA antibodies can also be used in anticancer therapy as a vaccine to elicit an immune response to cells expressing a PSCA-related protein. In particular, the formation of antiidiotypic antibodies is well known in the art; this methodology can easily be adapted to generate anti-idiotypic anti-PSCA antibodies that mimic an epitope on a PSCA-related protein (see, for example, Wagner et al., 1997, Hybridoma 16: 33-40; Foon et al., 1995, J. Clin. Invest. 96:334-342; Herlyn et al., 1996, Cancer Immunol. Immunother. 43:65-76). Such an anti-idiotypic antibody can be used in cancer vaccine strategies.

An object of the present invention is to provide PSCA antibodies, which inhibit or retard the growth of tumor cells expressing PSCA. A further object of the present invention is to provide methods for inhibiting angiogenesis and other biological functions and thereby reducing tumor growth in mammals, preferably humans, by using such PSCA antibodies and in particular by using such PSCA antibodies combined with radiation and chemotherapy or both.

In one embodiment, there is synergy when tumors, including human tumors, are treated with PSCA antibodies together with chemotherapeutic agents or radiation or combinations thereof. In other words, the inhibition of tumor growth by a PSCA antibody is enhanced more than expected when combined with chemotherapeutic agents or radiation or combinations thereof. Synergy can be shown, for example, by greater inhibition of tumor growth by combined treatment than would be expected from a treatment with only PSCA antibodies or the additive effect of treatment with a PSCA antibody and a chemotherapeutic agent or radiation. Preferably, synergy is demonstrated by remission of the cancer where remission is not expected from treatment either from a naked PSCA antibody or with treatment using an additive combination of a PSCA antibody and a chemotherapeutic agent or radiation.

The method of tumor cell growth inhibition using a PSCA antibody and a combination of chemotherapy or radiation or both comprises administering the PSCA antibody before, during, or after the initiation of chemotherapy or radiation therapy, as well as any combination thereof (ie, before and during, before and after, during and after or before, during and after starting chemotherapy and/or radiation therapy). For example, PSCA antibody is typically administered between 1 and 60 days, preferably between 3 and 40 days, more preferably between 5 and 12 days, before starting radiation therapy and/or chemotherapy. However, depending on the treatment protocol and the specific patient's needs, the procedure is performed in a way that will provide the most effective treatment and ultimately prolong the life of the patient.

The administration of chemotherapeutic agents can be carried out in a number of ways including systemically by the parenteral and enteral routes. In one embodiment, the PSCA antibody and the chemotherapeutic agent are administered as separate molecules. In another embodiment, the PSCA antibody is linked, for example by conjugation, to a chemotherapeutic agent. (See the example entitled "Human clinical trials for the treatment and diagnosis of human carcinomas using human anti-PSCA antibodies in vzvo") and (See section entitled "PSCA as a target for antibody-based therapy"). Specific examples of chemotherapeutic agents or chemotherapy include cisplatin, dacarbazine (DUC), dactinomycin, mechlorethamine (nitrogen mustard), streptozocin, cyclophosphamide, carmustine (BCNU), lomustine (CCNU), doxorubicin (adriamycin), daunorubicin, procarbazine, mitomycin, cytarabine, etoposide, methotrexate, 5-fluorouracil, vinblastine, vincristine, bleomycin, paclitaxel (taxol), docetaxel (taxotere), aldesleukin, asparaginase, busulfan, carboplatin, cladribine, dacarbazine, floxuridine, fludarabine, hydroxyurea, ifosfamide, interferon-alpha, leuprolide, megestrol, melphalan, mercaptopurine, plicamycin, mitotane, pegaspargase, pentostatin, pipobroman, plicamycin, streptozocin, tamoxifen, teniposide, testolactone, thioguanine, thiotepa, uracil mustard, vinorelbine, chlorambucil, taxol and combinations thereof.

The source of radiation, used in combination with a PSCA antibody, can be either external or internal to the patient being treated. When the source is external to the patient, the therapy is known as external beam radiation therapy (EBRT). When the source of radiation is internal to the patient, the treatment is called brachytherapy (BT).

The radiation is administered according to well-known standard techniques using standard equipment manufactured for this purpose, such as the AECL Theratron and Varian Clinac. The radiation dose depends on a number of factors which are well known in the field. Such factors include the organ being treated, the healthy organs in the radiation path that may be inadvertently adversely affected, the tolerance of the patient to radiation therapy, and the area of the body in need of treatment. The dose will typically be between 1 and 100 Gy and more particularly between 2 and 80 Gy. Some doses that have been reported include 35 Gy to the spinal cord, 15 Gy to the kidney, 20 Gy to the liver, and 65-80 Gy to the prostate. However, it should be emphasized that the invention is not limited to any particular dose. The dose will be determined by the treating physician according to the particular factors in a given situation, including the factors mentioned above.

The distance between the source of the external radiation and the point of entry into the patient can be any distance that represents an acceptable balance between killing target cells and minimizing side effects. The source of the external radiation is typically between 70 and 100 cm from the point of entry into the patient.

Brachytherapy is generally performed by placing the source of radiation in the patient. Typically, the source of radiation is placed approximately 0-3 cm from the tissue being treated. Known techniques include interstitial, internal cavity ("intercavitary") and surface brachytherapy. The radioactive sources can be implanted permanently or temporarily. Some typical radioactive atoms that have been used in permanent implants include iodine-125 and radon. Some typical radioactive atoms that have been used in temporary implants include radium, cesium-137 and iridium-192. Some additional radioactive atoms that have been used in brachytherapy include americium-241 and gold-198. The radiation dose for brachytherapy can be the same as mentioned above for external radiation therapy. In addition to the factors mentioned above to determine the dose of external radiation therapy, the type of radioactive atom used is also taken into account in determining the brachytherapy dose.

X.C.) PSCA as a measure of cellular immune responses

Vaccines and methods for producing vaccines that contain an immunogenically effective amount of one or more HLA binding peptides as described here are further embodiments according to the invention. Furthermore, vaccines according to the invention comprise compositions of one or more of the required peptides. A peptide may be present in a vaccine individually. Alternatively, the peptide may exist as a homopolymer comprising multiple copies of the same peptide or as a heteropolymer of different peptides. Polymers have the advantage of increased immunological response and, where different peptide epitopes are used to prepare the polymer, the additional ability to elicit antibodies and/or CTLs that react with different antigenic determinants of the pathogenic organism or tumor-related peptide targeted for an immune response. The composition may be a naturally occurring region of an antigen or may be prepared, e.g. recombinantly or by chemical synthesis.

Carriers that can be used with vaccines according to the invention are well known in the field and include e.g. thyroglobulin, albumins such as human serum albumin, tetanus toxoid, polyamino acids such as poly-L-lysine, poly-L-glutamic acid, influenza, hepatitis B virus core protein and the like. The vaccines may contain a physiologically tolerable (ie acceptable) diluent such as water or saline, preferably phosphate-buffered saline. The vaccines also typically include an adjuvant. Adjuvants such as incomplete Freund's adjuvant, aluminum phosphate, aluminum hydroxide or alum are examples of materials well known in the art. In addition, as described herein, CTL responses can be primed by conjugation of peptides of the invention to lipids, such as tripalmitoyl-S-glycerylcysteine-lyseryl-serine (P3CSS). Furthermore, an adjuvant such as synthetic cytosine phosphorothiolated guanine-containing (CpG) oligonucleotides has been found to increase CTL responses 10- to 100-fold (see, e.g., Davila and Celis, J. Immunol. 165:539-547 (2000)).

Upon immunization with a peptide composition according to the invention, via injection, aerosol, oral, transdermal, transmucosal, intrapleural, intrathecal or other suitable routes, the immune system of the host responds to the vaccine by producing large amounts of CTLs and/or HTLs specific for the desired antigen . Accordingly, the host becomes at least partially immune to subsequent development of cells expressing or overexpressing PSCA antigen or derives at least some therapeutic benefit once the antigen has become tumor associated.

In some embodiments, it may be desirable to combine class I peptide components with components that elicit or facilitate neutralizing antibody and or helper T cell responses directed against the target antigen. A preferred embodiment of such a composition comprises class I and class U epitopes according to the invention. An alternative embodiment of such a composition comprises a class I and/or class U epitope according to the invention, together with a cross-reactive HTL epitope such as PADRE™ (Epimmune, San Diego, CA) molecule (described e.g. .in US patent number 5,736,142).

A vaccine according to the invention may also comprise antigen-presenting cells (APC), such as dendritic cells (DC), as a constituent for presenting peptides according to the invention. Vaccine compositions can be formed in vitro, following dendritic cell mobilization and harvesting, whereby charging of dendritic cells occurs in vitro. For example, dendritic cells are transfected, e.g. with a minigene according to the invention or are pulsed with peptides. The dendritic cell can then be administered to a patient to elicit immune responses in vivo. Vaccine compositions, either DNA- or peptide-based, can also be administered in vivo in combination with dendritic cell mobilization whereby charging of dendritic cells occurs in vivo.

Preferably, the following principles are applied when selecting an array of epitopes for inclusion in a polyepitope composition for use in a vaccine or for selecting distinct epitopes to be included in a vaccine and/or to be encoded by nucleic acids such as a minigene . It is preferred that each of the following principles be balanced to produce the selection. The multiple epitopes incorporated into a given vaccine composition may be, but need not be, consecutive in sequence in the natural antigen from which the epitopes are derived. 1. ) Epitopes are selected that, upon administration, mimic immune responses observed to be consistent with tumor shedding. For HLA class I, this includes 3-4 epitopes that come from at least one tumor-associated antigen (TAA). For HLA class II, a similar rationale is applied; again, 3-4 epitopes are selected from at least one TAA (see, e.g., Rosenberg et al., Science 278:1447-1450). Epitopes from one TAA can be used in combination with epitopes from one or more additional TAAs to produce a vaccine that targets tumors with varying expression patterns of commonly expressed TAAs. 2. ) Epitopes are selected that have the desired binding affinity established to be consistent with immunogenicity: for HLA class I an IC50 of 500 nM or less, often 200 nM or less; and for class II an IC50 of 1000 nM or less. 3. ) Sufficient supermotif-bearing peptides or a sufficient array of allele-specific motif-bearing peptides are selected, providing broad population coverage. For example, it is preferred to have at least 80% population coverage. A Monte Carlo analysis, a statistical evaluation known in the art, can be used to judge the breadth or excess of population coverage. 4. ) When selecting epitopes from cancer-related antigens, it is often useful to choose analogues because the patient may have developed tolerance to the natural epitope. 5. ) Of particular relevance, epitopes are referred to as "nested epitopes". Packed epitopes occur where at least two epitopes overlap in a given peptide sequence. A packaged peptide sequence may comprise B cell, HLA class I and/or HLA class LI epitopes. When providing packaged epitopes, a general objective is to provide the largest number of epitopes per sequence. Thus, one aspect is to avoid providing a peptide that is any longer than the amino terminus of the amino-terminal epitope and the carboxyl terminus of the carboxyl-terminal epitope in the peptide. When providing a multi-epitopic sequence, such as a sequence comprising packaged epitopes, it is generally important to screen the sequence to ensure that it does not have pathological or other deleterious biological properties. 6. ) If a polyepitopic protein is formed, or when generating a minigene, an aim is to form the smallest peptide that includes the interesting epitopes. This principle is similar if not the same as that used when selecting a peptide comprising packaged epitopes. However, with an artificial polyepitopic peptide, size minimization purposes are balanced against the need to integrate any spacer sequences between epitopes in the polyepitopic protein. Spacer amino acid residues can, for example, be introduced to avoid joining epitopes (an epitope recognized by the immune system, not present in the target antigen and only formed by the man-made juxtaposition of epitopes) or to facilitate cleavage between epitopes and thereby enhance epitope presentation. Joining epitopes are generally avoided because the recipient may generate an immune response to the non-natural epitope. Of particular importance is a junctional epitope that is a "dominant epitope." A dominant epitope can lead to such a zealous response that immune responses to other epitopes are reduced or suppressed. 7. ) Where the sequences of multiple variants of the same target protein are present, potential peptide epitopes can also be selected on the basis of their conservation. For example, a criterion for conservation may be defined as the entire sequence of an HLA class I binding peptide or the entire 9-mer core of a class U binding peptide being conserved in a designated percentage of the sequences evaluated for a specific protein antigen.

X. C. 1. Minigene vaccines

Several different approaches are available that allow simultaneous delivery of multiple epitopes. Nucleic acids that code for the peptides according to the invention are a particularly applicable embodiment according to the invention. Epitopes for inclusion in a minigene are preferably selected according to the guidelines set forth in the previous section. A preferred means for administering nucleic acids which encode the peptides according to the invention uses minigene constructs which encode a peptide comprising one or multiple epitopes according to the invention.

Use of multi-epitope minigenes is described below and in Ishioka et al, J. Immunol. 162:3915-3925, 1999; An, L. and Whitton, J.L., J. Virol. 71:2292, 1997; Thomson, S.A. et al., J. Immunol. 157:822, 1996; Whitton, J.L. et al., J. Virol. 67:348, 1993; Hanke, R. etal, Vaccine 16:426, 1998. For example, a multi-epitope DNA plasmid encoding supermotif and/or motif-bearing epitopes may derive PSCA, the PADRE™ universal helper T-cell epitope, or multiple HTL -epitopes from PSCA (see, e.g., Tables V-XVLTI and XXLT to LI) and an endoplasmic reticulum translocation signal sequence are constructed. A vaccine may also include epitopes derived from other TAAs.

The immunogenicity of a multi-epitopic minigene can be confirmed in transgenic mice to assess the magnitude of CTL induction responses against tested epitopes. Furthermore, the immunogenicity of DNA-encoded epitopes in vivo may be consistent with the in vivo responses of specific CTL lines against target cells transfected with the DNA plasmid. Thus, these experiments can show that the minigene serves to both: 1.) generate a CTL response and 2.) as the induced CTLs recognize cells expressing the encoded epitopes.

For example, to create a DNA sequence that codes for the selected epitopes (minigene) for expression in human cells, the amino acid sequences of epitopes can be reverse translated. A human codon usage table can be used to guide the codon choice for each amino acid. These epitope-encoding DNA sequences may be directly contiguous, so that when translated, a continuous polypeptide sequence is formed. To optimize expression and/or immunogenicity, additional elements can be introduced into the minigene design. Examples of amino acid sequences that can be reverse translated and comprised in the minigene sequence include: HLA class I epitopes, HLA class II epitopes, antibody epitopes, a ubiquitination signal sequence and/or an endoplasmic reticulum targeting signal. In addition, HLA presentation of CTL and HTL epitopes can be enhanced by including synthetic (eg, poly-alanine) or naturally occurring flanking sequences adjacent to the CTL or HTL epitopes; these larger peptides comprising the epitope(s) are within the scope of the invention.

The minigene sequence can be converted into DNA by assembling oligonucleotides that code for the plus and minus strands of the minigene. Overlapping oligonucleotides (30-100 bases long) can be synthesized, phosphorylated, purified and hybridized under appropriate conditions using well known techniques. The ends of the oligonucleotides can be ligated, for example, by using T4 DNA ligase. This synthetic minigene, encoding the epitope polypeptide, can then be cloned into a desired expression vector.

Standard regulatory sequences well known to those skilled in the art are preferably included in the vector to ensure expression in the target cells. Many vector elements are desirable: a promoter with a downstream cloning site for minigene insertion; a polyadenylation signal for efficient transcription termination; a£. co//' origin of replication; and eriE. co//'-selectable marker (eg ampicillin or kanamycin resistance). A number of promoters can be used for this purpose, e.g. the human cytomegalovirus (hCMV) promoter. See e.g. US Patent Nos. 5,580,859 and 5,589,466 for other suitable promoter sequences.

Additional vector modifications may be desired to optimize minigene expression and immunogenicity. In some cases, introns are necessary for efficient gene expression and one or more synthetic or naturally occurring introns could be introduced into the transcribed region of the minigene. The inclusion of mRNA stabilization sequences and sequences for replication in mammalian cells can also be considered for increasing minigene expression.

Once an expression vector is selected, the minigene is cloned into the polylinker region downstream of the promoter. This plasmid is transformed into an appropriate E. coli strain and DNA is prepared using standard techniques. The orientation and DNA sequence of the minigene, as well as all other elements comprised in the vector, have been confirmed using restriction mapping and DNA sequence analysis. Bacterial cells carrying the correct plasmid can be stored as a master cell bank and a worker cell bank.

In addition, immunostimulatory sequences (IS Ser or CpGs) appear to play a role in the immunogenicity of DNA vaccines. These sequences can be included in the vector, outside of the minigene coding sequence, if desired to improve immunogenicity.

In some embodiments, a bicistronic ("bicistronic") expression vector that allows production of both the minigene-encoded epitopes and another protein (included to enhance or reduce immunogenicity) may be used. Examples of proteins or polypeptides that could advantageously enhance the immune response if co-expressed include cytokines (e.g. IL-2, IL-12, GM-CSF), cytokine-inducing molecules (e.g. LelF), costimulatory molecules or for HTL- responses, pan-DR binding proteins (PADRE™, Epiimmune, San Diego, CA). Helper (HTL) epitopes may be bound to intracellular targeting signals and expressed separately from expressed CTL epitopes; this allows targeting of the HTL epitopes to a cell compartment different from that of the CTL epitopes. If necessary, this could facilitate more efficient entry of HTL epitopes into the HLA class II reaction pathway, thereby enhancing HTL induction. In contrast to HTL or CTL induction, specific attenuation of the immune response by co-expression of immunosuppressive molecules (eg TGF-β) may be beneficial in certain diseases.

Therapeutic amounts of plasmid DNA can be produced, for example, by fermentation in E. coli, followed by purification. Aliquots from the working cell bank are used to inoculate growth medium and grown to saturation in shake flasks or a bioreactor according to well-known techniques. Plasmid DNA can be purified using standard bioseparation technologies such as solid phase anion exchange resins supplied by QIAGEN, Inc. (Valencia, California). If necessary, supercoiled DNA can be isolated from the open circular and linear forms using gel electrophoresis or other methods.

Purified plasmid DNA can be prepared for injection using a variety of formulations. The simplest of these is the reconstitution of freeze-dried DNA in sterile phosphate-buffered saline (PBS). This approach, known as "naked DNA," has currently been used for intramuscular (LM) administration in clinical trials. In order to maximize the immunotherapeutic effects of the minigene DNA vaccines, an alternative method of formulating purified plasmid DNA may be desirable. A number of methods are described and new techniques may become available. Cationic lipids, glycolipids and fusogenic liposomes can also be used in the formulation (see e.g. as described by WO 93/24640; Mannino & Gould-Fogerite, BioTechniques 6(7): 682

(1988); US Pat No. 5,279,833; WO 91/06309; and Felgner, et al., Proe. Nat'l Acad. Pollock. USA 84:7413 (1987). In addition, peptides and compounds referred to collectively as protective, interactive, non-condensing compounds (PINCs) could also be complexed to purified plasmid DNA to affect variables such as stability, intramuscular dispersion, or trafficking to specific organs or cell types.

Target cell sensitization can be used as a functional assay for expression and HLA class I presentation of minigene-encoded CTL epitopes. For example, the plasmid DNA is introduced into a mammalian cell line suitable as a target for standard CTL chromium release assays. The transfection method used will depend on the final formulation. Electroporation can be used for "naked" DNA, while cationic lipids allow direct in vitro transfection. A plasmid expressing green fluorescent protein (GFP) can be cotransfected to allow enrichment of transfected cells using fluorescence-activated cell sorting (FACS). These cells are then chromium-51 (<51>Cr) labeled and used as target cells for epitope-specific CTL lines; cytolysis, detected by<51>Cr release, indicates both production and HLA presentation of minigene-encoded CTL epitopes. Expression of HTL epitopes can be evaluated analogously using assays to assess HTL activity.

In v/vo immunogenicity is another approach for functional testing of minigene DNA formulations. Transgenic mice expressing appropriate human HLA proteins are immunized with the DNA product. The dose and route of administration are formulation dependent (e.g. IM DNA in PBS, intraperitoneal (i.p.) for lipid-complexed DNA). Twenty-one days after immunization, spleen cells are harvested and restimulated for one week in the presence of peptides encoding each epitope tested. Next, for CTL effector cells, assays are performed for cytolysis of peptide-loaded,<51>Cr-labeled target cells using standard techniques. Lysis of target cells sensitized by HLA-loaded peptide epitopes corresponding to minigene-encoded epitopes demonstrates DNA vaccine function for in vivo induction of CTLs. Immunogenicity of HTL epitopes has been confirmed in transgenic mice in an analogous manner.

Alternatively, the nucleic acids can be administered using ballistic delivery as described, for example, in US Patent No. 5,204,253. When using this technique, particles comprised solely of DNA are administered. In a further alternative embodiment, the DNA may be adhered to particles such as gold particles.

Minigenes can also be delivered using other bacterial or viral delivery systems well known in the art, e.g. an expression construct which codes for epitopes according to the invention can be introduced into a viral vector such as vaccinia.

X. C. 2. Combinations of CTL peptides with helper peptides Vaccine compositions comprising CTL peptides according to the invention can be modified, e.g. analoges, to provide desired characteristics, such as improved serum half-life, extended population coverage or improved immunogenicity.

For example, the ability of a peptide to elicit CTL activity can be enhanced by linking the peptide to a sequence containing at least one epitope capable of eliciting a T helper cell response. Although a CTL peptide can be directly bound to a T helper peptide, CTL epitope/HTL epitope conjugates are often bound by a spacer molecule. The spacer is typically comprised of relatively small, neutral molecules, such as amino acids or amino acid mimetics, which are mainly uncharged under physiological conditions. The spacers are typically selected from e.g. Ala, Gly or other neutral spacers of non-polar amino acids or neutral polar amino acids. It will be understood that the possibly present spacer need not be comprised of the same residues and can thus be a hetero- or homo-oligomer. When present, the spacer will usually be at least one or two residues, more commonly three to six residues and sometimes 10 or more residues. The CTL peptide epitope can be linked to the T helper peptide epitope either directly or via a spacer either at the amino or carboxy terminus of the CTL peptide. The amino terminus of either the immunogenic peptide or the T-helper peptide can be acylated.

HTL peptide epitopes can also be modified to alter their biological properties. For example, they can be modified to include D-amino acids to increase their resistance to proteases and thus extend their serum half-life, or they can be conjugated to other molecules such as lipids, proteins, carbohydrates and the like to increase their biological activity. For example, a T-helper peptide may be conjugated to one or more palmitic acid chains at either the amino or carboxyl terminus.

X. C. 3. Combinations of CTL peptides with T-cell priming agents

In some embodiments, it may be desirable to include in the pharmaceutical compositions according to the invention at least one component that primes B lymphocytes or T lymphocytes. Lipids have been identified as agents capable of priming CTL in vivo. For example, palmitic acid residues can be attached to the 8- and α-amino groups of a lysine residue and then attached, e.g. via one or more linking residues, such as Gly, Gly-Gly-, Ser, Ser-Ser or the like, to an immunogenic peptide. The lipidated peptide can then be administered either directly in a micelle or particle, introduced into a liposome or emulsified in an adjuvant, e.g. incomplete Freund's adjuvant. In a preferred embodiment, a particularly effective immunogen composition comprises palmitic acid bound to 8- and α-amino groups of Lys, which are linked via binding, e.g. Ser-Ser, to the amino terminus of the immunogenic peptide.

As another example of lipid priming of CTL responses, E. co//' lipoproteins, such as tripalmitoyl-S-glycerylcysteine lyseryl-serine (P3CSS) can be used to prime virus-specific CTL when covalently bound to an appropriate peptide (see e.g. eg Deres, etal, Nature 342:561,1989). Peptides according to the invention can be linked to, for example, P3CSS, and the lipopeptide is administered to an individual to specifically prime an immune response to the target antigen. Furthermore, because induction of neutralizing antibodies can also be primed with P3CSS-conjugated epitopes, two such compositions can be combined to more effectively elicit both humoral and cell-mediated responses.

X.C.4. Vaccine compositions comprising DC pulsed with CTL and/or HTL peptides

An embodiment of a vaccine composition according to the invention comprises ex vivo administration of a cocktail of epitope-bearing peptides to PBMC, or isolated DC therefrom, from the patient's blood. A pharmaceutical agent to facilitate harvest of DC can be used such as Progenipoietin™ (Pharmacia-Monsanto, St. Louis, MO) or GM-CSF/IL-4. After pulsing the DC with peptides and before reinfusion into patients, the DC are washed to remove unbound peptides. In this embodiment, a vaccine comprises peptide-pulsed DCs that present the pulsed peptide epitopes complexed with HLA molecules on their surfaces.

The DC can be pulsed ex vivo with a cocktail of peptides, some of which stimulate CTL responses to PSCA. Optionally, a helper T cell (HTL) peptide, such as a natural or artificial loosely restricted HLA class II peptide, may be included to facilitate the CTL response. Thus, a vaccine according to the invention is used to treat cancer that expresses or overexpresses PSCA.

X.D.) Adoptive immunotherapy

Antigenic PSCA-related peptides are also used to elicit a CTL and/or HTL response ex vivo. The resulting CTL or HTL cells can be used to treat tumors in patients who do not respond to other conventional forms of therapy or will not respond to a therapeutic vaccine peptide or nucleic acid according to the invention. Ex vivo CTL or HTL responses to a particular antigen are induced by incubating in tissue culture the patient's, or genetically compatible, CTL or HTL progenitor cells together with a source of antigen-presenting cells (APC), such as dendritic cells, and the appropriate immunogen the peptide. After an appropriate incubation period (typically approx. 7-28 days), during which the precursor cells are activated and expanded into effector cells, the cells are infused back into the patient, where they will destroy (CTL) or facilitate destruction (HTL) of their specific target cell { e.g. e.g. a tumor cell). Transfected dendritic cells can also be used as antigen-presenting cells.

X.E.) Administration of vaccines for therapeutic or prophylactic purposes Pharmaceutical agents and vaccine compositions according to the invention are typically used to treat and/or prevent cancers that express or overexpress PSCA. In therapeutic applications, peptide and/or nucleic acid compositions are administered to a patient in an amount sufficient to elicit an effective B-cell, CTL, and/or HTL response to the antigen and to cure or at least partially arrest or abate symptoms and /or complications. An amount sufficient to accomplish this is defined as "therapeutically effective dose." Amounts effective for this application will depend on e.g. the particular composition administered, the method of administration, the stage and severity of the disease being treated, the weight and general health of the patient and the judgment of the prescribing physician.

For pharmaceutical compositions, the immunogenic peptides of the invention, or DNA encoding them, are generally administered to an individual already carrying a tumor expressing PSCA. The peptides or DNA encoding them can be administered individually or as fusions of one or more peptide sequences. Patients may be treated with the immunogenic peptides separately or in conjunction with other treatments such as surgery as appropriate.

For therapeutic use, administration should generally begin at the first diagnosis of PSCA-associated cancer. This is followed by boosting doses until at least symptoms are mainly subdued and for a period thereafter. The embodiment of the vaccine composition (ie, including but not limited to embodiments such as peptide cocktails, polyepitopic polypeptides, minigenes or TAA-specific CTLs or pulsed dendritic cells) delivered to the patient may vary according to the stage of the disease or the health status of the patient. For example, in a patient with a tumor expressing PSCA, a vaccine comprising PSCA-specific CTL may be more effective in killing tumor cells in the patient with advanced disease than alternative embodiments.

It is generally important to provide an amount of the peptide epitope delivered by a method of administration sufficient to effectively stimulate a cytotoxic T cell response; compositions that stimulate helper T cell responses may also be provided according to this embodiment of the invention.

The dose for an initial therapeutic immunization generally occurs in a unit dose range where the lower value is approx. 1, 5, 50, 500 or 1,000 ug and the higher value is approx. 10,000; 20,000; 30,000; or 50,000 ug. Dose values for a human typically vary from approx. 500 ug to approx. 50,000 ug per 70 kg. patient. Reinforcement doses of between approx. 1.0 ug to approx. 50,000 (ag peptide in accordance with a boost regimen over weeks to months can be administered depending on the patient's response and condition as determined by measuring the specific activity of CTL and HTL obtained from the patient's blood. Administration should be continued until at least clinical symptoms or laboratory tests indicate that the neoplasia is removed or reduced and for a period thereafter.The doses, methods of administration and dosage regimens are adjusted according to methods known in the art.

In certain embodiments, the peptides and compositions according to the present invention are used in serious disease states, i.e. life-threatening or potentially life-threatening situations. In such cases, as a result of the minimal amounts of extraneous substances and the relatively non-toxic nature of the peptides in preferred compositions according to the invention, it is possible and may be felt desirable by the treating physician to administer significant excesses of these peptide compositions in relation to to these stated dose amounts.

The vaccine compositions according to the invention can also be used exclusively as prophylactic agents. In general, the dose for an initial prophylactic immunization generally occurs in a unit dose range where the lower value is approx. 1, 5, 50, 500 or 1000 ug and the higher value is approx. 10,000; 20,000; 30,000; or 50,000 ug. Dose values for a human typically vary from approx. 500 ug to approx. 50,000 ug per 70 kg. patient. This is followed by booster doses of between approx. 1.0 ug to approx. 50,000 µg of peptide administered at defined intervals from approximately four weeks to six months after the initial administration of vaccine. The immunogenicity of the vaccine can be assessed by measuring the specific activity of CTL and HTL obtained from a sample of the patient's blood.

The pharmaceutical compositions for therapeutic treatment are intended for parenteral, topical, oral, nasal, intrathecal or local (eg, as a cream or topical ointment) administration. Preferably, the pharmaceutical compositions are administered parenterally. intravenously, subcutaneously, intradermally or intramuscularly. Thus, the invention provides compositions for parenteral administration comprising a solution of the immunogenic peptides dissolved or suspended in an acceptable carrier, preferably an aqueous carrier.

A variety of aqueous carriers can be used, e.g. water, buffered water, 0.8% saline solution, 0.3% glycine, hyaluronic acid and the like. These compositions can be sterilized by conventional, well-known sterilization techniques or can be sterile filtered. The resulting aqueous solutions may be packaged for use as is or lyophilized, the lyophilized preparation being combined with a sterile solution prior to administration.

The compositions may contain pharmaceutically acceptable auxiliary substances as necessary for approximate physiological conditions, such as pH regulation and buffering agents, tonicity regulating agents, wetting agents, preservatives and the like, for example sodium acetate, sodium lactate, sodium chloride, potassium chloride, calcium chloride, sorbitan monolaurate, triethanolamine oleate, etc.

The concentration of peptides according to the invention in the pharmaceutical formulations can vary greatly, i.e. from less than approx. 0.1%, usually at or at least approx. 2% to as much as 20% to 50% or more by weight and will be selected primarily by liquid volumes, viscosities, etc., according to the particular method of administration selected.

A human unit dosage form of a composition is typically comprised in a pharmaceutical composition comprising a human unit dose of an acceptable carrier, in one embodiment an aqueous carrier, and is administered in a volume/amount known to those skilled in the art to be used for administration of such compositions to humans (see, e.g., Remington's Pharmaceutical Sciences, 17th Edition, A. Gennaro, Editor, Mack Publishing Co., Easton, Pennsylvania, 1985). For example, a peptide dose for initial immunization can be from approx. 1 to approx. 50,000 ug, generally 100-5,000 ug, for a 70 kg patient. For example, for nucleic acids, an initial immunization can be performed using an expression vector in the form of naked nucleic acid administered IM (or SC or ID) in amounts of 0.5-5 mg at multiple sites. The nucleic acid (0.1 to 1000 µg) can also be administered using a gene gun. After an incubation period of 3-4 weeks, a booster dose is then administered. The booster may be recombinant fowlpox virus administered at a dose of 5-10<7> to 5x10<9>pfu.

For antibodies, a treatment generally involves repeated administration of the anti-PSCA antibody preparation, via an acceptable route of administration such as intravenous (IV) injection, typically at a dose in the range of approx. 0.1 to approx. 10 mg/kg body weight. In general, doses are in the range of 10-500 mg MAb per week effective and well tolerated. Furthermore, an initial "loading dose" of approximately 4 mg/kg patient body weight represents IV, followed by weekly doses of approx. 2 mg/kg IV of the anti-PSCA MAb formulation is an acceptable dosing regimen. As recognized by those skilled in the art, various factors may influence the ideal dose in a particular case. Such factors include, for example, the half-life of a compound, the binding affinity of an Ab, the immunogenicity of a substance, the degree of PSCA expression in the patient, the extent of circulating field PSCA antigen, the desired equilibrium concentration level, treatment frequency and the impact of chemotherapeutic agents or other agents used in combination with the treatment method according to the invention, as well as the health status of a particular patient. Non-limiting preferred human unit doses are, for example, 500ug -1 mg, 1 mg - 50 mg, 50 mg - 100 mg, 100 mg - 200 mg, 200 mg - 300 mg, 400 mg - 500 mg, 500 mg - 600 mg, 600 mg - 700 mg, 700 mg - 800 mg, 800 mg - 900 mg, 900 mg - 1 g or 1 mg - 700 mg. In certain embodiments, the dose is in the range of 2-5 mg/kg body weight, e.g. followed by weekly doses of 1-3 mg/kg; 0.5 mg, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 mg/kg body weight followed, e.g. for two, three or four weeks of weekly doses; 0.5-10 mg/kg body weight, e.g. followed for two, three or four weeks by weekly doses; 225, 250, 275, 300, 325, 350, 375, 400 mg m<2>of body surface weekly; 1-600 mg m<2>of body surface weekly; 225-400 mg m<2>of body surface weekly; these doses may be followed by weekly doses for 2, 3, 4, 5, 6, 7, 8, 9, 19, 11, 12 or more weeks.

In one embodiment, human unit dosage forms of polynucleotides comprise a suitable dosage range or effective amount that provides any therapeutic effect. As recognized by one skilled in the art, a therapeutic effect depends on several factors, including the sequence of the polynucleotide, molecular weight of the polynucleotide, and route of administration. Doses are generally chosen by the doctor or other professional healthcare personnel according to a number of parameters known in the field, such as severity of symptoms, patient history and the like. In general, for a polynucleotide of approx. 20 bases, a dose range can be selected from, for example, an independently selected lower limit such as approx. 0.1, 0.25, 0.5, 1, 2, 5, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 200, 300, 400 or 500 mg/kg up to a independently chosen upper limit, greater than the lower limit, of approx. 60, 80, 100, 200, 300, 400, 500, 750, 1000, 1500, 2000, 3000, 4000, 5000, 6000, 7000, 8000, 9000 or 10,000 mg/kg. For example, a dose can be about any of the following: 0.1 to 100 mg/kg, 0.1 to 50 mg/kg, 0.1 to 25 mg/kg, 0.1 to 10 mg/kg, 1 to 500 mg/kg, 100 to 400 mg/kg, 200 to 300 mg/kg, 1 to 100 mg/kg, 100 to 200 mg/kg, 300 to 400 mg/kg, 400 to 500 mg/kg, 500 to 1000 mg/kg, 500 to 5000 mg/kg or 500 to 10,000 mg/kg. In general, parenteral administration methods may require higher doses of polynucleotide compared to more direct application of the nucleotide to disease tissue, which renders polynucleotides of increasing length.

In one embodiment, human unit dosage forms of T cells comprise a suitable dose range or effective amount that provides any therapeutic effect. As recognized by one skilled in the art, a therapeutic effect depends on several factors. Doses are generally chosen by the doctor or other professional healthcare personnel according to a number of parameters known in the field, such as severity of symptoms, patient history and the like. A dose can be approx. IO<4>cells to approx. IO<6>cells, approx. IO<6>cells to approx. 10<8> cells, approx. 10<8> to approx. 10<11> cells or approx. 10<8> to approx. 5 x 10<10> cells. A dose can also be approx. IO<6>cells/m<2>to approx. 10<10>cells/m<2>or approx. IO<6>cells/m<2>to approx. 108 cells/m<2>.

The protein(s) according to the invention and/or nucleic acids encoding the protein(s) can also be administered via liposomes, which can also serve to: 1) measure the protein(s) to a particular tissue, such as lymphoid tissue; 2) to measure selectivity to disease cells; or 3) to increase the half-life of the peptide composition. Liposomes include emulsions, foams, micelles, insoluble monolayers, liquid crystals, phospholipid dispersions, lamellar layers and the like. In these preparations, the peptide that is delivered is incorporated as part of a liposome, alone or together with a molecule that binds to a receptor prevalent among lymphoid cells, such as monoclonal antibodies that bind to the CD45 antigen or with other therapeutic or immunogenic compositions. Thus, liposomes either filled or decorated with a desired peptide according to the invention can be directed to the seat of lymphoid cells, where the liposomes then deliver the peptide compositions. Liposomes for use according to the invention are formed from standard vesicle-forming lipids, which generally comprise neutral and negatively charged phospholipids and a sterol such as cholesterol. The selection of lipids is generally guided by consideration of e.g. liposome size, acid lability and stability of the liposomes in the bloodstream. A number of methods are available for the preparation of liposomes, as described in e.g. Szoka, et al., Ann. Fox. Biophys. Bio meadow. 9:467 (1980) and US Patent Nos. 4,235,871, 4,501,728, 4,837,028 and 5,019,369.

For targeted cells in the immune system, a ligand that is introduced into the liposome can include e.g. antibodies or fragments thereof specific for cell surface determinants of the desired immune system cells. A liposome suspension containing a peptide can be administered intravenously, locally, topically, etc. in a dose that varies according to, e.g. a. the method of administration, the peptide being delivered and the stage of the disease being treated.

For solid compositions, conventional non-toxic solid carriers can be used which include, for example, pharmaceutical grades of mannitol, lactose, starch, magnesium stearate, sodium saccharin, talc, cellulose, glucose, sucrose, magnesium carbonate and the like. For oral administration, a pharmaceutically acceptable non-toxic composition is formed by incorporating any of the normally used additives, such as those carriers previously listed and generally 10-95% of active ingredient, i.e. one or more peptides according to the invention and more preferably in a concentration of 25%-75%.

For aerosol administration, immunogenic peptides are preferably supplied in finely divided form together with a surfactant and propellant. Typical percentages of peptides are approx. 0.01%-20% by weight, preferably approx. 1%-10%. The surfactant must of course be non-toxic and preferably soluble in the propellant. Representative of such agents are the esters or partial esters of fatty acids containing from approx. 6 to 22 carbon atoms, such as caproic, octanoic, lauric, palmitic, stearic, linoleic, linolenic, olesteric ("olesteric") and oleic acids with an aliphatic polyhydric alcohol or its cyclic anhydride. Mixed esters such as mixed or natural glycerides can be used. The surfactant can amount to approx. 0.1%-20% by weight of the composition, preferably approx. 0.25-5%. The balance of the composition is usually propellant. A carrier can also be included if desired, as with e.g. lecithin for intranasal delivery.

XI.) Diagnostic and prognostic embodiments of PSCA.

As described herein, PSCA polynucleotides, polypeptides, reactive cytotoxic T cells (CTL), reactive helper T cells (HTL) and antipolypeptide antibodies are used in well-known diagnostic, prognostic and therapeutic trials that examine disorders associated with dysregulated cell growth such as cancer, particularly the cancer listed in Table I (see, e.g., both its specific pattern of tissue expression as well as its overexpression in certain cancers as described, for example, in the example entitled "Expression Analysis of PSCA in Normal Tissue and Patient Samples").

PSCA can be analogized to a prostate associated antigen PSA, the archetypal marker that has been used by medical practitioners for years to identify and monitor the presence of prostate cancer (see, eg, Merrill etal, J. Uroi. 163(2): 503-5120 (2000); Polascik et al, J. Uroi. Aug; 162(2):293-306 (1999) and Fortier et al, J. Nat. Cancer Inst. 91(19): 1635-1640(1999)) . A number of other diagnostic markers have also been used in similar contexts including p53 and K-ras (see e.g. Tulchinsky etal, Int J Mol Med 1999 Jul 4(1):99-102 and Minimoto etal, Cancer Detect Prev 2000;24 (1):1-12). Therefore, this description of PSCA polynucleotides and polypeptides (as well as PSCA polynucleotide probes and anti-PSCA antibodies used to identify the presence of these molecules) and their properties allow those skilled in the art to use these molecules by methods analogous to those used, for example in a number of diagnostic trials aimed at investigating disorders associated with cancer.

Typical embodiments of diagnostic methods using the PSCA polynucleotides, polypeptides, reactive T cells and antibodies are analogous to those methods from well-established diagnostic tests, such as using e.g. PS A polynucleotides, polypeptides, reactive T cells and antibodies. For example, like PSA polynucleotides are used as probes (for example in Northern analysis, see e.g. Sharief et al., Biochem. Mol. Biol. Int. 33(3): 567-74(1994)) and primers (for example in PCR analysis, see, e.g., Okegawa et al, J. Uroi. 163(4): 1189-1190 (2000)) to observe the presence and/or level of PSA mRNAs in methods for monitoring PSA- overexpression or the metastasis of prostate cancer, the PSCA polynucleotides described herein can be used similarly to detect PSCA overexpression or the metastasis of prostate and other cancers that express this gene. Alternatively, as PSA polypeptides are used to generate antibodies specific for PSA which can then be used to observe the presence and/or level of PSA proteins in methods for monitoring PSA protein overexpression (see, e.g., Stephan et al, Urology 55(4):560-3 (2000)) or the metastasis of prostate cells (see, e.g., Alanen et al., Pathol. Res. Pract. 192(3):233-7 (1996)), the PSCA polypeptides described herein can are used to generate antibodies for use in detecting PSCA overexpression or the metastasis of prostate cells and cells of other cancers that express this gene.

Specifically, because metastases involve the movement of cancer cells from an organ of origin (such as the lung or prostate gland, etc.) to a different area of the body (such as a lymph node), tests that examine a biological sample for the presence of cells expressing PSCA polynucleotides and and/or polypeptides are used to provide evidence of metastasis. For example, when a biological sample from tissue that does not normally contain PSCA-expressing cells (lymph node) is found to contain PSCA-expressing cells such as the PSCA expression seen in LAPC4 and LAPC9, xenografts isolated from lymph node and bone metastasis respectively, this finding is indicative for metastasis.

Alternatively, PSCA polynucleotides and/or polypeptides can be used to provide evidence of cancer, for example when cells in a biological sample that do not normally express PSCA or express PSCA at a different level are found to express PSCA or have an increased expression of PSCA ( see, e.g., PSCA expression in the cancers listed in Table I and in patient samples etc. shown in the accompanying figures). In such trials, those skilled in the art may further wish to form supplemental evidence of metastasis by testing the biological sample for the presence of another tissue-restricted marker (in addition to PSCA) such as PSA, PSCA, etc. (see, e.g., Alanen et al, Pathol . Res. Pract. 192(3): 233-237(1996)).

The use of immunohistochemistry to identify the presence of a PSCA polypeptide within a tissue section may indicate an altered state of certain cells within that tissue. It is well understood in the art that the ability of an antibody to localize to a polypeptide expressed in cancer cells is a means of diagnosing the presence of disease, disease stage, progression and/or tumor aggressiveness. Such an antibody can also detect an altered distribution of the polypeptide within the cancer cells compared to corresponding non-malignant tissue.

The PSCA polypeptide and immunogenic compositions are also useful in considering the phenomena of altered subcellular protein localization in disease states. Change of cells from normal to disease state causes changes in cellular morphology and is often associated with changes in subcellular protein localization/distribution. For example, cell membrane proteins that are expressed in a polarized manner in normal cells can be altered in disease, resulting in distribution of the protein in a non-polar manner across the entire cell surface.

The phenomenon of altered subcellular protein localization in a disease state has been demonstrated with MUC1 and Her2 protein expression using immunohistochemical means. Normal epithelial cells have a typical apical distribution of MUC1, in addition to some supranuclear localization of the glycoprotein, whereas malignant lesions often demonstrate an apolar labeling pattern (Diaz et al, Breast Journal, 7; 40-45 (2001); Zhang et al, Clinical Cancer Research, 4; 2669-2676 (1998): Cao, et al, The Journal of Histochemistry and Cytochemistry, 45: 1547-1557 (1997)). In addition, normal mammary epithelium is either negative for Her2 protein or exhibits only a basolateral distribution while malignant cells may express the protein over the entire cell surface (De Potter, et al, International Journal of Cancer, 44; 969-974 (1989): McCormick, et al , 117;935-943 (2002)). Alternatively, distribution of the protein may change from a surface-only localization to include diffuse cytoplasmic expression in the disease state. One such example can be seen with MUC1 (Diaz, et al, The Breast Journal, 7: 40-45 (2001)).

Alteration in the localization/distribution of a protein within the cell, as detected by immunohistochemical methods, may also provide valuable information regarding the benefit of certain treatment modalities. This last point is illustrated by a situation where a protein can be intracellular in normal tissue, but cell surface in malignant cells; the cell surface localization makes the cells advantageously amenable to antibody-based diagnostics and treatment regimens. When such a change of protein localization occurs for PSCA, the PSCA protein and immune responses related thereto are very useful. Accordingly, the ability to determine whether alteration of subcellular protein localization occurred in the 24P4C12-PSCA protein and immune responses related thereto proves very useful. Application of the PSCA compositions allows professionals in the field to make important diagnostic and therapeutic decisions.

Immunohistochemical reagents specific for PSCA are also useful for detecting metastases of tumors expressing PSCA when the polypeptide appears in tissues where PSCA is not normally produced.

Thus, PSCA polypeptides and antibodies which are the result of immune responses are therefore applicable in a number of important contexts such as diagnostic, prognostic, preventive and/or therapeutic purposes known to professionals in the field.

Just as PSA polynucleotide fragments and polynucleotide variants are used by those skilled in the art for use in methods of monitoring PSA, PSCA polynucleotide fragments and polynucleotide variants are used in an analogous manner. In particular, typical PSA polynucleotides are used in methods for monitoring PSA probes or primers consisting of fragments of the PSA cDNA sequence. To illustrate this, primers used to PCR amplify a PS A polynucleotide must encompass less than the entire PSA sequence to function in the polymerase chain reaction. In the context of such PCR reactions, those skilled in the art generally generate a variety of different polynucleotide fragments that can be used as primers to amplify different parts of a polynucleotide of interest or to optimize amplification reactions (see, e.g., Caetano-Anolles, G. Biotechniques 25(3) : 472-476, 478-480 (1998), Robertson et al., Methodes Mol. Biol.

98:121-154 (1998)). A further illustration of the use of such fragments is provided in the example entitled "Expression Analysis of PSCA in Normal Tissue and Patient Samples," where a PSCA polynucleotide fragment is used as a probe to show expression of PSCA RNAs in cancer cells. In addition, variant polynucleotide sequences are typically used as primers and probes for the corresponding mRNAs in PCR and Northern analyzes (see e.g. Sawai et al, Fetal Diagn. Ther. 1996 Nov-Dec 11(6):407-13 and Current Protocols In Molecular Biology, Volume 2, Unit 2, Frederick M. Ausubel et al. eds., 1995)). Polynucleotide fragments and variants are useful in this context where they are able to bind to a target polynucleotide sequence (e.g. a PSCA polynucleotide shown in figure 1 or variant thereof) under conditions of high stringency.

Furthermore, PSA polypeptides containing an epitope recognized by an antibody or T cell that specifically binds to that epitope can be used in methods of monitoring PSA. PSCA polypeptide fragments and polypeptide analogs or variants can also be used in an analogous manner. This practice of using polypeptide fragments or polypeptide variants to form antibodies (such as anti-PSA antibodies or T cells) is typical in the field of a variety of systems such as fusion proteins used by medical practitioners (see, e.g., Current Protocols In Molecular Biology, Volume 2, Unit 16, Frederick M. Ausubel et al eds., 1995). In this context, each epitope(s) functions to provide the architecture with which an antibody or T cell is reactive. Those skilled in the art typically generate a number of different polypeptide fragments that can be used to generate immune responses specific to different parts of a polypeptide of interest (see, e.g., US Patent No. 5,840,501 and US Patent No. 5,939,533). For example, it may be preferred to use a polypeptide comprising one of the PSCA biological motifs described herein or a motif-bearing subsequence that is readily identified by those skilled in the art based on motifs available in the art. Polypeptide fragments, variants or analogues are typically useful in this context as long as they comprise an epitope capable of generating an antibody or T cell specific for a target polypeptide sequence (eg a PSCA polypeptide shown in Figure 1).

As shown here, the PSCA polynucleotides and polypeptides (as well as the PSCA polynucleotide probes and the anti-PSCA antibodies or T cells used to identify the presence of these molecules) exhibit specific properties that make them useful in diagnosing cancers such as those listed in Table I. Diagnostic tests that measure the presence of PSCA gene products, to evaluate the presence or onset of a disease state described herein, such as prostate cancer, are used to identify patients for preventive measurements or further monitoring, which is then successfully performed with PSA. Furthermore, these materials satisfy a need in the field for molecules that have similar or complementary characteristics to PSA in situations where, for example, a definitive diagnosis of metastasis of prostatic origin cannot be made on the basis of a test for PSA alone (see e.g. Alanen etal., Pathol. Res. Pract. 192(3): 233-237 (1996)) and thus materials such as PSCA polynucleotides and polypeptides (as well as the PSCA polynucleotide probes and anti-PSCA antibodies used to identify the presence of these molecules) to be used to confirm a metastasis of prostatic origin.

Finally, in addition to their use in diagnostic assays, the PSCA polynucleotides described herein have several other practical uses such as their use in the identification of oncogenetically associated chromosomal abnormalities in the chromosomal region to which the PSCA gene maps (see the example designated " Chromosomal mapping of PSCA" below). Furthermore, in addition to their use in diagnostic assays, the PSCA-related proteins and polynucleotides described herein have other practical uses such as their use in the forensic analysis of tissues of unknown origin (see, e.g., Takahama K Forensic Sei Int 1996 Jun 28;80(l-2): 63-9).

In addition, PSCA-related proteins or polynucleotides according to the invention can be used to treat a pathological condition characterized by the overexpression of PSCA. For example, the amino acid or nucleic acid sequence of Figure 1, or fragments thereof, can be used to generate an immune response to a PSCA antigen. Antibodies or other molecules that react with PSCA can be used to modulate the function of this molecule and thereby provide a therapeutic benefit.

XII.) Inhibition of PSCA protein function

The invention encompasses various methods and compositions for inhibiting the binding of PSCA to its binding partner or its association with other protein(s) as well as methods for inhibiting PSCA function.

XE. A.) Inhibition of PSCA with intracellular antibodies

In one approach, a recombinant vector encoding single-chain antibodies that specifically bind to PSCA is introduced into PSCA-expressing cells via gene transfer technologies. Accordingly, the encoded single chain anti-PSCA antibody is expressed intracellularly, binds to PSCA protein and thereby inhibits its function. Methods for the construction of such intracellular single chain antibodies are well known. Such intracellular antibodies, also known as "intrabodies", are specifically targeted to a particular compartment within the cell, providing control over where the inhibitory activity of the treatment is focused. This technology has been successfully applied in the field (for overview see Richardson and Marasco, 1995, TIBTECH vol. 13). Intrabodies have been shown to virtually eliminate the expression of otherwise abundant cell surface receptors (see, eg, Richardson et al, 1995, Proe. Nati. Acad. Sei. USA 92: 3137-3141; Beerli et al, 1994, J. Biol. Chem . 289: 23931-23936; Deshanera/., 1994, Gene Ther. 1: 332-337).

Single chain antibodies comprise the variable domains of the heavy and light chain linked by a flexible linker polypeptide and are expressed as a single polypeptide. Optionally, single chain antibodies are expressed as a single chain variable region fragment linked to the light chain constant region. Well-known intracellular trafficking signals are engineered into recombinant polynucleotide vectors encoding such single-chain antibodies to precisely measure the intrabody of the desired intracellular compartment. For example, intrabodies targeted to the endoplasmic reticulum (IS) are engineered to introduce a leader peptide and optionally a C-terminal IS retention signal, such as the KDEL amino acid motif. Intrabodies intended to exert activity in the nucleus are engineered to include a nuclear localization signal. Lipid moieties are attached to intrabodies to tether the intrabodies to the cytosolic side of the plasma membrane. Intrabodies can also be targeted to exert function in the cytosol. For example, cytosolic intrabodies are used to sequester factors within the cytosol, thereby preventing them from being transported to their natural cellular destination.

In one embodiment, intrabodies are used to trap PSCA in the nucleus, thereby preventing its activity within the nucleus. Nuclear targeting signals are engineered into such PSCA intrabodies to achieve the desired targeting. Such PSCA intrabodies are designed to bind specifically to a particular PSCA domain. In another embodiment, cytosolic intrabodies that specifically bind to a PSCA protein are used to prevent PSCA from gaining access to the nucleus, thereby preventing it from exerting any biological activity within the nucleus (eg, preventing PSCA from formation of transcription complexes with other factors).

To specifically direct the expression of such intrabodies to particular cells, the transcription of the intrabodies is placed under the regulatory control of an appropriate tumor-specific promoter and/or enhancer. To measure intrabody expression specific to the prostate, for example, the PSA promoter and/or promoter/enhancer can be used (see, for example, US patent no. 5,919,652 published July 6, 1999).

XJJ.B.) Inhibition of PSCA by recombinant proteins

In another approach, recombinant molecules bind to PSCA and thereby inhibit PSCA function. For example, these recombinant molecules prevent or inhibit PSCA from availability/binding to its binding partner(s) or association with other protein(s). Such recombinant molecules can, for example, contain the reactive part(s) of a PSCA-specific antibody molecule. In a particular embodiment, the PSCA binding domain of a PSCA binding partner is engineered into a dimeric fusion protein, whereby the fusion protein comprises two PSCA ligand binding domains linked to the Fc portion of a human IgG, such as human IgGi. Such an IgG part may contain, for example, the CH2 and CHs domains and the hinge region, but not the CHi domain. Such dimeric fusion proteins are administered in soluble form to patients suffering from cancers associated with expression of PSCA, whereby the dimeric fusion protein specifically binds to PSCA and blocks PSCA interaction with a binding partner. Such dimeric fusion proteins are further combined into multimeric proteins using known methods

antibody linking technologies.

XLl.C.) Inhibition of PSCA transcription or translation

The present invention also includes various methods and compositions for inhibiting the transcription of the PSCA gene. Correspondingly, the invention also provides methods and compositions for inhibiting the translation of PSCA mRNA into protein.

In one approach, a method of inhibiting the transcription of the PSCA gene comprises contacting the PSCA gene with a PSCA antisense polynucleotide. In another approach, a method of inhibiting PSCA mRNA translation comprises contacting a PSCA mRNA with an antisense polynucleotide. In another approach, a PSCA-specific ribozyme is used to cleave a PSCA message, thereby inhibiting translation. Such antisense and ribozyme-based methods may also target the regulatory regions of the PSCA gene, such as PSCA promoter and/or enhancer elements. Similarly, proteins capable of inhibiting a PSCA gene transcription factor are used to inhibit PSCA mRNA transcription. The various polynucleotides and compositions applicable in the above methods are described above. The use of antisense and ribozyme molecules to inhibit transcription and translation are well known in the art.

Other factors that inhibit the transcription of PSCA by affecting PSCA transcriptional activation are also useful for treating cancers that express PSCA. Similarly, factors that interfere with PSCA processing are useful for treating cancers that express PSCA. Cancer treatment methods using such factors are also within the scope of the invention.

XJJ.D.)General Considerations for Therapeutic Strategies

Gene transfer and gene therapy technologies can be used to deliver therapeutic polynucleotide molecules to tumor cells that synthesize PSCA (ie, antisense, ribozyme, polynucleotides encoding intrabodies and other PSCA-inhibiting molecules). Several gene therapy procedures are known in the art. Recombinant vectors encoding PSCA antisense polynucleotides, ribozymes, factors capable of affecting PSCA transcription, and so on, can be delivered to target tumor cells using such gene therapy methods.

The above therapeutic approaches can be combined with any one of a variety of surgical, chemotherapy, or radiation therapy regimens. The therapeutic methods of attack according to the invention can enable the use of reduced doses of chemotherapy (or other therapies) and/or less frequent administration, an advantage for all patients and especially for those who do not tolerate the toxicity of the chemotherapeutic agent well.

The antitumor activity of a particular composition (eg, antisense, ribozyme, intrabody), or a combination of such compositions, can be evaluated using various in vitro and in vitro assay systems. In vitro tests that evaluate therapeutic activity include cell growth tests, mycagar tests and other tests indicative of tumor-promoting activity, binding tests capable of determining the extent to which a therapeutic composition will inhibit the binding of PSCA to a binding partner, etc.

In vivo, the effect of a PSCA therapeutic composition can be evaluated in a suitable animal model. For example, xenogenic prostate cancer models can be used, where human prostate cancer explants or passaged xenograft tissues are introduced into immune compromised animals, such as nude or SCID mice (Klein et al, 1997, Nature Medicine 3: 402-408). For example, PCT patent application WO98/16628 and US patent 6,107,540 describe various xenograft models of human prostate cancer capable of recapitulating the development of primary tumors, micrometastasis and the formation of osteoblastic metastases characteristic of late stage disease. Efficacy can be predicted using assays that measure inhibition of tumor formation, tumor regression or metastasis and the like.

In vitro assays evaluating the promotion of apoptosis are useful in the evaluation of therapeutic compositions. In one embodiment, xenografts from tumor-bearing mice treated with the therapeutic composition can be examined for the presence of apoptotic foci and compared to untreated control xenograft-bearing mice. The extent of apoptotic foci found in the tumors of the treated mice provides an indication of the therapeutic effectiveness of the composition.

The therapeutic compositions used in the practice of the foregoing methods may be formulated into pharmaceutical compositions comprising a carrier suitable for the desired delivery method. Suitable carriers include any material which, when combined with the therapeutic composition, retains the antitumor function of the therapeutic composition and is generally not reactive with the patient's immune system. Examples include, but are not limited to, any of several standard pharmaceutical carriers such as sterile phosphate buffered saline solutions, bacteriostatic water, and the like (see generally Remington's Pharmaceutical Sciences 16thEdition, A. Osal., Ed., 1980).

Therapeutic formulations can be solubilized and administered via any route capable of delivering the therapeutic composition to the tumor site. Potentially effective methods of administration include, but are not limited to, intravenous, parenteral, intraperitoneal, intramuscular, intratumoral, intradermal, intraorgan, orthotopic, and the like. A preferred formulation for intravenous injection comprises the therapeutic composition in a solution of preserved bacteriostatic water, sterile unpreserved water and/or diluted in polyvinyl chloride or polyethylene bags containing 0.9% sterile sodium chloride for injection, USP. Therapeutic protein preparations can be freeze-dried and stored as sterile powders, preferably under vacuum and then reconstituted in bacteriostatic water (containing, for example, benzyl alcohol preservative) or in sterile water before injection.

Doses and administration protocols for the treatment of cancer using the foregoing methods will vary with the method and the target cancer and will generally depend on several other factors recognized in the art.

Xin.) Identification, characterization and application of modulators of PSCA

Procedures for identifying and applying modulators

In one embodiment, screening is performed to identify modulators that induce or suppress a particular expression profile, suppress or induce specific pathways, preferentially producing the associated phenotype thereby. In another embodiment, having identified differentially expressed genes important in a particular condition; screens are performed to identify modulators that alter the expression of individual genes, either increasing or decreasing. In another embodiment, screening is performed to identify modulators that alter a biological function of the expressed product of a differentially expressed gene. Having in turn identified the significance of a gene in a particular condition, screens are performed to identify agents that bind and/or modulate the biological activity of the gene product.

In addition, screens are performed for genes that are induced in response to a candidate agent. After identification of a modulator (one that suppresses a cancer expression pattern leading to a normal expression pattern or a modulator of a cancer gene that leads to expression of the gene as in normal tissue) a screen is performed to identify genes that are specifically modulated in response to the agent. Comparison of expression profiles between normal tissue and drug-treated cancer tissue reveals genes that are not expressed in normal tissue or cancer tissue, but are expressed in drug-treated tissue and vice versa. These agent-specific sequences have been identified and used by methods described here for cancer genes or proteins. In particular, these sequences and the proteins they encode are used in the labeling or identification of drug-treated cells. In addition, antibodies have been raised against the agent-induced proteins and used to measure new therapeutic agents for the treated cancer tissue sample.

Modulator-related identification and screening tests:

Gene expression-related experiments

Proteins, nucleic acids and antibodies according to the invention are used in screening experiments. The cancer-associated proteins, antibodies, nucleic acids, modified proteins and cells containing these sequences are used in screening experiments, such as evaluation of the effect of drug candidates on a "gene expression profile," expression profile of polypeptides or change of biological function. In one embodiment, expression profiling is used, preferably in conjunction with high-throughput screening techniques to allow monitoring of expression profiling genes following treatment with a candidate agent (eg, Davis, GF, et al, J Biol Screen 7:69 (2002); Zlokarnik, et al, Science 279:84-8 (1998); Heid, GenomeRes 6:986-94,1996).

The cancer proteins, antibodies, nucleic acids, modified proteins and cells containing the natural or modified cancer proteins or genes are used in screening experiments. The present invention includes methods for screening for compositions that modulate the cancer phenotype or a physiological function of a cancer protein according to the invention. This is done on a gene itself or by assessing the effect of drug candidates on a "gene expression profile" or biological function. In one embodiment, expression profiles are used, preferably in conjunction with high-throughput screening techniques to allow monitoring after treatment with a candidate agent, see Zlokamik, supra.

A number of tests have been carried out targeting the genes and proteins according to the invention. Experiments are run on an individual nucleic acid or protein level. That is, having identified a particular gene as upregulated in cancer, test compounds are screened for the ability to modulate gene expression or for binding to the cancer protein according to the invention. "Modulation" in this context includes an increase or a decrease in gene expression. The preferred amount of modulation will depend on the initial change in gene expression in normal versus tissue undergoing cancer, with changes of at least 10%, preferably 50%, more preferably 100-300% and in some embodiments 300-1000% or greater. Thus, if a gene exhibits a 4-fold increase in cancer tissue compared to normal tissue, a reduction of approximately four-fold is often desired; corresponding to a 10-fold reduction in cancer tissue compared to normal tissue, a target value of a 10-fold increase in expression of the test compound is often desired. Modulators that exacerbate the type of gene expression seen in cancer are also useful, e.g. as an up-regulated target in further analyses.

The amount of gene expression is monitored using nucleic acid probes and the quantification of gene expression levels, or alternatively a gene product itself is monitored, e.g. through the use of antibodies to the cancer protein and standard immunoassays. Proteomics and separation techniques also allow quantification of expression.

Expression monitoring to identify compounds that modify gene expression

In one embodiment, gene expression monitoring, i.e., an expression profile, is monitored simultaneously for multiple entities. Such profiles will typically involve one or more of the genes from figure 1.1, this embodiment is e.g. cancer nucleic acid probes attached to biochips to detect and quantify cancer sequences in a particular cell. Alternatively, PCR can be used. Thus, a series, e.g. wells in a microtiter plate are used with distributed primers in the desired wells. A PCR reaction can then be performed and analyzed for each well.

Expression monitoring is performed to identify compounds that modify the expression of one or more cancer-associated sequences, e.g. a polynucleotide sequence set forth in Figure 1. Generally, a test modulator is added to the cells prior to analysis. Furthermore, screens are also provided to identify agents that modulate cancer, modulate cancer proteins according to the invention, bind to cancer protein according to the invention or intervene in the binding of cancer protein according to the invention and an antibody or other binding partner.

In one embodiment, high-throughput screening methods involve providing a library containing a large number of potential therapeutic compounds (candidate compounds). Such "combinatorial chemical libraries" are then screened in one or more experiments to identify those library members (specifically chemical species or subclasses) that exhibit a desired characteristic activity. The compounds thus identified may serve as conventional "lead compounds," as compounds for screening, or as therapeutic agents.

In certain embodiments, combinatorial libraries of potential modulators are screened for an ability to bind to the cancer polypeptide or to modulate activity. Conventionally, new chemical entities with useful properties are formed by the identification of a chemical compound (termed a "lead compound") with some desirable property or activity, e.g. inhibitory activity, producing variants of the leader compound and evaluating the property and activity of those variant compounds. Often, high-throughput screening (HTS) methods are used for such an analysis.

As noted above, gene expression monitoring is conveniently used to test candidate modulators (eg, protein, nucleic acid, or small molecule). After the candidate agent has been added and the cells allowed to incubate for a period, the sample containing a target sequence that has been analyzed e.g. set to a biochip.

If necessary, the target sequence is prepared using known techniques. For example, a sample is processed to lyse the cells, using known lysis buffers, electroporation, etc., with purification and/or amplification such as PCR performed as appropriate. For example, an in v/7/*o transcription with tags covalently bound to the nucleotides is performed. In general, the nucleic acids are labeled with biotin-FITC or PE or with cy3 or cy5.

The target sequence can be marked with e.g. a fluorescent, a chemiluminescent, a chemical or a radioactive signal to provide a means of detecting the target sequence-specific binding of a probe. The label can also be an enzyme, such as alkaline phosphatase or horseradish peroxidase, which when provided with an appropriate substrate provides a product that is detected. Alternatively, the label is a labeled compound or small molecule, such as an enzyme inhibitor, that binds but is not catalyzed or altered by the enzyme. The label can also be a group or compound, such as an epitope tag or biotin that specifically binds to streptavidin. For the biotin example, the streptavidin is labeled as described above, thereby providing a detectable signal for the bound target sequence. Unbound labeled streptavidin is typically removed before analysis.

As will be recognized by those skilled in the art, these assays may be direct hybridization assays or may comprise "sandwich assays" involving the use of multiple probes, as generally described in US Patent Nos. 5,681,702; 5,597,909; 5,545,730; 5,594,117; 5,591,584; 5,571,670; 5,580,731; 5,571,670; 5,591,584; 5,624,802; 5,635,352; 5,594,118; 5,359,100; 5,124,246; and 5,681,697.1 this embodiment, the target nucleic acid is generally prepared as described above and then added to the biochip comprising an array of nucleic acid probes, under conditions that allow the formation of a hybridization complex.

A number of hybridization conditions are used according to the present invention, including high, moderate and low stringency conditions as described above. The experiments are generally run under stringent conditions that allow formation of the labeled probe hybridization complex only in the presence of target. Stringency can be controlled by changing a step parameter which is a thermodynamic variable, including but not limited to temperature, formamide concentration, salt concentration, chaotropic salt concentration, pH, organic solvent concentration, etc. These parameters can also be used to control non-specific binding which is generally described in US Patent No. 5,681,697. Thus, it may be desirable to perform certain steps at higher stringency conditions to reduce non-specific binding.

The reactions described here can be carried out in a number of ways. Components of the reaction may be added simultaneously or sequentially, in different orders, with preferred embodiments described below. In addition, the reaction may include a variety of other reagents. These include salts, buffers, neutral proteins, e.g. albumin, detergents, etc. which can be used to facilitate optimal hybridization and detection and/or reduce non-specific or background interactions. Reagents that otherwise improve the efficiency of the assay, such as protease inhibitors, nuclease inhibitors, antimicrobial agents, etc., may also be used as appropriate, depending on the sample preparation procedures and purity of the target. The experimental data is analyzed to determine the expression levels of individual genes and changes in expression levels between conditions, forming a gene expression profile.

Biological activity-related experiments

The invention provides methods for identifying or screening a compound that modulates the activity of a cancer-related gene or protein according to the invention. The methods comprise adding a test compound, as defined above, to a cell comprising a cancer protein according to the invention. The cells contain a recombinant nucleic acid which codes for a cancer protein according to the invention. In another embodiment, a library of candidate agents is tested on a variety of cells.

In one aspect, the experiments are evaluated in the presence or absence or prior or subsequent exposure of physiological cues, e.g. hormones, antibodies, peptides, antigens, cytokines, growth factors, action potentials, pharmacological agents including chemotherapeutics, radiation, carcinogenic or other cells (ie cell-cell contacts). In another example, the determinations are made at different stages of the cell cycle process. In this way, compounds that modulate genes or proteins according to the invention have been identified. Compounds with pharmacological activity are capable of enhancing or interfering with the activity of the cancer protein according to the invention. Once identified, similar structures are evaluated to identify critical structural features of the compound.

In one embodiment, a method of modulating (eg, inhibiting) cancer cell division is provided; the method comprises the administration of a cancer modulator. In another embodiment, a method of modulating (eg, inhibiting) cancer is provided; the method comprises administering a cancer modulator. In a further embodiment, methods of treating cells or individuals with cancer are provided; the method comprises administering a cancer modulator.

In one embodiment, a method for modulating the status of a cell expressing a gene according to the invention is provided. As used herein, status includes such species-accepted parameters as growth, proliferation, survival, function, apoptosis, aging, localization, enzymatic activity, signal transduction, etc. of a cell. In one embodiment, the cancer inhibitor is an antibody as described above. In another embodiment, the cancer inhibitor is an antisense molecule. A number of cell growth, proliferation and metastasis assays are known to those skilled in the art as described herein.

High-throughput screening to identify modulators

The attempts to identify suitable modulators are amenable to high-throughput screening. Thus, preferred assays detect enhancement or inhibition of cancer gene transcription, inhibition or enhancement of polypeptide expression, and inhibition or enhancement of polypeptide activity.

In one embodiment, modulators are evaluated in high throughput

screening methods proteins, often naturally occurring proteins or fragments of naturally occurring proteins. Thus, e.g. cellular extracts containing proteins

or random or directed digestions of proteinaceous cellular extracts used. In this way, libraries of proteins are made for screening by the methods according to the invention. Particularly preferred in this embodiment are libraries of bacterial, fungal, viral and mammalian proteins, with the latter being preferred and human proteins being particularly preferred. Particularly useful test compound will be directed against the class of proteins to which the target belongs, e.g. substrates for enzymes or ligands and receptors.

Application of soft agar growth and colony formation to identify and characterize modulators

Normal cells require a solid substrate to attach and grow. When cells are transformed, they lose this phenotype and grow separately from the substrate. For example, transformed cells can be grown in stirred suspension culture or suspended in semi-solid media, such as semi-solid or soft agar. The transformed cells, when transfected with tumor suppressor genes, can regenerate normal phenotype and again require a solid substrate to attach to and grow. Soft agar growth or colony formation assays are used to identify modulators of cancer sequences that, when expressed in host cells, inhibit abnormal cellular proliferation and transformation. A modulator reduces or eliminates

the host cell's ability to grow suspended in solid or semisolid media, such as agar.

Techniques for soft agar growth or colony formation in suspension experiments are described in Freshney, Culture of Animal Cells a Manual of Basic Technoque (3rd ed., 1994). See also the procedure section of Garkavtsev et al. (1996), supra.

Evaluation of contact inhibition and growth density limitation to identify and characterize modulators

Normal cells typically grow in a flat and organized pattern in cell culture until they touch other cells. When the cells touch each other, they are contact inhibited and stop waxing. However, transformed cells are not contact-inhibited and continue to grow to high densities in disorganized foci. Thus, transformed cells grow to a higher saturation density than corresponding normal cells. This is detected morphologically by the formation of a disoriented monolayer of cells or cells in foci. Alternatively, labeling index with (<3>H)-thymidine at saturation density is used to measure density limitation of growth, corresponding to an MTT or Alamar Blue assay, will reveal the proliferative capacity of cells and the ability of modulators to affect same. See Freshney (1994), supra. Transformed cells, when transfected with tumor suppressor genes, can regenerate a normal phenotype and become contact-inhibited and wildly grow to a lower density.

In this experiment, labeling index with 3H)-thymidine at saturation density is a preferred method for measuring density limitation of growth. Transformed host cells are transfected with a cancer-associated sequence and are grown for 24 hours at saturating density in non-limiting medium conditions. The percentage of cell labeling with (<3>H)-thymidine is determined by incorporated cpm.

Contact-independent growth is used to identify modulators of cancer sequences, which had led to abnormal cellular proliferation and transformation. A modulator reduces or eliminates contact-dependent growth and returns the cells to a normal phenotype.

Evaluation of growth factor or serum dependence to identify and characterize modulators

Transformed cells have lower serum dependence than their normal counterparts (see, e.g., Temin, J. Nat. Cancer Inst. 37:167-175 (1966); Eagle et al., J. Exp. Med 131:836-879 (1970)) ; Freshney, supra. This is partly due to the release of various growth factors from the transformed cells. The degree of growth factor or serum dependence of transformed host cells can be compared with that of control. For example, growth factor or serum dependence of a cell is monitored by methods for identifying and characterizing compounds that modulate cancer-associated sequences according to the invention.

Application of tumor-specific marker levels to identify and characterize modulators

Tumor cells release an increased amount of certain factors (called "tumor-specific markers") than their normal counterparts. For example, plasminogen activator (PA) is released from human glioma at a higher level than from normal brain cells (see, e.g., Gullino, Angiogenesis, Tumor Vascularization, and Potential Interference with Tumor Growth, in Biological Responses in Cancer, pp. 178- 184 (Mihich (ed.) 1985)). Similarly, tumor angiogenesis factor (TAF) is released at a higher level in tumor cells than their normal counterparts. See e.g. Folkman, Angiogenesis and Cancer, Sem. Cancer Biol. (1992)), while bFGF is released from endothelial tumors (Ensoli, B etal).

Various techniques that measure the release of these factors are described in Freshney (1994), supra. See also Unkless et al, J. Biol. Chem. 249:4295-4305 (1974); Strickland & Beers, J. Biol. Chem. 251:5694-5702 (1976); Whur etal., Br. J. Cancer 42:305 312 (1980); Gullino, Angiogenesis, Tumor Vascularization, and Potential Interference with Tumor Growth, in Biological Responses in Cancer, pp. 178-184 (Mihich (ed.) 1985); Freshney, Anticancer Res.

5:111-130 (1985). For example, tumor-specific marker levels are monitored by methods for identifying and characterizing compounds that modulate cancer-associated sequences according to the invention.

Invasibility into Matrigel to identify and characterize modulators

The degree of invasion ability into Matrigel or an extracellular matrix component can be used as an attempt to identify and characterize compounds that modulate cancer-associated sequences. Tumor cells show a positive correlation between malignant disease and the ability of cells to invade Matrigel or any other extracellular matrix component. In this experiment, tumorigenic cells are typically used as host cells. Expression of a tumor suppressor gene in these host cells would reduce the invasiveness of the host cells. Techniques described in Cancer Res. 1999; 59:6010; Freshney (1994), supra, can be applied. Briefly, the level of host cell invasion is measured using filters coated with Matrigel or some other extracellular matrix component. Penetration into the gel or through to the distal side of the filter is rated as invasion ability and rated histologically by the number of cells and distance moved or by labeling the cells with <125>1 and counting the radioactivity on the distal side of the filter or bottom of the dish. See e.g. Freshney (1984), supra.

Evaluation of tumor growth in vivo to identify and characterize modulators

Effects of cancer-associated sequences on cell growth are tested in transgenic or immunosuppressive organisms. Transgenic organisms are produced in a number of species-accepted ways. For example, knock-out transgenic organisms, e.g. mammals such as mice made, where a cancer gene is destroyed or where a cancer gene is inserted. Knock-out transgenic mice are produced by insertion of a marker gene or other heterologous gene into the endogenous cancer gene site in the mouse genome via homologous recombination. Such mice can also be made by substituting the endogenous cancer gene with a mutated version of the cancer gene or by mutating the endogenous cancer gene, e.g. by exposure to carcinogens.

To produce transgenic chimeric animals, e.g. mice, a DNA construct is introduced into the nuclei of embryonic stem cells. Cells containing the newly engineered genetic lesion are injected into a host mouse embryo, which is reimplanted into a recipient female mouse. Some of these embryos develop into chimeric mice that have germ cells some of which are derived from the mutant cell line. Therefore, by breeding the chimeric mice, it is possible to obtain a new line of mice containing the introduced genetic lesion (see, e.g., Capecchi et al., Science 244:1288

(1989)). Chimeric mice can be derived according to US Patent 6,365,797, published April 2, 2002; US Patent 6,107,540 published August 22, 2000; Hogan et al, Manipulating the Mouse Embryo: A laboratory Manual, Cold Spring Harbor Laboratory (1988) and Teratocarcinomas and Embryonic Stem Cells: A Practical Approach, Robertson, ed., IRL Press, Washington, D.C., (1987).

Alternatively, various immunosuppressive or immunodeficient host animals can be used. For example, a genetically athymic "nude" mouse (see, e.g., Giovanella et al., J. Nat. Cancer Inst. 52:921 (1974)), an SCLD mouse, a thymectornized mouse or an irradiated mouse (see, e.g., Bradley et al., Br. J. Cancer 38:263 (1978); Selby et al., Br. J. Cancer 41:52 (1980)) is used as a host. Transplantable tumor cells (typically about 106 cells) injected into isogenic hosts produce invasive tumors in a high proportion of cases, whereas normal cells of similar origin will not. In hosts that develop invasive tumors, cells expressing cancer-associated sequences are injected subcutaneously or orthotopically. Mice are then separated into groups, comprising control groups and treated experimental groups) e.g. treated with a modulator). After a suitable period of time, preferably 4-8 weeks, tumor growth is measured (eg by volume or by its two largest dimensions or weight) and compared to the control. Tumors that have a statistically significant reduction (using, for example, Student's t-test) are indicated as having inhibited growth.

In vitro experiments to identify and characterize modulators

Experiments to identify compounds with modulating activity can be performed in vitro. For example, a cancer polypeptide is first contacted with a potential modulator and incubated for a suitable period of time, e.g. from 0.5 to 48 hours. In one embodiment, the cancer polypeptide levels are determined in vitro by measuring the level of protein or mRNA. The level of protein is measured using immunoassays such as Western blotting, ELISA and the like with an antibody that selectively binds to the cancer polypeptide or a fragment thereof. For measurement of mRNA, amplification, e.g. when using PCR, LCR or hybridization experiments, e.g. Northern hybridization, RNAse protection, dot blotting preferred. The level of protein or mRNA is detected using directly or indirectly labeled detection means, e.g. fluorescent or radioactively labeled nucleic acids, radioactively or enzymatically labeled antibodies and the like, as described herein.

Alternatively, a reporter gene system can be designed using a cancer protein promoter operably linked to a reporter gene such as luciferase, green fluorescent protein, CAT or β-gal. The reporter construct is typically transfected into a cell. After treatment with a potential modulator, the amount of reporter gene transcription, translation or activity is measured according to standard techniques known to those skilled in the art (Davis GF, supra, Gonzalez, J. & Negulescu, P. Curr. Opin. Biotechnol. 1998: 9:624 ).

As described above, in v/7/*o screens are performed on individual genes and gene products. It is who have identified a particularly differentially expressed gene that is important in a particular condition, screening for modulators of the expression of the gene or the gene product itself is carried out.

In one embodiment, screening for modulators of expression of specific gene(s) is performed. Typically, expression of only one or a few genes is evaluated. In another embodiment, screens are designed to first find compounds that bind to differentially expressed proteins. These compounds are then evaluated for the ability to modulate differentially expressed activity. Furthermore, once initial candidate compounds are identified, variants can be further screened to better evaluate structure-activity relationships.

Binding experiments to identify and characterize modulators

In binding experiments according to the invention, a purified or isolated gene product according to the invention is generally used. For example, antibodies are formed to a protein according to the invention and immunoassays are run to determine the amount and/or localization of the protein. Alternatively, cells containing the cancer proteins are used in the experiments.

Thus, the methods comprise combining the cancer protein according to the invention and a candidate compound such as a ligand and determining the binding of the compound to the cancer protein according to the invention. Preferred embodiments employ the human cancer protein; Animal models of human disease can also be developed and used. Other analogous mammalian proteins can also be used as recognized by those skilled in the art. In some embodiments, further variants or derivatives of cancer proteins are used.

In general, the cancer protein according to the invention or the ligand is non-diffusibly bound to an insoluble carrier. The carrier can e.g. be one that has isolated sample receiving areas (a microtiter plate, an array, etc). The insoluble carrier may be made of any composition to which the compositions may be bound, is readily separated from soluble material, and is otherwise compatible with the overall screening procedure. The surface of such carriers may be solid or porous and of any suitable shape.

Examples of suitable insoluble supports include microtiter plates, arrays, membranes and beads. These are typically made of glass, plastic (e.g. polystyrene), polysaccharide, nylon, nitrocellulose or Teflon™, etc. Microtiter plates and arrays are particularly convenient because a large number of experiments can be performed simultaneously, using small amounts of reagents and samples. The particular manner of binding the composition to the carrier is not important as long as it is compatible with the reagents and overall methods of the invention, maintains the activity of the composition and is non-diffusible. Preferred methods of binding include the use of antibodies that do not sterically block either the ligand binding site or the activation sequence when binding the protein to the support, direct binding to "sticky" or ionic supports, chemical cross-linking, the synthesis of the protein or agent on the surface, etc. After binding the protein or the ligand/binding agent to the carrier, excess unbound material is removed by washing. The sample receiving areas can then be blocked through incubation with bovine serum albumin (BSA), casein, or other innocuous protein or group.

Once a cancer protein according to the invention is bound to the carrier and a test compound is added to the experiment. Alternatively, the candidate binding agent is bound to the carrier and the cancer protein according to the invention is then added. Binding agents include specific antibodies, non-natural binding agents identified in chemical library screenings, peptide analogs, etc.

Of particular interest are attempts to identify agents that have a low toxicity to human cells. A number of experiments can be used for this purpose, including proliferation experiments, cAMP experiments, labeled in vitro protein-protein binding experiments, electrophoretic mobility shift experiments, immunoassays for protein binding, functional experiments (phosphorylation experiments, etc.) and the like.

A determination of binding of the test compound (ligand, binding agent, modulator, etc.) to a cancer protein according to the invention can be carried out in several ways. The test compound can be labeled and binding determined directly, e.g. by binding all or part of the cancer protein according to the invention to a solid support, adding a labeled candidate compound (e.g. a fluorescent label), washing off excess reagent and determining whether the label is present on the solid support. Different blocking and washing steps can be used as appropriate.

In certain embodiments, only one of the components is labeled, e.g. a protein according to the invention or ligands labeled. Alternatively, more than one component is marked with different marks,/e£s. I<125> for the proteins and a fluorophore for the compound. Proximity reagents,/e£s. quenching or energy transfer reagents are also useful.

Competitive binding to identify and characterize modulators

In one embodiment, the binding of the "test compound" is determined by competitive binding assay with a "competitor." The competitor is a binding group that binds to the target molecule (e.g. a cancer protein according to the invention). Competitors include compounds such as antibodies, peptides, binding partners, ligands, etc. Under certain circumstances, the competitive binding between the test compound and the competitor displaces the test compound. In one embodiment, the test compound is labeled. Either the test compound, the competitor, or both are added to the protein for a time sufficient to allow binding. Incubations are carried out at a temperature that facilitates optimal activity, typically between four and 40°C. Incubation periods are typically optimized, e.g. to facilitate rapid high-throughput screening; typically between zero and one hour will be sufficient. Excess reagent is generally removed or washed away. The second component is then added and the presence or absence of the labeled component is followed to indicate binding.

In one embodiment, the competitor is added first, followed by the test compound. Displacement of the competitor is an indication that the test compound binds to the cancer protein and is thus able to bind to and potentially modulate the activity of the cancer protein. In this embodiment, any component may be labeled. Thus, e.g. if the competitor is labeled, the presence of label in the post-test compound wash solution indicates displacement of the test compound. Alternatively, if the test compound is labeled, the presence of the label on the carrier indicates displacement.

In an alternative embodiment, the test compound is added first, with incubation and washing, followed by the competitor. The absence of binding by the competitor indicates that the test compound binds to the cancer protein with higher affinity than the competitor. Thus, if the test compound is labeled, the presence of the label on the carrier, coupled with a lack of competitor binding, indicates that the test compound binds to and thus potentially modulates the cancer protein of the invention.

Consequently, the competitive binding methods include differential screening to identify agents that can modulate the activity of the cancer proteins according to the invention. In this embodiment, the methods comprise combining a cancer protein and a competitor in a first sample. Another sample comprises a test compound, the cancer protein and a competitor. The binding of the competitor is determined for both samples and a change or difference in binding between the two samples indicates the presence of an agent capable of binding to the cancer protein and potentially modulating its activity. It is if the binding of the competitor is different in the second sample compared to the first sample that the agent is able to bind to the cancer protein.

Alternatively, differential screening is used to identify drug candidates that bind to the natural cancer protein but cannot bind to modified cancer proteins. For example, the structure of the cancer protein has been modeled and used in rational drug design to synthesize agents that interact with that site, agents that generally do not bind to site-modified proteins. Furthermore, such drug candidates that affect the activity of a natural cancer protein have also been identified by screening drugs for the ability to either enhance or reduce the activity of such proteins.

Positive controls and negative controls can be used in the experiments. Preferably, control and test samples are carried out in at least triplicate to obtain statistically significant results. Incubation of all samples occurs for a time sufficient to allow binding of the agent to the protein. After incubation, samples are washed free of non-specifically bound material and the amount of bound, generally labeled agent determined. For example, where a radiolabel is used, the samples can be counted in a scintillation counter to determine the amount of bound compound.

A number of other reagents can be included in the screening test. These include reagents such as salts, neutral proteins, e.g. albumin, detergents, etc. which are used to facilitate optimal protein-protein binding and/or reduce non-specific or background interactions. Also reagents which otherwise improve the efficiency of the experiment, such as protease inhibitors, nuclease inhibitors, antimicrobial agents, etc., can be used. The mixture of components is added in an order that provides the desired bond.

Use of polynucleotides to down-regulate or inhibit a protein according to the invention.

Polynucleotide cancer modulators can be introduced into a cell containing the target nucleotide sequence by forming a conjugate with a ligand binding molecule, as described in WO 91/04753. Suitable ligand binding molecules include, but are not limited to, cell surface receptors, growth factors, other cytokines, or other ligands that bind to cell surface receptors. Preferably, conjugation of the ligand binding molecule does not substantially interfere with the ability of the ligand binding molecule to bind to its corresponding molecule or receptor or block entry of the sense or antisense oligonucleotide or its conjugated version into the cell. Alternatively, a polynucleotide modulator of cancer can be introduced into a cell containing the target nucleic acid sequence, /e£s. by forming a polynucleotide-lipid complex, as described in WO 90/10448. It is understood that the use of antisense molecules or "knock out" and "knock in" models can also be used in screening experiments as described above, in addition to methods of treatment.

Inhibitory and antisense nucleotides

In certain embodiments, the activity of cancer-associated protein is down-regulated or completely inhibited, by the use of antisense polynucleotide or inhibitory small nuclear RNA (snRNA), i.e. a nucleic acid complementary to and which can preferentially hybridize specifically to, a coding mRNA nucleic acid sequence,/e£s. a cancer protein according to the invention, mRNA or a subsequence thereof. Binding of the antisense polynucleotide to the mRNA reduces translation and/or stability of the RNA.

In the context of the present invention, antisense polynucleotides may comprise naturally occurring nucleotides or synthetic species formed from naturally occurring subunits or their close homologues. Antisense polynucleotides may also have altered sugar groups or intersugar linkages. Examples among these are the phosphorothioate and other sulfur-containing species known for use in the area. Analogues are encompassed by the present invention as long as they function effectively to hybridize with nucleotides according to the invention. See e.g. Isis Pharmaceuticals, Carlsbad, CA; Sequitor, Inc., Natick, MA.

Such antisense polynucleotides can be readily synthesized using recombinant means or can be synthesized in vitro. Equipment for such synthesis is sold by many vendors including Applied Biosystems. The preparation of other oligonucleotides such as phosphorothioates and alkylated derivatives are also well known to those skilled in the art.

Antisense molecules as used herein include antisense or sense oligonucleotides. Sense oligonucleotides can e.g. is used to block transcription by binding to the antisense strand. The antisense and sense oligonucleotide comprises a single-stranded nucleic acid sequence (either RNA or DNA) capable of binding to target mRNA (sense) or DNA (antisense) sequences of cancer molecules. Antisense or sense oligonucleotides, according to the present invention, comprise a fragment generally of at least approx. 12 nucleotides, preferably from approx. 12 to 30 nucleotides. The ability to derive an antisense or a sense oligonucleotide, based on a cDNA sequence encoding a given protein, is described in e.g. Stein & Cohen (Cancer Res. 48:2659 (1988 and van der Krol et al. (BioTechniques 6:958 (1988))).

Ribose<y>mer

In addition to antisense polynucleotides, ribozymes can be used to target and inhibit transcription of cancer-associated nucleotide sequences. A ribozyme is an RNA molecule that catalytically cleaves other RNA molecules. Various types of ribozymes have been described, including group I ribozymes, hammerhead ribozymes, hairpin ribozymes, RNase P, and axehead ribozymes (see, e.g., Castanotto et al., Adv. in Pharmacology 25: 289-317 (1994) for a general overview of the properties of different ribozymes).

The general features of hairpin ribozymes are described e.g. in Hampel et al., Nucl. Acids Res. 18:299-304 (1990); European Patent Publication No. 0360257; US Patent No. 5,254,678. Methods of preparation are well known to those skilled in the art (see, e.g., WO 94/26877; Ojwang et al., Proe. Nati. Acad. Sei. USA 90:6340-6344 (1993); Yamada et al., Human Gene Theraphy 1:39 -45 (1994); Leavitt et al, Proe. Nat. Acad Sci. USA 92:699-703 (1995); Leavitt et al, Human Gene Theraphy 5: 1151-120 (1994); and Yamada et al, Virology 205: 121 -126

(1994)).

Application of modulators in phenotypic screening

In one embodiment, a test compound is administered to a population of cancer cells, which have an associated cancer expression profile. By "administering" or "putting in contact with" is meant here that the modulator is added to the cells in such a way as to allow the modulator to act on the cell, whether by uptake and intracellular action or by action on the cell surface. In some embodiments, a nucleic acid encoding a proteinaceous agent (ie, a peptide) is inserted into a viral construct such as an adenoviral or retroviral construct and added to the cell, such that expression of the peptide agent is completed, e.g. PCT US97/01019. Adjustable gene therapy systems can also be used. Once the modulator is administered to the cells, the cells are washed if desired and allowed to incubate under preferably physiological conditions for a period. The cells are then harvested and a new gene expression profile is formed. Thus, e.g. cancer tissue screened for agents that modulate, e.g. induces, or suppresses, the cancer phenotype. A change in at least one gene, preferably many, from the expression profile indicates that the agent has an effect on cancer activity. Similarly, alteration of a biological function or a signaling pathway is indicative of modulator activity. By defining such a signature for the cancer phenotype, screens for new drugs that change the phenotype are planned. With this approach, the drug target need not be known and need not be represented in the original gene/protein expression screening platform, nor does the level of transcript that the target protein needs to change. The modulator inhibitory function will serve as a surrogate marker.

As described above, screens are performed to assess genes or gene products. It is who have identified a particularly differentially expressed gene that is important in a particular condition, screening of modulators for either expression of the gene or the gene product itself is carried out.

Use of modulators to influence peptides according to the invention

Measurements of cancer polypeptide activity or of the cancer phenotype are performed using a variety of assays. For example, the effects of modulators on the function of a cancer polypeptide(s) are measured by examining parameters described above. A physiological change affecting activity is used to assess the effect of a test compound on the polypeptides of the present invention. When the functional results are determined using intact cells or animals, a number of effects can become available such as, in the case of cancer associated with solid tumors, tumor growth, tumor metastasis, neovascularization, hormone release, transcriptional changes to both known and uncharacterized genetic markers (f .eg by Northern blots), changes in cell metabolism such as cell growth or pH changes and changes in intracellular "second messengers" such as cGNIP.

Methods for Identification of Characteristic Cancer-Associated Sequences Expression of various gene sequences is consistent with cancer. Accordingly, disorders based on mutant or variant cancer genes are determined. In one embodiment, the invention provides methods for identifying cells containing variant cancer genes,/e£s. determining the presence of, in whole or in part, the sequence of at least one endogenous cancer gene in a cell. This is accomplished using any number of sequencing techniques. The invention includes methods for identifying the cancer genotype of an individual, e.g. determination of all or part of the sequence of at least one gene according to the invention in that individual. This is generally carried out in at least one tissue from that individual, e.g. a tissue listed in Table I and may include the evaluation of several tissues or different samples of the same tissue. The method may comprise comparing the sequence of the sequenced gene to a known cancer gene, i.e. a wild type gene to determine the presence of family members, homologies, mutations or variants. The sequence of all or part of the gene can then be compared to the sequence of a known cancer gene to determine if any differences exist. This is performed using any number of known homology programs, such as BLAST, Bestfit, etc. The presence of a difference in the sequence between the cancer gene of the patient and the known cancer gene correlates with a disease state or a predisposition to a disease state, as described herein .

In a preferred embodiment, the cancer genes are used as probes to determine the number of copies of the cancer gene in the genome. The cancer genes are used as probes to determine the chromosomal localization of the cancer genes. Information such as chromosomal localization finds applications in providing a diagnosis or prognosis especially when chromosomal abnormalities such as translocations and the like are identified in the locus of the cancer gene.

XIV.) RN Ai and therapeutic application of small interfering RNAs (siRNAs)

The present invention is also directed to siRNA oligonucleotides, especially double-stranded RNAs comprising at least a fragment of the PSCA coding region or 5"-UTR regions or complement or any antisense oligonucleotide specific for the PSCA sequence. In one embodiment, such oligonucleotides are used to elucidate a function of PSCA or is used to screen for or evaluate modulators of PSCA function or expression. In another embodiment, gene expression of PSCA is reduced using siRNA transfection and results in significantly reduced proliferative capacity of transformed cancer cells which endogenously express the antigen; cells treated with specific PSCA siRNAs show reduced survival as measured, eg, by a metabolic decay of cell viability correlated to the reduced proliferative capacity. Thus, PSCA siRNA compositions comprise siRNA (double-stranded RNA) corresponding to nucleic acid -ORF sequence of PSCA protein one or subsequences thereof; these subsequences are generally 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30,31, 32, 33, 34, 35 or more than 35 consecutive RNA nucleotides in length and contain sequences complementary and non-complementary to at least part of the mRNA coding sequence. In a preferred embodiment, the subsequences are 19-25 nucleotides in length, most preferably 21-23 nucleotides in length.

RNA interference is a new approach to silencing genes in vitro and in vivo, thus small double-stranded RNAs (siRNAs) are valuable therapeutic agents. The power of siRNAs to silence specific gene activities has now been brought to animal models of disease and is also being used in humans. For example, hydrodynamic infusion of a solution of siRNA into a mouse with an siRNA against a particular target is recognized to be therapeutically effective.

The pioneering work of Song et al. indicate that one type of fully natural nucleic acid, small interfering RNAs (siRNAs), serve as therapeutic agents even without further chemical modification (Song, E., etal. "RNA interference targeting Fas protects mice from fulminant hepatitis" Nat. Med. 9( 3): 347-51(2003)). This work provided the first in vitro evidence that infusion of siRNAs into an animal could alleviate disease. In that case, the authors provided mice with injections of siRNA designed to silence the FAS protein (a cell death receptor that, when overactivated during an inflammatory response, induces hepatocytes and other cells to die). The next day, the animals were given an antibody specific to Fas. Control mice died of acute liver failure within a few days, while over 80% of the siRNA-treated mice remained free of severe disease and survived. About 80% to 90% of their liver cells incorporated the naked siRNA oligonucleotides. Furthermore, the RNA molecules worked for 10 days before losing effect after 3 weeks.

For use in human therapy, siRNA is delivered by efficient systems that elicit long-lasting RNAi activity. A major caveat ("caveat") for clinical application is the delivery of siRNAs to the appropriate cells. Hepatocytes appear to be particularly susceptible to exogenous RNA. Today, targets located in the liver are attractive because the liver is an organ that can be easily targeted by nucleic acid molecules and viral vectors. However, other tissue and organ targets are also preferred.

Formulations of siRNAs with compounds that promote transfer across cell membranes are used to improve administration of siRNAs in therapy. Chemically modified synthetic siRNA, which is resistant to nucleases and has serum stability with accompanying improved duration of RNAi effects, is a further embodiment.

Thus, siRNA technology is a therapeutic agent for human malignancy by delivering siRNA molecules targeting PSCA to individuals with the cancers as listed in Table 1. Such administration of siRNAs leads to reduced growth of cancer cells expressing PSCA and provides an antitumor therapy, reducing the morbidity and/or mortality associated with malignant disease.

The efficiency of this modality of gene product knockdown is significant when measured in vitro or in vivo. Efficacy in vitro is readily demonstrable through the application of siRNAs to cells in culture (as described above) or to aliquots of cancer patient biopsies when in v/7/*o methods are used to detect the reduced expression of PSCA protein.

XV.) Putty/articles for manufacture

For use in the laboratory, the prognostic, prophylactic, diagnostic and therapeutic applications described here are kits within the scope of the invention. Such kits may comprise a carrier, packaging or container compartmentalized to receive one or more containers such as vials, tubes and the like, each container(s) comprising one of the separate elements to be used in the method, together with a mark or insert comprising instructions for use, such as a use described herein. For example, the container(s) may comprise a probe that is or may be detectably labeled. Such a probe can be an antibody or polynucleotide specific for a protein or a gene or message according to the invention, respectively. Where the method uses nucleic acid hybridization to detect the target nucleic acid, the kit may also have containers containing nucleotide(s) for amplification of the target nucleic acid sequence. Kits may comprise a container comprising a reporter such as a biotin-binding protein, such as avidin or streptavidin, bound to a reporter molecule, such as an enzymatic, fluorescent or radioisotope label; as a reporter can be used with, e.g. a nucleic acid or antibody. The kit may comprise all or part of the amino acid sequences in figure 1, figure 2 or figure 3 or analogues thereof or a nucleic acid molecule which codes for such amino acid sequences.

The kit according to the invention will typically comprise the container described above and one or more other containers connected thereto comprising materials desirable from a commercial and user standpoint, including buffers, diluents, filters, needles, syringes; carrier, packaging, container, vial and/or tube labels listing contents and/or instructions for use and packaging inserts with instructions for use.

A label may be present on or with the container to indicate that the composition is being used for a specific therapeutic or non-therapeutic use, such as a prognostic, prophylactic, diagnostic or laboratory use and may also indicate directions for either in vivo or in vz'£ro application, such as those described here. Directions and or other information may also be included on an insert(s) or mark(s) included with or on the kit. The mark may be on or associated with the container. A mark may be on a container when letters, numbers or other characters forming the mark are cast or etched into the container itself; a mark may be associated with a container when it is present within a container or carrier that also contains the container, e.g. as a packaging insert. The mark may indicate that the composition is used for the diagnosis of, treatment of, prophylaxis ("prophylaxing") or prognosis of a condition, such as a neoplasia of a tissue listed in Table I.

The terms "putty" and "article for manufacture" can be used as synonyms.

In another embodiment according to the invention, an article(s) for manufacture containing compositions, such as amino acid sequence(s), small molecule(s), nucleic acid sequence(s) and/or antibody(s), e.g. materials useful for the diagnosis, prognosis, prevention and/or treatment of tissue neoplasias as indicated in Table I provided. The article of manufacture typically comprises at least one container and at least one label. Suitable containers include, for example, bottles, vials, syringes and test tubes. The containers can be formed from a variety of materials such as glass, metal or plastic. The container may contain amino acid sequence(s), small molecule(s), nucleic acid sequence(s), cell population(s) and/or antibody(s). In one embodiment, the container contains a polynucleotide for use in examining the mRNA expression profile of a cell, together with reagents used for this purpose. In another embodiment, a container comprises an antibody, binding fragment thereof or specific binding protein for use in evaluating protein expression of PSCA in cells and tissues or for relevant laboratory, prognostic, diagnostic, prophylactic and therapeutic purposes; indications and/or directions for such applications may be included on or with such containers, likewise reagents and other compositions or tools may be used for these purposes. In another embodiment, a container comprises materials for eliciting a cellular or humoral immune response, along with associated indications and/or directions. In another embodiment, a container comprises materials for adoptive immunotherapy, such as cytotoxic T cells (CTL) or helper T cells (HTL), along with associated indications and/or directions; reagents and other compositions or tools used for such purposes may also be covered.

Alternatively, the container may contain a composition effective for the treatment, diagnosis, prognosis or prophylaxis of a condition and may have a sterile access port (for example, the container may be an intravenous solution bag or a vial having a stopper pierceable at a hypodermic injection needle). The active agents in the composition may be an antibody capable of specific binding of PSCA and modulation of the function of PSCA.

The article for manufacture may further comprise another container comprising a pharmaceutically acceptable buffer, such as phosphate-buffered saline, Ringer's solution and/or dextrose solution. It may further include other materials desirable from a commercial and user standpoint, including other buffers, diluents, filters, stirrers, needles, syringes and/or packaging inserts with indications and/or instructions for use.

EXAMPLES:

Various aspects of the invention are further described and illustrated by way of the many examples that follow, none of which shall limit the scope of the invention.

Example 1

Expression analysis of PSCA variants in normal tissue and patient samples

Previously, PSCA, here referred to as PSCA-v.l, was identified as an antigen expressed in prostate cancer. Its expression was detected in more than 80% of primary prostate cancers and in the majority of prostate metastases. It has also been shown to be expressed in bladder cancer, ovarian cancer and pancreatic cancer; these cancers are listed in Table I. By immunohistochemical analysis, PSCA has been shown to be overexpressed on the cell surface of most urothelial transitional carcinomas and in 60% of primary pancreatic adenocarcinomas. The PSCA expression data have been reported in patent publications (PCT/US98/04664, PCT/US/28883, PCT/US00/19967) and in peer-reviewed articles (Saffran etal, Proe Nati Acad Sei USA. 2001 Feb 27; 98(5) : 2658-2663; Amara etal, Cancer Res. 2001 Jun 15; 61(12): 4660-65; Reiter etal, Proe Nati Acad Sei USA. 1998 Feb 17; 95(4): 1735-40; Argani etal, Cancer Res. 2001 Jun 1; 61(11): 4320-24).

Specific expression of different PSCA variants has been studied in normal and cancer patient samples. Primers were designed to differentiate between PSCA-v.l/v.2/v.4, PSCA-v.3 and PSCA-v.5. PSCA v. l/v.2/v.4 leads to a PCR product of 425 bp, PSCA-v.3 leads to a PCR product of 300 bp, while PSCA-v.5 leads to a PCR product of 910 bp in size (Figure ll(a).

First-strand cDNA was prepared from normal bladder, brain, heart, kidney, liver, lung, prostate, spleen, skeletal muscle, testis, pancreas, colon, stomach, collections of prostate cancer, bladder cancer, kidney cancer, colon cancer, lung cancer, ovarian cancer, breast cancer, cancer metastasis and pancreatic cancer (Figure 11(b). Normalization was performed by PCR using primers for actin. Semiquantitative PCR using the variant-specific primers was performed at 30 cycles of amplification.

Results show expression of PSCA-v.5 mainly in breast cancer, cancer metastasis and pancreatic cancer and at a lower level in colon cancer and lung cancer. PSCA v. l/v.2/v.4 PCR product was detected in prostate cancer, bladder cancer, kidney cancer, colon cancer, lung cancer, ovarian cancer, breast cancer, cancer metastasis and pancreatic cancer. Among normal tissues, PSCA-v. l/v.2/v.4-PCR product detected only in prostate, stomach and at a lower level in kidney and lung, while PSCA-v.5 was not detected in any normal tissues. PSCA v. 3-PCR-detected product was not detected in any of the samples tested.

Primers were designed to differentiate between PSCA-v.4 and PSCA-v.5 (Figure 1 J(a). PSCA-v.4 leads to a PCR product of 460 bp, while PSCA-v.5 leads to a PCR product of 945 bp in size.

First strand cDNA was prepared from normal bladder, brain, heart, kidney, liver, lung, prostate, spleen, skeletal muscle, testis, pancreas, colon, stomach, prostate cancer collections, bladder cancer and multi-xenograft collection (prostate cancer, kidney cancer and bladder cancer xenografts ) (Figure 1 J(b). Normalization was performed by PCR using primers for actin. Semiquantitative PCR using the variant-specific primers was performed at 30 cycles of amplification.

Results show expression of PSCA-v.4 in prostate cancer, bladder cancer and multi-xenograft collection, normal kidney and prostate. PSCA-v.5 was detected only in normal prostate and bladder cancer.

The restricted expression of PSCA variants in normal tissues and the expression detected in cancer patient samples indicate that PSCA variants are therapeutic, prognostic, laboratory, prophylactic and diagnostic targets for human cancer.

Example 2

Splicing variants of PSCA

As used herein, the term variant includes transcript variants and single-nucleotide polymorphisms (SNPs). Transcript variants are variants of mature mRNA from the same gene that arise from alternative transcription or alternative splicing. Alternative transcripts are transcripts from the same gene but start transcription at different points. Splice variants are mRNA variants spliced differently from the same transcript. In eukaryotes, when a multi-exon gene is transcribed from genomic DNA, the initial RNA is spliced to produce functional mRNA, which has only exons and is used for translation into an amino acid sequence. Consequently, a given gene can have zero to many alternative transcripts and each transcript can have zero to many splice variants. Each transcript variant has a unique exon composition and may have different coding and/or non-coding (5'- or 3'-end) parts from the original transcript. Transcript variants can code for the same, similar or different proteins, such proteins that have the same or a similar function or a different function. The variant proteins can be expressed in the same tissue at the same time, in a different tissue at the same time or in the same tissue at different times or in a different tissue at a different time. Proteins encoded by a transcript variant may have similar or different subcellular or extracellular localizations (eg, secreted versus intracellular).

Transcript variants have been identified by a number of species-accepted methods. For example, alternative transcripts and splice variants have been identified by full-length cloning or by using full-length transcript and EST sequences. First, all human ESTs were grouped into clusters that show direct or indirect identity with each other. Then, ESTs in the same cluster were further grouped into subclusters and assembled into a consensus sequence. The original gene sequence is compared to the consensus sequence(s) or other full-length sequences. Each consensus sequence is a potential splice variant for that gene. Many confirmation modalities are known in the art, such as identification of the variant by Northern analysis, full-length cloning or by using probe libraries, etc. Even when a variant is identified, which is not yet a full-length clone, that part of the variant is very useful as a research tool,/e£s. for antigen formation or for further cloning of the full-length splice variant using techniques known in the field.

Furthermore, computer programs are available in the field that identify transcript variants based on genomic sequences. Genomic-based transcript variant identification programs include FgenesH (A. Salamov and V. Solovyev, "Ab initio gene finding in Drosophila genomic DNA," Genome Research. 2000 April; 10(4): 516-22); Grail (URL compbio.ornl.gov/Grail-bin/EmptyGrailForm) and GenScan (URL genes.mit.edu/GENSCAN.html). For a general discussion of splice variant identification protocols see e.g. Southan, C, A genomic perspective on human proteases, FEBS Lett. 2001 Jun 8; 498(2-3):214-8; de Souza, S.J., etal, Identification of human chromosome 22 transcribed sequences with ORF expressed sequence tags, Proe. Nat Acad Sei U S A. 2000 Nov 7; 97(23): 12690-3.

To further confirm the parameters of a transcript variant, a number of techniques are available in the field such as full-length cloning, proteomics validation, PCR-based validation and 5'-RACE validation, etc. (see e.g. Proteomic Validation: Brennan, S.O., et al, Albumin banks peninsula: a new termination variantcharacterized byelectrospray mass spectrometry, Biochem Biophys Acta. 1999 Aug 17;1433(l-2):321-6; Ferranti P, et al, Differential splicing of pre-messenger RNA produces multiple forms of mature caprine alpha(sl)-casein, Eur J Biochem. 1997 Oet 1;249(1):1-7. For PCR-based validation: Wellmann S, et al, Specific reverse transcription-PCR quantification of vascular endothelial growth factor (VEGF) splice variants by LightCycler technology, Clin Chem. 2001 Apr;47(4):654-60; Jia, H.P., etal, Discovery of new human beta-defensins using a genomics-based approach, Gene. 2001 Jan 24;263(1-2):211-8. For PCR-based and 5'-RACE validation: Brigle, K.E., et al, Organization of the murine r edused folate carrier gene and identification of variant splice forms, Biochem Biophys Acta. 1997 Aug 7; 1353(2): 191-8).

It is known in the art that genomic regions are modulated in cancer. When the genomic region to which a gene maps is modulated in a particular cancer, the alternative transcripts or splice variants of the gene are also modulated. Described here is that PSCA has a particular expression profile related to cancer (see e.g. Table I). Alternative transcripts and splice variants of PSCA are also implicated in cancer, for example in one or more of these tissues and also in certain additional tissues. The variants thus serve as tumor-associated markers/antigens.

Using the full-length PSCA gene together with EST sequences, four additional transcript variants were identified, designated as PSCA-v.2, v.3, v.4 and v.5. The boundaries of exons in the original transcript were PSCA-v. 1 shown in Table VI. The sequences for PSCA and the PSCA variants are indicated in Figure 1.

Example 3

Single-nucleotide polymorphisms of PSCA

A single-nucleotide polymorphism (SNP) is a single base pair variation in a nucleotide sequence at a specific location. At any given point in the genome, there are four possible nucleotide base pairs: A/T, C/G, G/C, and T/A. As used herein, an allele is one of a series of alternative forms of a given gene, differing in DNA sequence and effect of a product (RNA and/or protein).

A SNP occurring on a cDNA is termed a cSNP. This cSNP can change amino acids of the protein coded for by the gene and thus change the function of the protein. Some SNPs cause inherited diseases; others contribute to quantitative variations in phenotype and responses to environmental factors including diet and drugs among individuals. Therefore, the existence of a SNP and/or combinations of alleles (termed haplotypes) has many useful applications, such as diagnosis of inherited diseases, determination of drug reactions and doses, identification of genes responsible for diseases and analysis of the genetic relatedness between individuals (P . Nowotny, J. M. Kwon and A. M. Goate, "SNP analysis to dissect human traits," Curr. Opin. Neurobiol. 2001 Oet; 11(5):637-641; M. Pirmohamed and B. K. Park, "Genetic susceptibility to adverse drug reactions ," Trends Pharmacol. Sei. 2001 Jun; 22(6):298-305; J. H. Riley, C. J. Allan, E. Lai and A. Roses, "The use of single nucleotide polymorphisms in the isolation of common disease genes," Pharmacogenomics . 2000 Feb; l(l):39-47; R. Judson, J. C. Stefens and A. Windemuth, "The predictive power of haplotypes in clinical response," Pharmacogenomics. 2000 Feb; 1(1): 15-26).

SNPs have been identified by a number of species-accepted methods (P. Bean, "The promising voyage of SNP target discovery," Am. Clin. Lab. 2001 Oct-Nov; 20(9): 18-20; K. M. Weiss, "In search of human variation," GenomeRes. 1998 Jul; 8(7): 691-697; M. M. She, "Enabling large-scale pharmacogenetic studies by high-throughput mutation detection and genotyping technologies," Clin. Chem. 2001 Feb; 47(2) : 164-172). For example, SNPs are identified by sequencing DNA fragments that show polymorphism by gel-based methods such as restriction fragment length polymorphism (RFLP) and denaturing gradient gel electrophoresis (DGGE). They have also been discovered by direct sequencing of DNA samples collected from different individuals or by comparing sequences from different DNA samples. With the rapid accumulation of sequence data in published and private databases, one also discovers SNPs by comparing sequences using computer programs (Z. Gu, L. Hillier and P. Y. Kwok, "Single nucleotide polymorphism hunting in cyberspace," Hum. Mutat. 1998; 12(4):221-225). SNPs can be verified and the genotype or haplotype of an individual can be determined by a variety of methods including direct sequencing and high-throughput microarrays (P. Y. Kwok, "Methods for genotyping single nucleotide polymorphisms," Annu. Rev. Genomics Hum. Genet. 2001; 2: 235-258; M. Kokoris, K. Dix, K. Moynihan, J. Mathis, B. Erwin, P. Grass, B. Hines and A. Duesterhoeft, "High-throughput SNP genotyping with the Masscode system," Mol. Diagn 2000 Dec;5(4):329-340).

Using the methods described above, thirteen SNPs were identified in the transcript for PSCA-v.2. Variant 2 was used, rather than, for example, variant 1, as it had fewer ambiguous bases than variant 1. Accordingly, SNPs were identified for PSCA-v.2 in positions 57 (t/c), 367 (c/t), 424 ( a/c), 495 (c/g), 499 (c/t), 563 (c/t), 567 (g/a), 627 (g/a), 634 (t/g), 835 (g /a), 847 (g/a), 878 (g/a) and 978 (c/g). The transcripts or proteins with alternative alleles were designated variant PSCA-v.6 through v. 18, as shown in Figure 1B and Figure 1G.

The nucleotide change in v.6 changed the start codon to v.l and thus translation would not start until the next ATG (AUG in mRNA), resulting in a protein 9 AA shorter than the v.l protein. The nucleotide changes for v. 7 and v. 8 were muted at the protein level.

Twelve of these 13 SNPs were also present in variant 4. The 12 SNP variants in relative to PSCA-v. 4 is designated PSCA-v. 19 through v.30. Variants 19 through 27 encode alternative amino acids as shown in Figure 1H.

Example 4

Production of recombinant PSCA in prokaryotes systems

To express recombinant PSCA and PSCA variants in prokaryotic cells, the full or partial length PSCA and PSCA variant cDNA sequences are cloned into any one of a variety of expression vectors known in the art. One or more of the following regions of PSCA variants are expressed: the full-length sequence presented in Figure 1 or any 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30 or more consecutive amino acids from PSCA, variants or analogues thereof.

In vitro transcription and translation constructs:

pCRJJ: To generate PSCA sense and antisense RNA probes for RNA in situ studies, pCRII constructs (Invitrogen, Carlsbad CA) were generated encoding either whole or fragments of the PSCA cDNA. The pCRII vector has Sp6 and T7 promoters flanking the insert to drive the transcription of PSCA RNA for use as probes in RNA in situ hybridization experiments. These probes are used to analyze the cellular and tissue expression of PSCA at the RNA level. Transcribed PSCA RNA representing the cDNA amino acid coding region of the PSCA gene is used in in v/7/*o translation systems such as the TnT™ Coupled Reticulolysate System (Promega, Corp., Madison, WI) to synthesize PSCA protein.

Bacterial constructs:

pGEX constructs: To generate recombinant PSCA proteins in bacteria fused to the glutathione S-transferase (GST) protein, all or part of the PSCA cDNA protein coding sequence is cloned into the pGEX family of GST fusion vectors ( Amersham Pharmacia Biotech, Piscataway, NJ). These constructs allow controlled expression of recombinant PSCA protein sequences with GST fused to the amino terminus and a six-histidine epitope (6XHis) to the carboxyl terminus. The GST and 6XHis tags allow purification of the recombinant fusion protein from induced bacteria with the appropriate affinity matrix and allow recognition of the fusion protein by anti-GST and anti-His antibodies. The 6XHis tag is formed by adding 6 histidine codons to the cloning primer at the 3' end, /e£s. of the open reading frame (ORF). A proteolytic cleavage site, such as the PreScission™ recognition site in pGEX-6P-1, can be used to allow cleavage of the GST tag from the PSCA-related protein. The ampicillin resistance gene and the pBR322 origin allow the selection and maintenance of the pGEX plasmids in E. coli.

pMAL constructs: To generate recombinant PSCA proteins, in bacteria, that are fused to maltose binding protein (MBP), all or part of the PSCA cDNA protein coding sequence is fused to the MBP gene by cloning into pMAL-c2X and pMAL -p2X vectors (New England Biolabs, Beverly, MA). These constructs allow controlled expression of recombinant PSCA protein sequences with MBP fused at the amino terminus and a 6XHis epitope tag at the carboxyl terminus. The MBP and 6XHis tags allow purification of the recombinant protein from induced bacteria with the appropriate affinity matrix and allow recognition of the fusion protein by anti-MBP and anti-His antibodies. The 6XHis epitope is formed by adding 6 histidine codons to the 3' cloning primer. A factor Xa recognition site allows cleavage of the pMAL tag from PSCA. The pMAL-c2X and pMAL-p2X vectors are optimized to express the recombinant protein in the cytoplasm or periplasm, respectively. Periplasmic expression enhances folding of proteins with disulfide bonds.

pET constructs: To express PSCA in bacterial cells, all or part of the PSCA cDNA protein coding sequence is cloned into the pET family of vectors (Novagen, Madison, WI). These vectors allow tightly controlled expression of recombinant PSCA protein in bacteria with and without fusion to solubility-enhancing proteins such as NusA and thioredoxin (Trx) and epitopages such as 6XHis- and S-Tag ™ that aid in the purification and detection of the recombinant protein. For example, constructs have been made using pET-NusA fusion system 43.1 such that regions of the PSCA protein are expressed as amino-terminal fusions to NusA.

Real constructs:

pESC constructs: To express PSCA in the yeast Saccharomyces cerevisiae for recombinant protein generation and functional studies, all or part of the PSCA cDNA protein coding sequence is cloned into the pESC family of vectors each containing 1 of 4 selectable markers, HJ .S3, TRP1, LEU2, and URA3 (Stratagene, La Jolla, CA). These vectors allow controlled expression from the same plasmid of up to 2 different genes or cloned sequences containing either Flag™ or Myc epitopes in the same yeast cell. This system is applicable to confirm protein-protein interactions of PSCA. In addition, expression in yeast introduces similar post-translational modifications, such as glycosylations and phosphorylations, that are found when expressed in eukaryotic cells.

pESP constructs: To express PSCA in the yeast species Saccharomyces pombe, all or part of the PSCA cDNA protein coding sequence is cloned into the pESP family of vectors. These

the vectors allow controlled high level expression of a PSCA protein sequence fused to either the amino terminus or the carboxyl terminus of GST which aids in purification of the recombinant protein. A Flag™ epitope tag allows detection of the recombinant protein with anti-Flag™ antibody.

Example 5

Production of recombinant PSCA in higher eukaryotic systems

A. Mammalian constructs:

To express recombinant PSCA in eukaryotic cells, the full or partial length PSCA cDNA sequences or variants thereof can be cloned into any one of a variety of expression vectors known in the art. One or more of the following regions of PSCA are expressed in these constructs, amino acids 1 to 123 or any 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21 , 22, 23, 24, 25, 26, 27, 28, 29, 30 or more consecutive amino acids from PSCA-v.l, PSCA variants or analogues thereof.

The constructs can be transfected into any one of a variety of mammalian cells such as 293 T cells. Transfected 293T cell lysates can be probed with the anti-PSCA polyclonal serum described here.

pcDNA4/HisMax constructs: To express PSCA in mammalian cells, a PSCA ORF or parts thereof of PSCA are cloned into pcDNA4/HisMax version A (Invitrogen, Carlsbad, CA). Protein expression is driven from the cytomegalovirus (CMV) promoter and the SP16 translational enhancer. The recombinant protein has XpressTM and six-histidine (6XHis) epitopes fused to the amino terminus. The pcDNA4/HisMax vector also contains the bovine growth hormone (BGH) polyadenylation signal and transcription termination sequence to enhance mRNA stability along with the SV40 origin for episomal replication and easy vector rescue in cell lines expressing the large T antigen. The zeocin resistance gene allows selection of mammalian cells expressing the protein and the ampicillin resistance gene and ColE1 origin allows selection and maintenance of the plasmid in E. coli.

pcDNA3. 1/ MvcHis constructs: To express PSCA in mammalian cells, a PSCA ORF or portions thereof of PSCA with a consensus Kozak translation initiation site was cloned into pcDNA3.1/MycHis version A (Invitrogen, Carlsbad, CA). Protein expression is driven from the cytomegalovirus (CMV) promoter. The recombinant proteins have the myc epitope and the 6XHis epitope fused to the carboxyl terminus. The pcDNA3.1/MycHis vector also contains bovine

the growth hormone (BGH) polyadenylation signal and transcription termination sequence to enhance mRNA stability, along with the SV40 origin for episomal replication and easy vector rescue in cell lines expressing the large T antigen. The neomycin resistance gene can be used as it allows selection of mammalian cells expressing the protein and the ampicillin resistance gene and ColE1 origin allows selection and maintenance of the plasmid in E. coli.

pcDNA3. 1/ CT- GFP- TOPO construct: To express PSCA in mammalian cells and to allow detection of the recombinant proteins using fluorescence, a PSCA ORF or parts thereof, with a consensus Kozak translation initiation site, is cloned into pcDNA3 .1/CT-GFP-TOPO (Invitrogen, CA). Protein expression is driven from the cytomegalovirus (CMV) promoter. The recombinant proteins have the green fluorescent protein (GFP) fused to the carboxyl terminus which facilitates non-invasive in vzvo detection and cell biology studies. The pcDNA3.1CT-GFP-TOPO vector also contains the bovine growth hormone (BGH) polyadenylation signal and transcription termination sequence to improve mRNA stability along with the SV40 origin for episomal replication and easy vector rescue in cell lines expressing the large T antigen. The neomycin resistance gene allows mammalian selection

cells expressing the protein and the ampicillin resistance gene and ColE1 origin allow selection and maintenance of the plasmid in E. coli. Additional constructs with an amino-terminal GFP fusion have been made in pcDNA3.1/NT-GFP-TOPO spanning the full length of a PSCA protein.

PAPtag: A PSCA ORF or parts thereof are cloned into pAPtag-5 (GenHunter Corp. Nashville, TN). This construct generates an alkaline phosphatase fusion on the carboxyl terminus of a PSCA protein while condensing the IgGic signal sequence to the amino terminus. Constructs have also been made in which alkaline phosphatase with an amino-terminal IgGic signal sequence is fused to the amino terminus of a PSCA protein. The resulting recombinant PSCA proteins are optimized for secretion into the medium of transfected mammalian cells and can be used to identify proteins such as ligands or receptors that interact with PSCA proteins. Protein expression is driven from the CMV promoter and the recombinant proteins also contain many and 6XHis epitopes condensed on the carboxyl terminus which facilitates detection and purification. The zeocin resistance gene present in the vector allows selection of mammalian cells expressing the recombinant protein and the ampicillin resistance gene allows selection of the plasmid in E. coli.

Ptag5: A PSCA ORF or parts thereof were cloned into pTag-5. This vector is similar to pAPtag but without the alkaline phosphatase fusion. This construct generates PSCA protein with an amino-terminal IgGic signal sequence and many and 6XHis epitopes on the carboxyl terminus that facilitate detection and affinity purification. The resulting recombinant PSCA protein is optimized for secretion into the medium of transfected mammalian cells and is used as an immunogen or ligand to identify proteins such as ligands or receptors that interact with the PSCA proteins. Protein expression is driven from the CMV promoter. The zeocin resistance gene present in the vector allows selection of mammalian cells expressing the protein and the ampicillin resistance gene allows selection of the plasmid in E. coli.

PsecFc: A PSCA ORF or parts thereof were cloned into psecFc. The psecFc vector was compared by cloning the human immunoglobulin G1 (IgG)-Fc (hinge, CH2, CH3 regions) into pSecTag2 (Invitrogen, California). This construct generates an IgG1-Fc fusion on the carboxyl terminus of the PSCA proteins, while condensing the IgGK signal sequence to the N-terminus. PSCA fusions using the murine IgG1-Fc region have also been used. The resulting recombinant PSCA proteins are optimized for secretion into the medium of transfected mammalian cells and can be used as immunogens or to identify proteins such as ligands or receptors that interact with PSCA protein. Protein expression is driven from the CMV promoter. The hygromycin resistance gene present in the vector allows selection of mammalian cells expressing the recombinant protein and the ampicillin resistance gene allows selection of the plasmid in E. coli.

Figure 8 shows expression and purification of PSCA.psecFc protein from 293T cells.

pSRα constructs: To generate mammalian cell lines constitutively expressing PSCA, the PSCA ORF or parts thereof of PSCA were cloned into pSRα constructs. Amphotropic and ecotropic retroviruses were generated by transfection of pSRα constructs, respectively, into the 293T-10A1 packaging line or cotransfection of pSRα and a helper plasmid (containing deleted packaging sequences) into the 293 cells. The retrovirus is used to infect a variety of mammalian cell lines, resulting in the integration of the cloned gene, PSCA, into the host cell lines. Protein expression is driven from a long terminal repeat (LTR) element. The neomycin resistance gene present in the vector allows selection of mammalian cells expressing the protein, and the ampicillin resistance gene and ColE1 origin allows selection and maintenance of the plasmid in E. coli. The retroviral vectors can then be used for infection and formation of different cell lines using, for example, PC3, NLH 3T3, TsuPrl, 293 or rat-1 cells.

Figure 6 shows expression of PSCA in recombinant murine, rat and human cell lines using the PSCA.pSRα construct. The indicated murine, rat and human cell lines were infected with retroviruses carrying the human PSCA cDNA and a neomycin resistance gene or with only the neomycin resistance gene. Stable recombinant cell lines were generated by G418 drug selection. PSCA expression was determined by FACS labeling with lG8 anti-PSCA MAb (5 µg/ml). Shown is the FACS profile of each cell line showing a fluorescent shift only in the PSCA-infected line indicating cell surface PSCA expression. These lines are useful in MAb development as immunogens, MAb screening reagents and in functional assays.

Additional pSRα constructs have been made that condense an epitope tag such as the FLAG™ tag to the carboxyl terminus of PSCA sequences to allow detection using anti-Flag antibodies. For example, the FLAG™ sequence 5'-gat tac aag gat gac gac gat aag 3' (SEQ ID NO:76) is added to the cloning primer at the 3' end of the ORF. Additional pSRα constructs have been made to produce both amino-terminal and carboxyl-terminal GFP and myc/6XHis fusion proteins of the full-length PSCA proteins.

Additional Viral Vectors: Additional constructs have been made for viral-mediated delivery and expression of PSCA. High virus titer leading to high level expression of PSCA is achieved in viral delivery systems such as adenoviral vectors and herpes amplicon vectors. PSCA coding sequences or fragments thereof are amplified by PCR and subcloned into the AdEasy shuttle vector (Stratagene). Recombination and viral packaging are performed according to the manufacturer's instructions to form adenoviral vectors. Alternatively, PSCA coding sequences or fragments thereof are cloned into the HSV-1 vector (Imgenex) to form herpes viral vectors. The viral vectors are then used for infection of different cell lines such as PC3, NLH 3T3, 293 or rat-1 cells.

Regulated expression systems: To control expression of PSCA in mammalian cells, the coding sequences of PSCA or parts thereof have been cloned into regulated mammalian expression systems such as the T-Rex system (Invitrogen), the GeneSwitch system (Invitrogen) and the tightly regulated Ecdyson system (Sratagen). These systems allow investigation of the temporal and concentration-dependent effects of recombinant PSCA. These vectors have then been used to control expression of PSCA in different cell lines such as PC3, NLH 3T3, 293 or rat-1 cells.

Baculovirus expression systems

To generate recombinant PSCA proteins in a baculovirus expression system, the PSCA-ORF or parts thereof are cloned into the baculovirus transfer vector pBlueBac 4.5 (Invitrogen), which provides a His-tag at the N-terminus. Specifically, pBlueBac-PSCA is cotransfected with helper plasmid pBac-N-Blue (Invitrogen) into SF9 (Spodoptera frugiperda) insect cells to generate recombinant baculoviruses (see Invitrogen's instruction manual for details). Baculovirus is then collected from cell supernatant and purified by plaque assay.

Recombinant PSCA protein is then produced by infection of HighFive insect cells (Invitrogen) with purified baculoviruses. Recombinant PSCA protein can be detected using anti-PSCA or anti-His-tag antibody. PSCA protein can be purified and used in various cell-based experiments or as an immunogen to generate polyclonal and monoclonal antibodies specific for PSCA.

Expression vectors for PSCA orthologs

Mouse and monkey orthologs of PSCA were cloned into pcDNA3.1/MycHis version A (Invitrogen, Carlsbad, CA). Protein expression is driven from the cytomegalovirus (CMV) promoter. The recombinant proteins have the myc epitope and the 6XHis epitope fused to the carboxyl terminus. These vectors allow expression of PSCA orthologs to assay cross-reactivity to monoclonal anti-human PSCA antibodies.

Mouse and monkey orthologs of PSCA were also cloned into pSRα constructs. The pSRα constructs allow the generation of mammalian cell lines constitutively expressing PSCA orthologs. Protein expression is driven from the cytomegalovirus (CMV) promoter. The recombinant proteins have the myc epitope and the 6XHis epitope fused to the carboxyl terminus. These vectors allow expression of PSCA orthologs to analyze cross-reactivity to monoclonal anti-human PSCA antibodies and to examine functional activity of PSCA orthologs. Amphotropic and ecotropic retroviruses were generated by transfection of pSRα constructs, respectively, into the 293T-10A1 packaging line or cotransfection of pSRα and a helper plasmid (containing deleted packaging sequences) into the 293 cells. The retrovirus is used to infect a variety of mammalian cell lines, resulting in the integration of the cloned gene, PSCA ortholog, into the host cell lines.

Figure 7 shows expression of mouse and simian PSCA.pcDNA3.1/MycHis after transfection into 293T cells. 293T cells were transfected with either mouse PSCA.pcDNA3.1/MycHis or simian

PSCA.pcDNA3.1/MycHis or pcDNA3.1/MycHis vector control. Forty hours later, cells were collected and analyzed by flow cytometry using anti-PSCA monoclonal antibodies.

Example 6

Antigenicity profiles and secondary structure

Amino acid profiles of PSCA variants 1, 3 and 4 were found available on the ProtScale web page on the World Wide Web at (.expasy.ch/cgi-bin/protscale.pl) on the ExPasy molecular biology server.

These profile: Hydrophilicity (Hopp T.P., Woods K.R., 1981. Proe. Nati. Acad. Sei. U.S.A. 78:3824-3828); hydrophaticity (Kyte J., Doolittle RF., 1982. J. Mol. Biol. 157:105-132); percentage available residues (Janin J., 1979 Nature 277:491-492); average flexibility (Bhaskaran R. and Ponnuswamy PK, 1988. Int. J. Pept. Protein Res. 32:242-255); beta-turn (Deleage, G., Roux B. 1987 Protein Engineering 1:289-294); and possibly others available in the field such as on the ProtScale website, were used to identify antigenic regions of each of the PSCA variant proteins. Each of the above amino acid profiles of PSCA variants was generated using the following ProtScale parameters for analysis: 1) A window size of 9; 2) 100% weight of the window edges compared to the window center; and 3) amino acid profile values normalized to lie between 0 and 1.

Hydrophilicity, hydrophaticity and percentage available residue profiles were used to determine stretches of hydrophilic amino acids (ie values greater than 0.5 on the hydrophilicity and percentage available residue profile and values less than 0.5 on the hydrophaticity profile). Such regions are likely to be exposed to the aqueous environment, to be present on the surface of the protein and thus accessible to immune recognition such as by antibodies.

Average flexibility and beta-turn profiles determine stretches of amino acids (ie values greater than 0.5 on the beta-turn profile and average flexibility profile) that are not forced into secondary structures such as beta-sheets and alpha-helices. Such regions are also more likely to be exposed on the protein and thus accessible to immune recognition such as by antibodies.

Antigenic sequences of the PSCA variant proteins indicated, e.g. by profilins described above, are used to prepare immunogens, either peptides or nucleic acids encoding them, to form therapeutic and diagnostic anti-PSCA antibodies. The immunogen can be any 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 30, 35, 40, 45, 50 or more than 50 consecutive amino acids or the corresponding nucleic acids encoding them, from the PSCA protein variants listed in Figure 1 from which the amino acid profiles can be inferred because the variant contains sequence identical to a shown variant. In particular, peptide immunogens according to the invention may comprise, a peptide region of at least 5 amino acids from Figure 1 in any whole number of increments comprising an amino acid position that has a value greater than 0.5 in the hydrophilicity profile; a peptide region of at least 5 amino acids from Figure 1 in any integer increment comprising an amino acid position having a value less than 0.5 in the hydrophilicity profile; a peptide region of at least 5 amino acids from Figure 1 in any integer increment comprising an amino acid position having a value greater than 0.5 in the percent available residue profiles; a peptide region of at least 5 amino acids of Figure 1 in any integer increment comprising an amino acid position having a value greater than 0.5 in the average flexibility profiles; and a peptide region of at least 5 amino acids from Figure 1 in any integer increment comprising an amino acid position having a value greater than 0.5 in the beta turn profile. Peptide immunogens according to the invention may also comprise nucleic acids which code for any of the foregoing.

All immunogens according to the invention, peptide or nucleic acid, can be concretized in human unit dose form or included in a composition comprising a pharmaceutical additive compatible with human physiology.

The secondary structure of PSCA protein variants 1, 3, 4 and 6, i.e. the predicted presence and location of alpha helices, extended strands and random coils, are predicted from the primary amino acid sequence using the HNN - Hierarchical Neural Network method (NPS@ : Network Protein Sequence Analysis TIBS 2000 March Vol. 25, No 3

[291]:147-150 Combet C, Blanchet C, Geourjon C. and Deléage G, http://pbil.ibcp.fr/cgi-bin/npsa_automat.pl?page=npsa_nn.html), available from the ExPasy molecular biology server located on the World Wide Web at (.expasy.ch/tools/). The analysis indicates that PSCA variant 1 is composed of 30.89% alpha helix, 21.95% extended strand and 47.15% random coil. PSCA protein variant 3 is composed of 14.89% alpha helix, 8.51% extended strand, and 76.60% random coil. PSCA protein variant 4 is composed of 9.52% alpha helix, 8.99% extended strand, and 81.48% random coil. PSCA protein variant 6 is composed of 24.56% alpha helix, 21.93% extended strand, and 53.51% random coil.

Analysis for the potential presence of transmembrane domains in the PSCA variant proteins was performed using a number of transmembrane prediction algorithms available from the ExPasy molecular biology server located on the World Wide Web at

(.expasy.ch/tools/).

Example 7

Formation of PSCA polyclonal antibodies

Polyclonal antibodies can be raised in a mammal, for example, by one or more injections of an immunizing agent and, if desired, an adjuvant. Typically, the immunizing agent and/or adjuvant will be injected into the mammal by multiple subcutaneous or intraperitoneal injections. In addition, to immunize with a full-length PSCA protein variant, computer algorithms are used in the design of immunogens which, based on amino acid sequence analysis, contain characteristics of being antigenic and available for recognition by the immune system of the immunized host (see the example labeled "Antigenicity Profiles and Secondary Structure"). Such regions would be predicted to be hydrophilic, flexible, in beta-turn conformations and exposed on the surface of the protein.

For example, recombinant bacterial fusion proteins or peptides containing hydrophilic, flexible beta-turn regions of PSCA protein variants are used as antigens to form

polyclonal antibodies in New Zealand White rabbits or monoclonal antibodies as described in the example labeled "Generation of PSCA Monoclonal Antibodies (MAbs)". For example, in PSCA variant 1, such regions include, but are not limited to, amino acids 28-56 and amino acids 66-94. For variant 3, such regions include, but are not limited to, amino acids 7-39 and amino acids 70-94. For variant 4, such regions include, but are not limited to, amino acids 6-18, amino acids 27-39, amino acids 103-133 and 177-189. For variant 6, such regions include, but are not limited to, amino acids 19-35 and amino acids 57-85. It is useful to conjugate the immunizing agent to a protein known to be immunogenic in the mammal being immunized. Examples of such immunogenic proteins include, but are not limited to, keyhole limpet hemocyanin (KLH), serum albumin, bovine thyroglobulin, and soybean trypsin inhibitor. In one embodiment, a peptide encoding amino acids 103-133 of PSCA variant 4 is conjugated to KLH and used to immunize a rabbit. Alternatively, the immunizing agent may comprise all or part of the PSCA variant proteins, analogues or fusion proteins thereof. For example, the PSCA variant amino acid sequences can be fused using recombinant DNA techniques to any one of a variety of fusion protein partners well known in the art, such as glutathione S-transferase (GST) and HIS-tagged fusion proteins. In a

embodiment, the PSCA variant 1 sequence, amino acids 18-98, fused to GST using recombinant techniques in the pGEX expression vector was expressed, purified and used to immunize both rabbits and mice to generate polyclonal and monoclonal antibodies, respectively. Such fusion proteins are purified from induced bacteria using the appropriate affinity matrix.

Other recombinant bacterial fusion proteins that can be used include maltose binding protein LacZ, thioredoxin, NusA or an immunoglobulin constant region (see the section entitled "Production of PSCA in Prokaryotic Systems" and Current Protocols In Molecular Biology, Volume 2, Unit 16, Frederick M. Ausubul et al .eds., 1995; Linsley, P.S., Brady, W., Urnes, M., Grosmaire, L., Damle, N. and Ledbetter, L.(1991) J.Exp. Med. 174, 561-566).

In addition to bacterially derived fusion proteins, mammalian expressed protein antigens are also used. These antigens are expressed from mammalian expression vectors such as the Tag5 and Fc fusion vectors (see the section entitled "Production of Recombinant PSCA in Eukaryotic Systems") and retain post-translational modifications such as glycosylations found in native protein. In one embodiment, the cDNA of PSCA variant 1, minus the N-terminal leader peptide and C-terminal GPI anchor, was cloned into the Tag5 mammalian secretion vector and expressed in 293T cells. The recombinant protein was purified by metal chelate chromatography from tissue culture supernatants of 293T cells stably expressing the recombinant vector. The purified Tag5 PSCA protein was then used as an immunogen.

During the immunization protocol, it is useful to mix or emulsify the antigen in adjuvants that enhance the immune response of the host animal. Examples of adjuvants include, but are not limited to, complete Freund's adjuvant (CFA) and MPL-TDM adjuvant (monophosphoryl lipid A, synthetic trehalose-dicorynomycolate).

In is typical protocol, rabbits are initially immunized subcutaneously with up to 200 µg, typically 100-200 µg, of fusion protein or peptide conjugated to KLH mixed in complete Freund's adjuvant (CFA). Rabbits are then injected subcutaneously every other week with up to 200 µg, typically 100-200 µg, of the immunogen in incomplete Freund's adjuvant (LFA).Test bleeds are taken approximately 7-10 days after each immunization and used to monitor the titer of the antiserum by ELISA.

To test the reactivity and specificity of immune serum, such as rabbit serum derived from immunization with a GST fusion of PSCA variant 3 or 4 protein, the respective full-length PSCA variant cDNA is cloned into the pCDNA 3.1 myc-his expression vector ( Invitrogen, see the example entitled "Production of Recombinant PSCA in Eukaryotic Systems"). After transfection of the constructs into 293T cells, cell lysates are probed with the anti-variant serum and with the anti-His antibody (Santa Cruz Biotechnologies, Santa Cruz, CA) to determine specific reactivity to denatured variant protein using the Western blot technique. In addition, the immune serum is tested by fluorescence microscopy, flow cytometry and immunoprecipitation against 293T and other recombinant PSCA variant expressing cells to determine specific recognition of native protein. Western blot, immunoprecipitation, fluorescent microscopy and flow cytometric techniques using cells endogenously expressing PSCA have also been performed to test reactivity and specificity.

Antiserum from rabbits immunized with PSCA variant fusion proteins, such as GST and MBP fusion proteins, is purified by depletion of antibodies reactive to the fusion partner sequence by passage over an affinity column containing the fusion partner either alone or in conjunction with an irrelevant fusion protein. For example, antiserum derived from a GST-PSCA variant 1 fusion protein is first purified by passage over a column of GST protein covalently coupled to AffiGel matrix (BioRad, Hercules, Calif). The antiserum is then affinity purified by passage over a column composed of an MBP-PSCA fusion protein covalently linked to Affigel matrix. The serum is then further purified by protein G affinity chromatography to isolate the IgG fraction. Sera from other His-tagged antigens and peptide-immunized rabbits as well as fusion partner-depleted sera are affinity purified by passage over a column matrix composed of the original protein immunogen or free peptide.

Example 8

Formation of PSCA monoclonal antibodies (MAbs)

In one embodiment, therapeutic monoclonal antibodies ("MAbs") to PSCA and PSCA variants include those that react with epitopes specific to each protein or specific to sequences common between the variants that would bind, internalize, destroy or modulate the biological function of PSCA or PSCA variants, such as those that would disrupt the interaction with ligands and binding partners. Immunogens for generation of such MAbs include those designed to encode or contain the extracellular domain or the entire PSCA protein sequence, regions predicted to contain functional motifs, and regions of the PSCA protein variants predicted to be antigenic from data analysis of the amino acid sequence. Immunogens include peptides, recombinant bacterial proteins such as GST-PSCA fusion proteins (Figure 8) and His-tagged PSCA-pET vector protein (Figure 6) and mammalian expressed purified His-tagged proteins (Figure 7) and human and murine IgG-Fc -fusion proteins. In addition, cells are engineered via retroviral transduction to express high levels of PSCA variant 1, such as RAT1-PSCA. 293T-PSCA, 3T3-PSCA or 300.19-PSCA, used to immunize mice (Figure 5).

To generate monoclonal antibodies to PSCA, mice were first immunized in the foot pad (FP) with typically 5-50 µg of immunogenic protein or between IO<6> and IO<7> PSCA-expressing cells mixed in a suitable adjuvant. Examples of suitable adjuvants for FP immunizations are TiterMax (Sigma) for the initial FP injection followed by alum gel with Immuneasy (Qiagen). After the initial injection, mice were then immunized twice weekly until the time they were sacrificed and B cells obtained from the lymph node used for fusion.

During the course of the immunization protocol, test bleeds were taken to monitor the titer and specificity of the immune response. In most cases, once appropriate reactivity and specificity were achieved, as determined by ELISA, Western blotting, immunoprecipitation, fluorescence microscopy, or flow cytometric assays, fusion and hybridoma transformation were then performed using electrocell fusion (BTX, ECM2000 ).

In one embodiment, the invention provides monoclonal antibodies designated Ha 1-1.16, Hal-5.99, Hal-4.117, Hal-4.120, Hal-4.121 and Hal-4.37. The antibodies were identified and shown to react and bind with cell surface or immobilized PSCA.

MAbs against PSCA were generated using xenomus technology<®>where the murine heavy and kappa light chain loci are inactivated and a majority of the human heavy and kappa light chain immunoglobulin loci are inserted: Ha 1-1.16 was generated after immunizing human gamma 1 -producing xenomus (13-fold with PSCA-GST); Hal-5.99 was generated after immunizing human gamma 2-producing xenomice 6 times with Rat1-PSCA cells followed by two injections of PSCA-tag5; Hal-4.117, HA1-4.37, Hal-4.120 and Hal-4.121 were generated after immunizing human gamma 1-producing xenomouse) 6 times with Rat1-PSCA cells followed by 4 injections of PSCA-tag5. Anti-PSCA MAbene, Hal-1.16, Hal-5.99, Hal-4.117, Hal-4.120 and Hal-4.121 bind endogenous cell surface PSCA expressed in prostate cancer xenograft cells.

The antibodies designated Hal-5.99, Hal-4.117, Hal-4.120, Hal-4.37, Hal-1.16 and Hal-4.121 were sent (via Federal Express) to the American Type Culture Collection (ATCC), P.O. Box 1549, Manassas, VA 20108 04-May-2004 and designated accession numbers PTA-6703 and PTA-6699 and PTA-6700 and PTA-6702 and PTA-6698 and PTA-6701, respectively.

DNA coding sequences for anti-PSCA-MAbs Hal-1.16, Hal-5.99, Hal-4.117, Hal-4.120, Hal-4.121 and Hal-4.37 were determined after isolation of mRNA from the respective hybridoma cells with Trizol reagent (Life Technologies , Gibco BRL). Total RNA was purified and quantified. First strand cDNAs were generated from total RNA with oligo (dT)12 18 priming using the Gibco BRL Superscript Preamplification system. First strand cDNA was amplified using human immunoglobulin variable heavy chain primers and human immunoglobulin variable light chain primers. PCR products were cloned into the pCRScript vector (Stratagene, La Jolla). Many clones were sequenced and the variable heavy and light chain regions determined. The nucleic acid and amino acid sequences of the variable heavy and light chain regions are listed in Figure 2 and Figure 3. Alignment of PSCA antibodies to germlin V-D-J sequences is shown in Figure 4A - Figure 4M.

Example 9

Screening and identification of PSCA antibodies

Antibodies generated using the methods set forth in the example entitled "Generation of PSCA Monoclonal Antibodies (MAbs)" were screened and identified using a combination of assays including ELISA, FACS, epitope clustering and affinity for PSCA expressed on the cell surface.

A. PSCA human MAb screening by FACS.

Primary hybridoma screening for MAbs against PSCA was performed by FACS. The protocol was as follows: 50 µl/well of hybridoma supernatant (exact) or purified antibodies (in serial dilutions) were added to 96-well FACS plates and mixed with PSCA-expressing cells (endogenous or recombinant, 50,000 cells/well). The mixture was incubated at 4°C for two hours. At the end of incubation, the cells were washed with FACS buffer and incubated with 100ul of detection antibody (anti-hlgG-PE) for 45 minutes at 4°C. At the end of incubation, cells were washed with FACS buffer, fixed with formaldehyde and analyzed using FACScan. Data were analyzed using CellQuest Pro software. Filled histograms indicate data from negative control cells and open histograms represent data from PSCA-positive cells (Figure 9).

Positive hybridomas identified from primary screens were transferred to 24-well plates and supernatants collected for confirmatory screens. Confirmatory screens included FACS analysis on B300.19-PSCA/300.19-neo, Ratl-PSCA/Ratl-neo, PC3-PSCA/neo, SW780 (bladder cancer cell line), LAPC9AI (prostate cancer cell line), HP AC (pancreatic cancer cell line) and ELIS A -experiments using Tag5-PSCA, GST-PSCA, GST-PSCA N-term, Med. C-term and pET-PSCA.

PSCA human MAb relative affinity assay

The hybridoma supernatants were tested to determine their relative binding affinity to cell surface PSCA. Hybridoma supernatants were serially diluted in FACS buffer (FB), from ug/ml to sub-ng/ml; and evaluated in a FACS binding assay using LAPC9AI cells. High affinity antibodies provided high MFI values. MFI values for each point were obtained using CellQuest Pro software and used for affinity calculation using Graphpad Prisme software (Table VU and Table VIII): sigmoidal dose response (variable slope) equation. The results of the relative affinity assay are shown in Figure 10.

Epitope grouping

PSCA antibodies were grouped according to epitope by judging their binding pattern on LAPC9AI cells. Briefly, a small amount of each of the antibodies was biotinylated; then each of the biotinylated antibodies was incubated with LAPC9AI in the presence of excess (100x) of non-biotinylated antibodies at 4°C for 1 hour during the incubation. In general, an excess amount of antibodies will compete with biotinylated antibodies if they bind to the same epitope. At the end of incubation, cells were washed and incubated with Streptavidin-PE for 45 min at 4°C. After washing away the unbound streptavidin-PE, the cells were analyzed using FACS. MFI regulations were used for data analysis (table VU). As shown in Table XI, cells illuminated in yellow indicate self-competition (100% competition), the MFI in these cells is the background control for each biotinylated antibody. Cells with no color indicate that the two antibodies compete with each other (low MFI), high MFI (highlighted in blue) indicates that the two antibodies bind to two distinct epitopes. The antibodies that have the same binding pattern bind to the same epitope among the antibodies. There are 6 epitope groups within the tested antibodies. Table XE shows that PSCA 4.121 binds to its unique epitope.

Example 10

Characterization and expression of PSCA antibodies

A. Cross-reactivity with monkey PSCA and mouse PSCA

MAbs were screened and characterized for their ability to react with PSCA of mouse and simian origin. This property is useful for understanding the consequences of MAb engagement of PSCA on cells and tissues using mouse and simian animal models. The cynomolgus monkey and mouse PSCA genes were cloned, expressed in a retrovirus, and transiently infected into 293-T cells.

were incubated with the respective antibodies using the following

the protocol. Test antibodies were incubated with 293-T cells expressing either cynomolgus monkey or murine PSCA or 293T cells expressing the neogene only as a negative control. Specific recognition was determined using anti-hlgG-PE secondary detection antibody. A representative histogram showing specified cross-reactivity is presented in Figure 11. The summary presented in Table X shows that all but one of the anti-human PSCA antibodies cross-react with monkey PSCA and only one anti-human PSCA antibody cross-reacts with mouse PSCA.

Affinity determination by FACS

A panel of seven (7) anti-human PSCA antibodies were tested for their binding affinity to PSCA on SW780 cells, a human bladder cancer cell line that expresses high levels of PSCA. 23 serial 1:2 dilutions of purified antibodies were incubated with SW780 cells (50,000 cells per well) overnight at 4°C with a final concentration of 167nM to 0.01pM. At the end of the incubation, cells were washed and incubated with anti-hlgG-PE detection antibody for 45 min at 4°C. After washing the unbound detection antibodies, the cells were analyzed by FACS; MFI values for each point were obtained using CellQuest Pro software and were used for affinity calculation using Graphpad Prisme software: sigmoidal dose response (variable slope) equation (Table VU and Table V<J>JI). Affinity values for the seven (7) antibodies are given in Table IX.

Example 11

PS C A- antibody internalization

Internalization of Hal-4,121 was studied using PC3-PSCA cells. Briefly, Hal-4,121 was incubated with cells at 4°C for 90 min to allow binding of the antibodies to the cell surface. The cells were then split into two groups and incubation continued at either 37°C to allow antibody internalization or at 4°C as a control (no internalization). An acid wash after 37°C/4°C incubation was used to remove PSCA 4.121 bound on the cell surface. Subsequent permeablization allowed detection of antibodies bound to internalized PSCA. After incubation with secondary detection antibodies, cells were analyzed using FACS or observed under a fluorescence microscope. Approximately 30% of Hal-4,121 internalized after incubation at 37°C for two hours (Figure 12).

Example 12

Antibody-mediated secondary killing

Antibodies to PSCA mediate saporin-dependent killing in PSCA-expressing cells. B300.19 - PSCA-expressing cells (750 cells/well) were seeded in a 96-well plate on day 1. The following day, an equal volume of medium containing 2X concentration of the indicated primary antibody together with a 2-fold excess of antihuman (Hum-Zap) or antigoat (Goat-Zap) polyclonal antibody conjugated with saporin toxin (Advanced Targeting Systems, San Diego, CA) was added to each well. The cells were allowed to incubate for 5 days at 37 degrees C. At the end of the incubation period, MTS (Promega) was added to each well and incubation continued for an additional 4 hours. The OD at 450 nM was determined. The results in Figures 13(A) show that PSCA antibodies HA1-4.121 and HA1-4.117 mediated saporin-dependent cytotoxicity in B300.19-PSCA cells while a control antibody, nonspecific human IgG1, had no effect. The results in Figure 13(B) show that addition of a secondary saporin-conjugated antibody, which does not recognize human Fc, fails to mediate cytotoxicity (Figure 13(A) and Figure 13(B)). These results indicate that drugs or cytotoxic proteins can be selectively delivered to PSCA-expressing cells using an appropriate anti-PSCA MAb.

Example 13

Antibody immune-mediated cytotoxicity

PSCA antibodies were evaluated to determine their ability to mediate immune-dependent cytotoxicity. PSCA antibodies (0-50 µg/ml) were diluted with RHB buffer (RPMI1640, Gibco Life Technologies, 20 mM HEPES). B300.19-PSCA-expressing cells were washed in RHB buffer and resuspended at a density of 10<6> cells/ml. In a typical experiment, 50 µl of PSCA antibody, 50 µl of diluted complement rabbit serum (Cedarlane, Ontario, Can) and 50 µl of a cell suspension were added together in a flat-bottom tissue culture 96-well plate. The mixture was incubated for 2 hours at 37°C in a 5% CO 2 incubator to facilitate complement-mediated cell lysis. Then 50 µl of Alamar Blue (Biosource Intl. Camarillo, CA) was added to each well and incubation continued for an additional 4-5 hours at 37°C. The fluorescence in each well was read using a 96-well fluorometer with excitation at 530 nm and emission at 590 nm. The results show that PSCA antibodies having an Ig1 (HA1-4.121) or an IgG2 isotype (HA1-5.99.1) but not an IgG4 isotype (HAI-6.46) were able to mediate complement-dependent lysis (Figure 14 ).

ADCC (antibody-dependent cellular cytotoxicity) is an immune-mediated lytic attack on cells bound by an antibody targeted to a specific cell surface antigen. In this case, it is the PSCA. Immune cells recognize the Fc part of the antibody through binding to Fcy receptors on the surface of leukocyte, monocyte and NK cells which triggers a lytic attack resulting in cell death. The ability of PSCA antibodies to mediate this reaction can be evaluated by labeling tumor cells in vitro with<51>chromium, europium, or a fluorescent molecule and incubating them in the presence of human PSCA-MAbs together with peripheral blood mononuclear cells. Specific lysis of the tumor cells can be determined by measuring the percentage lysis of the targeted tumor cells. Common endpoints that are determined include the release of radioactivity, europium, or fluorescent dye from the dead cells using an appropriate detection method. Alternatively, the release of an intracellular enzyme such as lactate dehydrogenase (LDH) can be measured.

Example 14

Formation of F(Ab') 2 fragments

Generation of F(Ab')2 fragments of MAbs is useful for investigating the effects of MAb molecules that retain their bivalent antigen binding site but lack the immune effector Fc domains in in vitro and in vivo therapeutic models. 20 mg of MAb-Hal-4,121 in 20 mM sodium acetate buffer pH 4.5 was incubated with and without immobilized pepsin (Pierce. Rockford IL) for the indicated times. Intact MAb and digested Fc fragments were removed by protein A chromatography. Shown in Figure 15 is an SDS-PAGE Coomasie-stained gel of intact undigested unreduced MAb, unreduced aliquots of digested material taken from the indicated times, and a reduced sample of the final digested F(ab') 2 product. This reagent can be used to treat animals bearing PSCA-expressing tumors. The antitumor activity observed with this antibody fragment may separate intrinsic biological activity from activity mediated by immune-dependent mechanisms.

Example 15

Expression of human antibodies using recombinant DNA methods To express anti-PSCA-MAbs recombinantly in transfected cells, anti-PSCA variable heavy and light chain sequences were cloned upstream of the human heavy chain IgG1 and light chain IgK constant regions, respectively. The complete anti-PSCA human heavy chain and light chain cassettes were cloned downstream of the CMV promoter/enhancer in a cloning vector. A polyadenylation site was included downstream of the MAb coding sequence. The recombinant anti-PSCA MAb expressing constructs were transfected into 293T, Cos and CHO cells. The HAl-4,121 antibody secreted from recombinant 293-T cells was evaluated for binding to cell surface PSCA in Figure Pia-3 A and compared to the same antibody produced from the original hybridoma (Figure 16).

Example 16

HLA class I and class II binding assays

HLA class I and class II binding assays using purified HLA molecules are performed according to described protocols (eg, PCT publications WO 94/20127 and WO 94/03205; Sidney et al., Current Protocols in lmmunology 18.3 .1 (1998); Sidney, et al., J. Immunol. 154:247

(1995); Sette, etal, Mol. Immunol. 31:813 (1994)). Briefly, purified MHC molecules (5 to 500 nM) are incubated with various unlabeled peptide inhibitors and 1-10 nM<125>I-radiolabeled probe peptides as described. After incubation, MHC-peptide complexes are separated from free peptide by gel filtration and the fraction of peptide bound is determined. In preliminary experiments, each MHC preparation is typically titrated in the presence of fixed amounts of radiolabeled peptides to determine the concentration of HLA molecules necessary to bind 10-20% of the total radioactivity. All subsequent inhibition and direct binding experiments are performed using these HLA concentrations.

Since under these conditions [label]<[HLA] and ICso>[HLA] the measured ICso values are reasonable approximations of the true KD values. Peptide inhibitors are typically tested at concentrations ranging from 120 µg/ml to 1.2 ng/ml and are tested in two to four completely independent experiments. To allow comparison of the data obtained in different experiments, a relative binding figure is calculated for each peptide by dividing the IC 50 of a positive control for inhibition by the IC 50 of each peptide tested (typically unlabeled versions of the radiolabeled probe peptide). For database purposes and intertrial comparisons, relative binding values are collected. These values can then be converted back to IC50 nM values by dividing the IC50 nM of the positive controls for inhibition by the relative binding of the peptide of interest. This method of data collection is accurate and consistent for comparison of peptides tested on different days or with different lots of purified MHC.

Binding experiments as described above can be used to analyze HLA-supermotif and/or HLA-motif-bearing peptides (see Table IV).

Example 17

Construction of "minigene" multi-epitope DNA plasmids

This example discusses the construction of a minigene expression plasmid. Minigene plasmids may of course contain different configurations of B cell, CTL and/or HTL epitopes or epitope analogs as described herein.

A minigene expression plasmid typically comprises multiple CTL and HTL peptide epitopes. In the present example, HLA-A2, -A3, -B7 supermotif bearing peptide epitopes and HLA-A1 and -A24 motif bearing peptide epitopes are used together with DR supermotif bearing epitopes and/or DR3 epitopes. The HLA class I supermotif or motif-bearing peptide epitope-derived PSCA is chosen so that multiple supermotifs/motifs are represented to ensure broad population coverage. Similarly, HLA class II epitopes are selected from PSCA to provide broad population coverage, i.e. both HLA-DR-1-4-7 supermotif bearing epitopes and HLA-DR-3 motif bearing epitopes are selected for inclusion in the minigene construct. The selected CTL and HTL epitopes are then inserted into a minigene for expression in an expression vector.

Such a construct may additionally comprise sequences that direct the HTL epitopes to the endoplasmic reticulum. For example, the Ii protein can be fused to one or more HTL epitopes as described in the site, where the CLIP sequence of the Ii protein is removed and replaced with an HLA class U epitope sequence so that the HLA class LT epitope is directed to it the endoplasmic reticulum, where the epitope binds to an HLA class II molecule.

This example illustrates the procedures to be used for the construction of a minigene-carrying expression plasmid. Other expression vectors that can be used for minigene compositions are available and known to those skilled in the art.

The minigene DNA plasmid from this example contains a consensus Kozak sequence and a consensus murine kappa Ig light chain signal sequence followed by CTL and/or HTL epitopes selected according to principles described herein. The sequence encodes an open reading frame fused to the Myc and His antibody epitope encoded by the pCDNA 3.1-Myc-His vector.

Overlapping oligonucleotides that can, for example, average approx. 70 nucleotides in length with 15 nucleotide overlaps, be synthesized and HPLC-purified. The oligonucleotides encode the selected peptide epitopes as well as appropriate linker nucleotides, Kozak sequence and signal sequence. The final multiepitopemin gene is assembled by extending the overlapping oligonucleotides in three sets of reactions using PCR. A Perkin/Elmer 9600 PCR machine is used and a total of 30 cycles are performed using the following conditions: 95°C for 15 sec, hybridization temperature (5° below the lowest calculated Tm of each primer pair) for 30 sec and 72°C for 1 min.

For example, a minigene is produced as follows. For a first PCR reaction, 5 µg of each of two oligonucleotides are hybridized and extended: In an example using eight oligonucleotides, i.e. four pairs of primers, oligonucleotides 1+2, 3+4, 5+6 and 7+8 are combined in 100 µl reactions containing Pfu polymerase buffer (lx= 10 mM KCl, 10 mM (NH 4 ) 2 SO 4 , 20 mM Tris chloride, pH 8.75, 2 mM MgSO 4 , 0.1% Triton X-100, 100 µg /ml BSA), 0.25 mM of each dNTP and 2.5 U of Pfu polymerase. The full-length dimer products are gel purified and two reactions containing the product of 1+2 and 3+4 and the product of 5+6 and 7+8 are mixed, hybridized and extended for 10 cycles. Half of the two reactions are then mixed and 5 cycles of hybridization and extension performed before flanking primers are added to amplify the full-length product. The full-length product is gel purified and cloned into pCR-blunt (Invitrogen) and individual clones are screened by sequencing.

Example 18

Plasmid construction and the degree to which it elicits immunogenicity.

The degree to which a plasmid construct, for example a plasmid constructed according to the previous example, is capable of eliciting immunogenicity is confirmed in vitro by determining epitope presentation of the APC after transduction or transfection of the APC with an epitope-expressing nucleic acid construct. Such an examination determines "antigenicity" and allows the use of human APC. The assay determines the ability of the epitope to be presented by the APC in a context recognized by a T cell by quantifying the density of epitope-HLA class I complexes on the cell surface. Quantification can be performed by direct measurement of the amount of peptide eluted from the APC (see, e.g., Sijts et al., J. Immunol. 156:683-692, 1996; Demotz et al., Nature 342:682-684, 1989); or the number of peptide-HLA class I complexes can be calculated by measuring the amount of lysis or lymphokine release induced by diseased or transfected target cells and then determining the peptide concentration required to achieve equivalent levels of lysis or lymphokine release (see, e.g., Kageyama et al., J. Immunol. 154:567-576, 1995).

Alternatively, immunogenicity is confirmed through in vzvo injections in mice and subsequent in v/7/*o assessment of CTL and HTL activity, which is analyzed using cytotoxicity and proliferation assays, respectively, as detailed e.g. in Alexander et al., Immunity 1:751-761, 1994.

For example, to confirm the capacity of a DNA minigene construct containing at least one HLA-A2 supermotif peptide to elicit CTLs in vivo, for example, HLA-A2.1/K<b> transgenic mice are immunized intramuscularly with 100 µg of naked cDNA. As a means of comparing the level of CTLs induced by cDNA immunization, a control group of animals is also immunized with a relevant peptide composition comprising multiple epitopes synthesized as a single polypeptide as if they would be encoded by the minigene.

Spleen cells from immunized animals are stimulated twice with each of the respective compositions (peptide epitopes encoded in the minigene or the polyepitopic peptide), then examined for peptide-specific cytotoxic activity in a<51>Cr release assay. The results indicate the magnitude of the CTL response directed against the A2-restricted epitope, thus indicating the in vivo immunogenicity of the minigene vaccine and the polyepitopic vaccine.

It has therefore been found that the minigene elicits immune responses directed against the HLA-A2 supermotif peptide epitopes that make up the polyepitopic peptide vaccine. A similar analysis has also been performed using other HLA-A3 and HLA-B7 transgenic mouse models to assess CTL induction at HLA-A3 and HLA-B7 motif or supermotif epitopes, whereby the minigene has also been found to elicit appropriate immune responses directed against the provided epitopes.

To confirm the capacity of a class II epitope-encoding minigene to elicit HTLs in vivo, DR transgenic mice, or for those epitopes that cross-react with the appropriate mouse MHC molecule I-Ab<->restricted mice, for example, are immunized intramuscularly with 100 µg of plasmid DNA. As a means of comparing the level of HTLs induced by DNA immunization, a group of control animals is also immunized with a relevant peptide composition emulsified in complete Freund's adjuvant. CD4+ T cells, i.e. HTLs, are purified from spleen cells of immunized animals and stimulated with each of the respective compositions (peptides encoded in the minigene). The HTL response is measured using a <3>H-thymidine insertion proliferation assay (see, e.g., Alexander et al. Immunity 1:751-761, 1994). The results indicate the magnitude of the HTL response thus demonstrating the in vivo immunogenicity of the minigene.

DNA minigenes, constructed as described in the previous example, can also be confirmed as a vaccine in combination with a research agent using a primary booster protocol. The research agent may consist of recombinant protein (eg Barnett et al, Aids Res. and Human Retroviruses 14, Supplement 3:S299-S309, 1998) or recombinant vaccinia, which for example expresses a minigene or DNA encoding the complete protein of interest (see e.g. Hanke et al., Vaccine 16:439-445, 1998; Sedegah et al., Proe. Nati. Acad. Sei USA 95:7648-53, 1998; Hanke and McMichael, Immunol. Letters 66:177-181, 1999; and Robinson et al., Nature Med. 5:526-34, 1999).

For example, the effectiveness of the DNA minigene used in a primary enhancer protocol has initially been evaluated in transgenic mice. In this example, A2 is. l/Kb transgenic mice immunized IM with 100 µg of a DNA minigene encoding the immunogenic peptides comprising at least one HLA-A2 supermotif-bearing peptide. After an incubation period (ranging from 3-9 weeks), the mouse is boosted IP with 10<7>pfu/mouse of a recombinant vaccinia virus expressing the same sequence encoded by the DNA minigene. Control mice are immunized with 100 µg of DNA or recombinant vaccinia without the minigene sequence or with DNA encoding the minigene but without the vaccinia boost. After a further incubation period of two weeks, spleen cells from the mouse are immediately examined for peptide-specific activity in an ELISPOT experiment. In addition, spleen cells are stimulated in vitro with the A2-restricted peptide epitopes encoded in the minigene and recombinant vaccinia, then assayed for peptide-specific activity in an alpha, beta, and/or gamma LFN

ELISA.

It has been found that the minigene used in a primary enhancer protocol elicits greater immune responses against the HLA-A2 supermotif peptides than with DNA alone. Such analysis can also be performed using HLA-A1 1 or HLA-B7 transgenic mouse models to assess CTL induction at HLA-A3 or HLA-B7 motif or supermotif epitopes. The use of primary booster protocols in humans is described below in the example entitled "Induction of CTL Responses Using a Primary Booster Protocol."

Example 19

Polyepitopic vaccine compositions from multiple antigens The PSCA peptide epitopes of the present invention are used together with epitopes from other target tumor-associated antigens, to create a vaccine composition that is useful for the prevention or treatment of cancer expressing PSCA and such other antigens. For example, a vaccine composition can be given as a single polypeptide that introduces multiple epitopes from PSCA as well as tumor-associated antigens that are often expressed with a target cancer associated with PSCA expression or can be administered as a composition comprising a cocktail of one or more distinct epitopes. Alternatively, the vaccine can be administered as a minigene construct or as dendritic cells loaded with the peptide epitopes in vitro.

Example 20

Application of peptides to assess an immune response

Peptides according to the invention can be used to analyze an immune response for the presence of specific antibodies, CTL or HTL directed against PSCA. Such an analysis can be performed in a manner described by Ogg et al., Science 279:2103-2106, 1998. In this example, peptides according to the invention are used as a reagent for diagnostic or prognostic purposes, not as an immunogen.

In this example, highly sensitive human leukocyte antigen tetrameric complexes ("tetramers") are used for a cross-sectional analysis of, for example, PSCA-HLA-A<*>0201-specific CTL frequencies from HLA A<*>0201-positive individuals on different stage of disease or after immunization comprising a PSCA peptide containing an A<*>0201 motif. Tetrameric complexes are synthesized as described (Musey et al, N. Engl. J. Med. 337:1267, 1997). Briefly, purified HLA heavy chain (A<*>0201 in this example) and P2 microglobulin are synthesized using a prokaryotic expression system. The heavy chain is modified by deletion of the transmembrane cytosolic tail and COOH-terminal addition of a sequence containing a BirA enzymatic biotinylation site. The heavy chain, P2-microglobulin, and peptide are refolded upon dilution. The 45-kD refolded product is isolated by solid protein-liquid chromatography and then biotinylated with BirA in the presence of biotin (Sigma, St. Louis, Missouri), adenosine 5'-triphosphate, and magnesium. Streptavidin-phycoerythrin conjugate is added in a 1:4 molar ratio and the tetrameric product is concentrated to 1 mg/ml. The resulting product is referred to as tetrameric phycoerythrin.

For the analysis of patient blood samples, approximately one million PBMCs are centrifuged at 300 g for 5 minutes and resuspended in 50 µl of cold phosphate-buffered saline. Tri-color analysis is performed with the tetramer phycoerythrin, together with anti-CD8 tricolor and anti-CD38. The PBMCs are incubated with tetramers and antibodies on ice for 30 to 60 min and then washed twice before formaldehyde fixation. Ports are used to contain >99.98% of control samples. Controls for the tetramers include both A<*>0201-negative individuals and A<*>0201-positive non-disease donors. The percentage of cells labeled with the tetramer is then determined by flow cytometry. The results indicate the number of cells in the PBMC sample that contain epitope-restricted CTLs, thereby readily indicating the extent of immune response to the PSCA epitope and thus the status of exposure to PSCA or exposure to a vaccine that elicits a protective or therapeutic response.

Example 21

Induction of immune responses using a primary booster protocol

A primary booster protocol similar in its underlying principle to that used to confirm the efficacy of a DNA vaccine in transgenic mice, as described above in the example entitled "The Plasmid Construction and the Degree of Inducing Immunogenicity," may also be used for administering the vaccine to human beings. Such a vaccine regimen may comprise an initial administration of, for example, naked DNA followed by a boost using recombinant virus encoding the vaccine or recombinant protein/polypeptide or a peptide mixture administered in an adjuvant.

For example, the initial immunization can be carried out using an expression vector, such as that constructed in the example entitled "Construction of "minigene" multi-epitope DNA plasmids" in the form of naked nucleic acid administered LM (or SC or LD) in the amounts of 0.5-5 mg at multiple sites. The nucleic acid (0.1 to 1000 (ag) can also be administered using a gene gun. After an incubation period of 3-4 weeks, a booster dose is then administered. The booster can be recombinant fowlpox virus administered at a dose of 5-10<7>to 5x10<9>pfu. An alternative recombinant virus, such as an MVA, canarypox, adenovirus or adeno-associated virus, can also be used for the enhancer or the polyepitopic protein or a mixture of the peptides can be administered. For evaluation of vaccine efficacy, patient blood samples are obtained before immunization so as well as at intervals after administration of the initial vaccine and booster doses of the vaccine. Peripheral blood mononuclear cells are isolated from fresh heparinized blood by Ficoll-Hypaque density gradient centrifugation, aliquoted into freezing media, and stored frozen. Samples are assayed for CTL and HTL activity.

Analysis of the results indicates that a magnitude of response sufficient to achieve a therapeutic or protective immunity against PSCA is generated.

Example 22

Complementary polynucleotides

Sequences complementary to the PSCA coding sequences (Figure 1 or Figure 3), or any part thereof, are used to detect, decrease or inhibit expression of naturally occurring PSCA. Although the use of oligonucleotides comprising from approx. 15 to 30 base pairs are described, essentially the same procedure is used with smaller or with larger sequence fragments. Suitable oligonucleotides are designed using e.g. OLIGO 4.06 software (National Biosciences) and the coding sequence of PSCA. To inhibit transcription, a complementary oligonucleotide is designed from the most unique 5' sequence and used to prevent promoter binding to the coding sequence. To inhibit translation, a complementary oligonucleotide is designed to prevent ribosomal binding to a PSCA-encoding transcript.

Example 23

Purification of naturally occurring or recombinant PSCA using PSCA-specific antibodies

Naturally occurring or recombinant PSCA are mainly purified by immunoaffinity chromatography using antibodies specific for PSCA. An immunoaffinity column is constructed by covalently coupling anti-PSCA antibody to an activated chromatographic resin, such as CNBr-activated SEPHAROSE (Amersham Pharmacia Biotech). After coupling, the resin is blocked and washed according to the manufacturer's instructions.

Media containing PSCA is passed over the immunoaffinity column and the column is washed under conditions that allow the preferential absorbance of PSCA (eg, high ionic strength buffers in the presence of detergent). The column is eluted under conditions that disrupt antibody/PSCA binding (eg, a buffer of pH 2 to pH 3 or a high concentration of a chaotrope, such as urea or thiocyanation) and GCR.P is collected.

Example 24

Identification of molecules that interact with PSCA

PSCA, or biologically active fragments thereof, are labeled with 121 1 Bolton-Hunter reagent. (see, e.g., Bolton et al. (1973) Biochem. J. 133:529.) Candidate molecules previously analyzed in the wells of a multiwell plate are incubated with the labeled PSCA, washed, and any wells with labeled PSCA complex are probed. . Data obtained using different concentrations of PSCA are used to calculate values for the number, affinity and association of PSCA with the candidate molecules.

Example 25

In vivo assay for PSCA tumor growth promotion

The effect of the PSCA protein on tumor cell growth has been evaluated in vivo by assessing tumor development and growth of cells expressing or lacking PSCA. For example, SCID mice are injected subcutaneously on each side with 1 x 10<6> of either 3T3 or prostate cancer cell lines (eg, PC3 cells) containing tkNeo-empty vector or PSCA. At least two strategies can be used: (1) Constitutive PSCA expression under the control of a promoter such as a constitutive promoter obtained from the genomes of viruses such as polyomavirus, fowlpoxvirus (UK 2,211,504 published 5 July 1989), adenovirus (such as adenovirus 2 ), bovine papillomavirus, avian sarcoma virus, cytomegalovirus, a retrovirus, hepatitis B virus and simian virus 40 (SV40) or from heterologous mammalian promoters, e.g. actin promoters or an immunoglobulin promoter, provided such promoters are compatible with the host cell systems and (2) Regulated expression under the control of an inducible vector system, such as ecdysone, tetracycline, etc, provided such promoters are compatible with the host cell systems. Tumor volume is then monitored by caliper measurement at the appearance of overt tumors and followed over time to determine whether PSCA-expressing cells grow at a faster rate and whether tumors produced by PSCA-expressing cells demonstrate characteristics of altered aggressiveness (eg, enhanced metastasis, vascularisation, reduced susceptibility to chemotherapeutic drugs).

In addition, mice can be implanted with 1 x 10<5> of the same cells orthotopically to determine whether PSCA has an effect on local growth in the prostate and whether PSCA affects the ability of the cells to metastasize, specifically to lymph nodes and bone (Miki T et al, Oncol Res. 2001;12:209; Fu X et al, Int J Cancer. 1991, 49:938). The effect of PSCA on bone tumor formation and growth can be assessed by injecting prostate tumor cells intratibially.

The experiment is also applicable to determine the PSCA-inhibitory effect of candidate therapeutic compounds, such as, for example, PSCA intrabodies, PSCA antisense molecules and ribozymes.

Example 26

PSCA monoclonal antibody-mediated inhibition of tumors in vivo

The significant expression of PSCA on the cell surface of tumor tissue, together with its restrictive expression in normal tissue makes PSCA a good target for antibody therapy. Similarly, PSCA is a target for T-cell-based immunotherapy. Thus, the therapeutic efficacy of anti-PSCA-MAbs in human prostate cancer xenograft mouse models and human pancreatic cancer xenograft mouse models has been evaluated using recombinant cell lines such as PC3-PSCA and 3T3-PSCA (see, e.g., Kaighn, M.E. , et al, Invest Uroi, 1979. 17(1): 16-23), as well as human prostate xenograft models such as LAPC 9AD (Saffran et al PNAS 1999,10:1073-1078).

Antibody effectiveness on tumor growth and metastasis formation has been studied, e.g. in a mouse orthotopic prostate or pancreatic cancer xenograft model. The antibodies may be unconjugated, such as

described in this example, or may be conjugated to a therapeutic modality as recognized in the art. Anti-PSCA-MAbs inhibit formation of both pancreatic and prostate xenografts. Anti-PSCA-MAbs also retard the growth of established orthotopic tumors and prolong the survival of tumor-bearing mice. These results indicate the utility of anti-PSCA-MAbs in the treatment of local and advanced stage prostate cancer, pancreatic cancer and those cancers listed in Table I. (see, e.g., Saffran, D., etal, PNAS 10:1073-1078 or "world wide web" URL pnas.org/cgi/doi/10.1073/pnas.051624698).

Administration of the anti-PSCA MAs led to retardation of established orthotopic tumor growth and inhibition of metastasis to scattered sites, resulting in a significant prolongation of the survival of tumor-bearing mice. These investigations indicate that PSCA is an attractive target for immunotherapy and demonstrate the therapeutic potential of anti-PSCA-MAbs for the treatment of local and metastatic prostate and pancreatic cancer. This example demonstrates that unconjugated PSCA monoclonal antibodies are effective in inhibiting the growth of human prostate tumor xenografts grown in SCID mice; therefore, a combination of such effective monoclonal antibodies is also effective.

Tumor inhibition using multiple PSCA-MAbs

Materials and methods

PSCA monoclonal antibodies:

Monoclonal antibodies were raised against PSCA as described in the example entitled "Generation of PSCA Monoclonal Antibodies (MAbs)." The antibodies are characterized by ELISA, Western blot, FACS and immunoprecipitation for their capacity to bind PSCA. Epitope mapping data for the anti-PSCA MAs, as determined by ELISA and Western analysis, recognize epitopes on the PSCA protein. Immunohistochemical analysis of prostate cancer tissue and cells with these antibodies is performed.

The monoclonal antibodies are purified from ascites or hybridoma tissue culture supernatants by protein G or protein A sepharose chromatography, dialyzed against PBS, filter sterilized and stored at -20°C. Protein determinations are performed by a Bradford assay (Bio-Rad, Hercules, CA). A therapeutic monoclonal antibody or a cocktail comprising a mixture of individual monoclonal antibodies is prepared and used to treat mice receiving subcutaneous or orthotopic injections of LAPC9 AD and HPAC tumor xenografts.

Cell lines and xenografts

The prostate cancer cell lines PC3 and LNCaP cell line as well as the fibroblast line NLH 3T3 (American Type Culture Collection) are maintained in RPMI and DMEM, respectively, supplemented with L-glutamine and 10% FBS.

PC3-PSCA and 3T3-PSCA cell populations are generated by retroviral gene transfer as described in Hubert, R.S., et al., Proe Nati Acad Sei USA, 1999. 96(25): 14523.

The LAPC-9 xenograft, which expresses a wild-type androgen receptor and provides prostate-specific antigen (PSA), was passaged into 6- to 8-week-old male ICR severe combined immunodeficiency (SCLD) mice (Taconic Farms) by s.c. trocar implant (Craft, N., et al, Nat Med. 1999, 5:280). Single cell suspensions of LAPC-9 tumor cells are prepared as described in Craft, et al.

Xenograft mouse models.

Subcutaneous (s.c.) tumors are formed by injection of 1 x 10<6> cancer cells mixed at a 1:1 dilution with Matrigel (Collaborative Research) into the right flank of male SCID mice. To test antibody effectiveness on tumor formation, i.e. antibody injections are started on the same day as tumor cell injections. As a control, mice are injected with either purified mouse IgG (ICN) or PBS; or a purified monoclonal antibody that recognizes an irrelevant antigen not expressed in human cells. In preliminary investigations, no difference has been found between mouse IgG or PBS on tumor growth. Tumor sizes are determined by caliper measurements and the tumor volume is calculated as length x width x height. Mice with subcutaneous tumors larger than 1.5 cm in diameter are euthanized.

Orthotopic injections are performed under anesthesia using ketamine/xylazine. For prostate orthotopic examinations, an incision is made through the abdomen to expose the prostate and LAPC or PC3 tumor cells (2 x IO<6>) mixed with Matrigel are injected into the prostate capsule in a 10-ul volume. To monitor tumor growth, mice are palpated and blood is collected on a weekly basis to measure PSA levels. The mice are allocated into groups for the appropriate treatments, with anti-PSCA or control MAbs being injected i.p.

Anti-PSCA-MAbs inhibit growth of PSCA-expressing xenograft cancer tumors

The effect of anti-PSCA-MAbs on tumorigenesis is tested using the HP AC and LAPC9 orthotopic models. As compared to s.c. the tumor model the orthotopic model, which requires injection of tumor cells directly into the mouse pancreas or prostate, respectively, results in local tumor growth, development of metastasis at distal sites, deterioration of mouse health and subsequent death (Saffran, D., et al, PNAS supra). These features render the orthotopic model more representative of human disease progression and allowed us to follow the therapeutic effect of MAbs on clinically relevant endpoints.

Consequently, tumor cells are injected into the mouse prostate and 2 days later the mouse is separated into two groups and treated with either: a) 250-1000ug of anti-PSCA-Ab or b) control antibody three times per week for two to five weeks.

A main advantage of the orthotopic cancer models is the ability to study the development of metastases. Formation of metastasis in mice bearing established orthotopic tumors has been examined by IHC analysis on lung sections using an antibody against a tumor-specific cell surface protein such as anti-CK20 for prostate cancer (Lin et al, Cancer Detect Prev. (2001) 25:202).

Another advantage of xenograft cancer models is the ability to study neovascularization and angiogenesis. Tumor growth is partly dependent on new blood vessel development. Although the capillary system and blood network development are of host origin, the initiation and architecture of the neovasculature is regulated by the xenograft tumor (Davidoff et al, Clin Cancer Res. (2001) 7:2870; Solesvik et al, Eur J Cancer Clin Oncol. (1984) 20 :1295). The effect of antibody and small molecule on neovascularization has been studied according to procedures known in the art, such as by IHC analysis of tumor tissues and their surrounding microenvironment.

Mice bearing established orthotopic tumors are administered injections of either anti-PSCA-MAb or control antibody over a 4-week period. Mice in both groups are allowed to establish a high tumor burden to ensure a high frequency of metastasis formation in mouse lungs. Mice are then sacrificed and their bladders, livers, legs and lungs are analyzed for the presence of tumor cells by IHC analysis. These studies demonstrate a broad antitumor efficacy of anti-PSCA antibodies on prostate cancer initiation and progression in xenograft mouse models. Anti-PSCA antibodies inhibit tumor formation of tumors as well as retard the growth of already established tumors and prolong the survival of treated mice. Furthermore, anti-PSCA-MAbs demonstrate a dramatic inhibitory effect on the spread of local prostate tumor to distal sites, even in the presence of a large tumor burden. Thus, anti-PSCA-MAbs are effective on the main clinically relevant endpoints (tumor growth), prolongation of survival and health.

Effect of PSCA-MAbs on the growth of human prostate cancer in mice

Using the above procedure, LAPC-9AI tumor cells (2.0 x 10<6> cells) were injected subcutaneously into male SCID mice. The mice were randomized into groups (n=10 in each group) and treatment started intraperitoneally (i.p.) on day 0 with HA 1-4,120 or isotype MAb control as indicated. Animals were treated twice weekly with a total of 7 doses until study day 28. Tumor growth was monitored using caliper measurements every 3 to 4 days as indicated. The results show that human anti-PSCA monoclonal antibody Hal-4,120 significantly inhibited the growth of human prostate cancer xenografts implanted subcutaneously in SCLD mice (p< 0.05) (Figure 18).

In another experiment, LAPC-9AI tumor cells (2.0 x 10<6> cells) were injected subcutaneously into male SCID mice. When tumor volume reached 50 mm3, mice were randomized into groups (n=10 in each group) and treatment initiated intraperitoneally (i.p.) with HAI-5.99.1 or isotype MAb control as indicated. Animals were treated twice weekly with a total of 5 doses until study day 14. Tumor growth was monitored using caliper measurements every 3 to 4 days as indicated. The results show that fully human anti-PSCA monoclonal antibody Hal-5.99 significantly inhibited the growth of established androgen-independent human prostate cancer xenografts implanted subcutaneously in SCLD mice (p<0.05). (figure 19).

In another experiment, LAPC-9AD tumor cells (2.5 x 10 6 cells) were injected subcutaneously into male SCID mice. When tumor volume reached 40 mm3, mice were randomized into groups (n=10 in each group) and treatment was initiated intraperitoneally (i.p.) with increasing concentrations of HA1-4,121 or isotype MAb control as indicated. Animals were treated twice weekly with a total of 7 doses until study day 21. Tumor growth was monitored using caliper measurements every 3 to 4 days as indicated. The results of this study demonstrated that HAI-4,121 inhibited the growth of established subcutaneous human androgen-dependent prostate xenografts in SCLD mice. The results were statistically significant for the 300 ug dose group on days 14, 17 and 21 (p< 0.05, Kruskal-Wallis test, two-sided with a=0.05) and for the 700 ug dose group on days 10, 14, 17 and 21 (p< 0.05, Kruskal-Wallis test, two-sided with a=0.05) (figure 20).

In another experiment, patient-derived androgen-dependent LAPC-9AD tumor cells (2.0 x 10 6 cells) were injected into the dorsal lobes of the prostates of male SCID mice. The tumors were allowed to grow for approximately 10 days at which time the mice were randomized into groups. Treatment with human 500 mg HA1-4,117, HA1-4,121 or isotype control MAb was initiated 10 days after tumor implantation. Antibodies were delivered intraperitoneally twice weekly with a total of 7 doses. Four days after the last dose, animals were sacrificed and primary tumors were excised and weighed. The results show that human anti-PSCA monoclonal antibodies Hal-4,121 (p< 0.01) and Hal-4,117 (p< 0.05) significantly inhibited the growth of LAPC-9AD prostate cancer xenografts orthotopically implanted in SCLD mice (Figure 21 ).

In another experiment, patient-derived androgen-dependent LAPC-9AD tumor cells (2.0 x 10 6 cells) were injected into the dorsal lobes of the prostates of male SCID mice. The tumors were allowed to grow for approximately 9 days at which time the mice were randomized into groups. The animals randomized into the survival groups include 11 mice in the isotype MAb control and 12 mice in the HAI-4,121-treated group. Animals were treated i.p. with 1000 µg of Hal-4,121 or 1000 µg of isotype control MAb twice weekly for a total of 9 doses. The results demonstrated that HAI-4,121 significantly (log-rank test: p<0.01) prolonged the survival of SCID mice bearing human androgen-dependent prostate tumors. Two mice in the HAI-4,121-treated group remained free of obvious tumors on day 110, the last day of the experiment (Figure 22).

Effect of PSCA-MAbs in combination with taxotere in mice

In another experiment, LAPC-9AI tumor cells (2 x 10 cells per animal) were injected subcutaneously into male SCID mice. When the tumor volume reached 65 mm<3>, animals were randomized and assigned to four different groups (n=10 in each group) as indicated. At the beginning of day 0, Hal-4,121 or isotype MAb control was administered i.p. twice weekly in a dose of 500 ug with a total of 6 doses. The last dose was given on day 17. Taxotere was given intravenously at a dose of 5 mg/kg on days 0, 3 and 7. Tumor growth was monitored every 3-4 days using caliper measurements. The results of this study demonstrate that HAI-4,121 as a single agent inhibited the growth of androgen-independent prostate xenografts in SCID mice by 45% when compared to control antibody treatment alone at day 28 (ANOVA/Tukey test: p<0.05). Administration of the isotype MAb control plus taxotere inhibited tumor growth by 28% when compared to control antibody treatment alone, which was not statistically significant. Administration of HA1-4,121 in combination with taxotere improved efficacy and resulted in a 69% inhibition of tumor growth when compared to control antibody alone (ANOVA/Tukey test: p<0.01). A statistically significant difference was also demonstrated when the HAI-4,121 plus taxotere combination group was compared to either the HAI-4,121 or isotype MAb control plus taxotere groups (ANOVA/Tukey test: p<0.05) (Figure 23).

Effect of PSCA-MAbs on the growth of human pancreatic cancer in mice

In another experiment, human HPAC pancreatic cancer cells (2 x 10<6>/mouse) were injected subcutaneously into immunodeficient ICR-SCLD mice (Taconic Farm, Germantown, NY). The mice were randomized into groups (n= 10 animals/group) and treatment with the indicated human PSCA monoclonal antibody started on the same day. Antibodies (500 mg/mouse) were delivered intraperitoneally twice weekly for a total of 8 doses. The results demonstrated that human anti-PSCA monoclonal antibodies Hal-4.121, Hal -4.117 and Hal -1.16 significantly inhibited the growth of human pancreatic cancer xenografts subcutaneously implanted in SCLD mice. Statistical analyzes were performed using a t-test (two-tailed, a=0.05) (Figure 24).

In another experiment, HPAC cells (3.0 x 10 6 cells) were implanted orthotopically into the pancreatic islet of SCLD mice. Mice were randomly assigned to three groups (n=9 in each group) as indicated. Treatment with HA1-4,121 (250 µg or 1000 µg) or isotype MAb control (1000 µg) was initiated on the day of implantation. Antibodies were administered i.p. twice weekly with a total of 10 doses. Thirteen days after the last dose, animals were sacrificed and primary tumors were excised and weighed. The results of this study demonstrated that HAI-4,121 significantly inhibited the orthotopic growth of human pancreatic cancer xenografts in SCID mice at both dose levels examined. Treatment with 250 µg and 1000 µg AGS-PSCA inhibited tumor growth by 66% and 70%, respectively (Kruskal-Wallis/Tukey test: p<0.01 and p<0.01, respectively) (Figure 25).

At autopsy, visible metastases to lymph nodes and scattered organs were observed in the control antibody-treated group. No visible metastases were observed in both HAI-4,121 treated groups. Lymph nodes, lungs and livers were removed from all animals and examined histologically for the presence of metastatic tumor. Sections from the lungs and lymph nodes removed from each animal were labeled for human cytokeratin and the number of metastases determined microscopically. The results of the histological analysis demonstrated a significant reduction in lymph node (LN) metastases in animals treated with HAI-4,121 (p= 0.0152 detected by Fisher's exact test). The incidence of metastasis and invasion was also significantly reduced in animals treated with both concentrations of HA1-4,121 (p= 0.0152 detected by Fisher's exact test). The number of lung metastases was significantly decreased in mice treated only with the 1.0 mg dose of HA1-4,121 (p= 0.0498 detected by Fisher's exact test) (Figure 26).

Effect of PSCA-MAbs on the growth of human bladder cancer in mice

In another experiment, human SW780 bladder cancer cells (2 x 10<6>/mouse) were injected subcutaneously into immunodeficient ICR-SCLD mice (Taconic Farm, Germantown, NY). The mice were randomized into groups (n= 10 animals/group) and treatment with the specified human PSCA-MAb started on the same day. Antibodies (250 mg/mouse) were delivered intraperitoneally twice weekly with a total of 7 doses. The results demonstrated that HAl-4.117 (p= 0.014), HA1-4.37 (p= 0.0056), HA1-1.78 (p= 0.001), Hal-5.99 (p= 0.0002) and HA1-4.5 (p= 0 .0008) significantly inhibited the growth of SW780 bladder tumors implanted subcutaneously in SCLD mice. Statistical analyzes were performed using a t-test (two-tailed, a=0.05) (Figure 27).

The results of these experiments demonstrate that the PSCA-MAbs can be used for therapeutic and diagnostic purposes to treat and manage the cancers listed in Table I.

Example 27

Therapeutic and diagnostic use of anti-PSCA antibodies in humans.

Anti-PSCA monoclonal antibodies have been safely and effectively used for diagnostic, prophylactic, prognostic and/or therapeutic purposes in humans. Western blot and immunohistochemical analysis of cancer tissue and cancer xenografts with anti-PSCA-MAb shows strong extensive labeling in carcinoma but significantly lower or undetectable levels in normal tissue. Detection of PSCA in carcinoma and in metastatic disease demonstrates the utility of MAbet as a diagnostic and/or prognostic indicator. Anti-PSCA antibodies have therefore been used in diagnostic applications such as immunohistochemistry of kidney biopsy specimens to detect cancer from suspected patients.

As determined by flow cytometry, anti-PSCA-MAb binds specifically to carcinoma cells. Thus, anti-PSCA antibodies are used in diagnostic whole-body imaging applications, such as radioimmunoscintigraphy and radioimmunotherapy (see, e.g., Potamianos S., et. al. Anticancer Res 20(2A):925-948 (2000)) for the detection of localized and metastatic cancer showing expression of PSCA. Shedding or release of an extracellular domain of PSCA into the extracellular environment, as seen for alkaline phosphodiesterase B10 (Meerson, N. R., Hepatology 27:563-568

(1998)), allows diagnostic detection of PSCA by anti-PSCA antibodies in serum and/or urine samples from suspected patients.

Anti-PSCA antibodies that specifically bind PSCA are used in therapeutic applications for the treatment of cancers that express PSCA. Anti-PSCA antibodies are used as an unconjugated modality and as a conjugated form where the antibodies are bound to one of various therapeutic or imaging modalities well known in the art, such as a prodrug, enzymes or radioisotopes. In preclinical studies, unconjugated and conjugated anti-PSCA antibodies are tested for tumor prevention and growth inhibition efficacy in the SCID mouse cancer xenograft models,/e£s. kidney cancer models AGS-K3 and AGS-K6 (see e.g. the example labeled "PSCA monoclonal antibody-mediated inhibition of tumors in vivo"). Either conjugated or unconjugated anti-PSCA antibodies are being used as a therapeutic modality in human clinical trials either alone or in combination with other treatments as described in the following examples.

Example 28

Human clinical trials for the treatment and diagnosis of human carcinomas using human anti-PSCA antibodies in vivo

Antibodies are used according to the present invention that recognize an epitope on PSCA and are used in the treatment of certain tumors as listed in Table I. Based on several factors, extensive PSCA expression levels, tumors as listed in Table I currently preferred indications. In connection with each of these indications, three clinical approaches have been successfully pursued. I.) Additional therapy: In additional therapy, patients are treated with anti-PSCA antibodies in combination with a chemotherapeutic or antineoplastic agent and/or radiation therapy. Primary cancer targets, such as those listed in Table I, are treated with standard protocols by the addition of anti-PSCA antibodies to standard first- and second-line therapy. Protocol design addresses efficacy as assessed by reduction in tumor mass as well as the ability to reduce common doses of standard chemotherapy. These dose reductions allow additional and/or prolonged therapy by reducing dose-related toxicity of the chemotherapeutic agent. Anti-PSCA antibodies have been used in many additional clinical trials in combination with the chemotherapeutic or antineoplastic agents adriamycin (advanced prostate carcinoma), cisplatin (advanced head and neck and lung carcinomas), taxol (breast cancer) and doxorubicin (preclinical).

n.) Monotherapy: In connection with the use of the anti-PSCA antibodies in the monotherapy of tumors, the antibodies are administered to patients without a chemotherapeutic or antineoplastic agent. In one embodiment, monotherapy is administered clinically in end-stage cancer patients with extensive metastatic disease. Patients show some disease stabilization. Trials demonstrate an effect in resistant patients with cancerous tumors.

Ul.) Imaging agent: By binding a radionuclide (e.g. iodine or yttrium (I131, Y<90>) to anti-PSCA antibodies, the radioactively labeled antibodies are used as a diagnostic and/or imaging agent. In such a role localizes the labeled antibodies to both solid tumors as well as metastatic lesions of cells expressing PSCA.In conjunction with the use of the anti-PSCA antibodies as imaging agents, the antibodies are used as an adjunct to the surgical treatment of solid tumors, as both a presurgical screen as well as a postoperative follow-up to determine which tumor remains and/or returns.In one embodiment, an (<in>In)-PSCA antibody is used as an imaging agent in a Phase I human clinical trial in patients who have a carcinoma that expresses PSCA (analogously see e.g. Divgi et al. J. Nati. Cancer Inst. 83:97-104 (1991)). Patients have been followed with standard anterior and posterior gamma cameras. The results indicate that primary le tions and metastatic lesions have been identified.

Dose and route of administration

As recognized by those skilled in the art, dosage considerations can be determined through comparison with the analog products that are in the clinic. Thus, anti-PSCA antibodies can be administered at doses in the range of 5 to 400 mg/m<2>, with the lower doses used e.g. in connection with security investigations. The affinity of anti-PSCA antibodies relative to the affinity of a known antibody for its target is one parameter used by those skilled in the art to determine analogous dosage regimens. Furthermore, compared to the chimeric antibody, anti-PSCA antibodies, which are fully human antibodies, have slower clearance; therefore, dosage in patients with such fully human anti-PSCA antibodies can be lower, perhaps in the range of 50 to 300 mg/m<2>, and still remain effective. Dosage in mg/m<2>, as opposed to the conventional measurement of dose in mg/kg, is a measurement based on surface area and is an appropriate dosage measurement designed to encompass patients of all sizes from infants to adults.

Three distinct delivery approaches are applicable for the delivery of anti-PSCA antibodies. Conventional intravenous delivery is one standard delivery technique for many tumors. However, in the context of tumors in the peritoneal cavity, such as tumors of the ovaries, bile ducts, other ducts and the like, intraperitoneal administration may be shown to be advantageous to obtain a high dose of antibody at the tumor and also to minimize antibody secretion. Similarly, certain solid tumors have vasculature suitable for regional perfusion. Regional perfusion allows a high dose of antibody at the site of a tumor and minimizes short-term shedding of the antibody.

Clinical Development Plan (CDP)

Overview: The CDP pursues and develops treatments for anti-PSCA antibodies in conjunction with adjunctive therapy, monotherapy and as an imaging agent. Trials initially demonstrate safety and then confirm efficacy of repeated doses. Trials are open-label comparisons of standard chemotherapy with standard therapy plus anti-PSCA antibodies. As will be recognized, one criterion that may be used in recruiting patients is PSCA expression levels in their tumors as determined by biopsy.

As with any protein or antibody infusion-based therapeutic agent, the safety concerns are primarily related to (i) cytokine release syndrome, ie, hypotension, fever, shaking, chills; (ii) the development of an immunogenic response to the material (ie, development of human antibodies by the patient to the therapeutic antibody or HAHA response); and, (iii) toxicity to normal cells expressing PSCA. Standard tests and follow-up are used to monitor each of these safety measures. Anti-PSCA antibodies have been found to be safe for human administration.

Example 29

Human clinical trial: monotherapy with human anti-PSCA antibody Anti-PSCA antibodies are safe in connection with the additional trial described above, a Phase II human clinical trial confirms the effectiveness and optimal dosage for monotherapy. Such trials are completed and carry the same safety and outcome analyzes as the additional trial described above with the exception of patients not receiving chemotherapy concurrently receiving doses of anti-PSCA antibodies.

Example 30

Human clinical trial: diagnostic imaging with anti-PSCA antibody

Once again, as the adjunctive therapy described above is safe within the safety criteria described above, a human clinical trial is being conducted regarding the use of anti-PSCA antibodies as a diagnostic imaging agent. The protocol is designed in a mainly similar way to those described in the area, such as in Divgi et al. J. Nati. Cancer Inst. 83:97-104

(1991). The antibodies have been found to be both safe and effective when used as a diagnostic modality.

Example 31

Human clinical trial adjunctive therapy with human anti-PSCA antibody and chemotherapeutic agent,

radiation and/or hormone ablation therapy

A phase I human clinical trial is being initiated to assess the safety of six intravenous doses of a human anti-PSCA antibody in the treatment of a solid tumor,/e£s. a cancer of a tissue listed in Table 1.1 the investigation is the safety of single doses of anti-PSCA antibodies when used as an adjunctive therapy to an antineoplastic or chemotherapeutic or hormone ablation agent as defined herein, such as, without limitation: cisplatin, topotecan, doxorubicin, adriamycin, taxol, Lupron, Zoladex, Eulexin, Casodex, Anandron or similarly rated. The trial design involves the delivery of approximately six single doses of an anti-PSCA antibody with the dose of antibody escalating from approximately 25 mg/m<2> to approx. 275 mg/m<2> over the course of treatment according to the following or similar programme:

Patients are followed closely for a week after each administration of antibody and chemotherapy. In particular, patients are assessed for the safety significance mentioned above: (i) cytokine release syndrome, ie, hypotension, fever, shaking, chills; (ii) the development of an immunogenic response to the material (ie, development of human antibodies by the patient to the human therapeutic antibody or HAHA response); and (iii) toxicity to normal cells expressing PSCA. Standard tests and follow-up are used to monitor each of these safety measures. Patients are also assessed for clinical outcome and especially reduction in tumor mass as evidenced by MRI or other imaging.

The anti-PSCA antibodies have been demonstrated to be safe and effective. Phase 11 trials confirm efficacy and refine optimal dosage.

Example 32

RNA interference (RNAi)

RNA interference (RNAi) technology has been implemented for a variety of cell experiments relevant to oncology. RN Ai is a posttranscriptional gene silencing mechanism activated by the doublet

RNA (dsRNA). RNAi induces specific mRNA degradation which leads to changes in protein expression and subsequently in gene function. In mammalian cells, these dsRNAs, short interfering RNA (siRNA), have the correct composition to activate the RNAi pathway for targeting degradation, specifically some mRNAs. See Elbashir S.M., et al, Duplexes of 21-nucleotide RNAs Mediate RNA interference in Cultured Mammalian Cells, Nature 411(6836):494-8 (2001). Thus, RNAi technology is successfully used in mammalian cells to turn off targeted genes.

Loss of cell proliferation control is a sticking point for cancerous cells; is thus relevant for determining the role of PSCA in cell survival/proliferation experiments. Accordingly, RNAi was used to investigate the function of the PSCA antigen. To generate siRNA for PSCA, algorithms were used to predict oligonucleotides that display the critical molecular parameters (G:C content, melting temperature, etc.) and have the ability to significantly reduce the expression levels of the PSCA protein when introduced into cells. According to this example, PSCA-siRNA compositions are used which comprise siRNA (double-stranded short interfering RNA) corresponding to the nucleic acid ORF sequence of the PSCA protein or subsequences thereof. Thus, siRNA subsequences used in this way are generally 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35 or more than 35 consecutive RNA nucleotides in length. These siRNA sequences are complementary and non-complementary to at least part of the mRNA coding sequence. In a preferred embodiment, the subsequences are 19-25 nucleotides in length, most preferably 21-23 nucleotides in length. In preferred embodiments, this siRNA achieves potent knockdown of PSCA antigen in cells expressing the protein and has functional effects as described below.

The selected siRNA (PSCA.b oligo) was tested in a variety of cell lines in the survival/proliferation MTS assay (measures cellular metabolic activity). The tetrazolium-based colorimetric assay (i.e., MTS) detects exclusively viable cells, since living cells are metabolically active and can therefore reduce tetrazolium salts to colored formazan compounds; however, dead cells do not. Furthermore, this PSCA.b oligo achieved robust knockdown of PSCA antigen in cells expressing the protein and had functional effects as described below using the following protocols.

Mammalian siRNA transfections: The day before siRNA transfection, the different cell lines were plated in media (RPMI 1640 with 10% FBS w/o antibiotics) at 2x10<3> cells/well in 80 (ul (96-well plate format) for the survival/ The MTS Experiment In parallel with the PSCA-specific siRNA oligo, the following sequences were included in each experiment as controls: a) Mock-transfected cells with Lipofectamin 2000 (Invitrogen, Carlsbad, CA) and hybridization buffer (no siRNA); b) Luciferase-4-specific siRNA (targeting sequence: 5'-AAGGGACGAAGACGAACACUUCTT-3') (SEQ ID NO:77); and c) Eg5-specific siRNA (targeting sequence: 5'-AACTGAAGACCTGAAGACAATAA-3') (SEQ ID NO:78). SiRNAs were used at 10 nM and 1 µg/ml Lipofectamine 2000 final concentration.

The procedure was as follows: the siRNAs were first diluted in OPTLMEM (serum-free transfection media, Invitrogen) at 0.1 µM (10-fold concentrated) and incubated for 5-10 min at RT. Lipofectamine 2000 was diluted to 10 µg/ml (10-fold concentrated) for the total number of transfections and incubated 5-10 minutes at room temperature (RT). Appropriate amounts of diluted 10-fold concentrated Lipofectamin 2000 were mixed 1:1 with diluted 10-fold concentrated siRNA and incubated at RT for 20-30" (5-fold concentrated transfection solution). 20 µl of the 5-fold concentrated transfection solutions were added to the respective samples and incubated at 37°C for 96 hours before analysis.

MTS assay: The MTS assay is a colorimetric method for determining the number of viable cells in proliferation, cytotoxicity or chemosensitivity assays based on a tetrazolium compound [3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl )-2-(4-sulfophenyl)-2H-tetrazolium, inner salt; MTS(b)] and an electron coupling reagent (phenazine ethosulfate; PES). Experiments were performed by adding a small amount of the dissolution reagent directly to culture wells, incubating for 1-4 hours and then measuring absorbance at 490nm with a 96-well plate reader. The amount of colored formazan product as measured by the amount of 490nm absorbance is directly proportional to the mitochondrial activity and/or number of live cells in culture.

To address the function of PSCA in cells, PSCA has been silenced by transfection of the endogenously expressing PSCA cell lines.

Another embodiment according to the invention is a method for analyzing PSCA-related cell proliferation is the measurement of DNA synthesis as a marker for proliferation. Labeled DNA precursors (ie 3H-thymidine) are used and their incorporation into DNA is quantified. Incorporation of the labeled precursor into DNA is directly proportional to the amount of cell division occurring in the culture. Another method used to measure cell proliferation is performing clonogenic assays. In these experiments, a defined number of cells are plated on the appropriate matrix and the number of colonies formed after a period of growth after siRNA treatment is counted.

In PSCA cancer target validation, complementary to the cell survival/proliferation analysis with apoptosis and cell cycle profile studies are considered. The biochemical anchor point for the apoptotic process is genomic DNA fragmentation, an irreversible event that commits the cell to death. One method for observing fragmented DNA in cells is the immunological detection of histone-complexed DNA fragments by an immunoassay (i.e. cell death detection ELISA) which measures the enrichment of histone-complexed DNA fragments (mono- and oligonucleosomes) in the cytoplasm of apoptotic cells. This experiment does not require pre-labeling of the cells and can detect DNA degradation in cells that do not proliferate in vitro (ie healthy isolated tumor cells).

The most important effector molecules for triggering apoptotic cell death are caspases. Caspases are proteases that, when activated, cleave a variety of substrates at the carboxy-terminal site of an aspartate residue, mediating very early steps of apoptosis upon activation. All caspases are synthesized as proenzymes and activation involves cleavage at aspartate residues. In particular, caspase-3 appears to play a central role in the initiation of cellular apoptosis events. Caspase-3 activation assays detect early apoptosis events. After RNAi treatments, Western blot detection of active caspase-3 presence or proteolytic cleavage products (ie, PARP) found in apoptotic cells further supports an active induction of apoptosis. Because the cellular mechanisms that result in apoptosis are complex, each has its own advantages and limitations. Consideration of other criteria/endpoints such as cellular morphology, chromatin condensation, membrane blebbing, apoptotic bodies help to further support cell death as apoptotic. Since not all the gene targets that regulate cell growth are antiapoptotic, the DNA content of permeabilized cells is measured to obtain the profile of DNA content or cell cycle profile. Nuclei of apoptotic cells contain less DNA due to the leakage out to the cytoplasm (sub-Gl population). In addition, the use of DNA dyes (ie propidium iodide) can also differentiate between the different phases of the cell cycle in the cell population due to the presence of different amounts of DNA in G0/G1, S and G2/M. In these surveys, the subpopulations are quantified.

For the PSCA gene, RNAi studies facilitate the understanding of the contribution of the gene product in cancer response pathways. Such active RNAi molecules have application in identification experiments to screen for MAbs that are active antitumor therapeutic agents. Furthermore, siRNAs are administered as therapeutic agents to cancer patients to reduce the malignant growth of many cancer types, including those listed in Table 1. When PSCA plays a role in cell survival, cell proliferation, tumorigenesis or apoptosis, it is used as a target for diagnostic, prognostic , preventive and/or therapeutic purposes.

Example 33

Detection of PSCA protein in cancer patient samples by IHC

Expression of PSCA protein in tumor samples from cancer patients was detected using the antibody HAI-4,117. Formalin-fixed, paraffin-embedded tissue was cut into 4 micron sections and placed on glass slides. The sections were awoken, rehydrated, and treated with antigen retrieval solution (Antigen Retrieval Citra Solution; BioGenex, 4600 Norris Canyon Road, San Ramon, CA, 94583) at high temperature. Sections were then incubated in fluorescein-conjugated human monoclonal anti-PSCA antibody, Hal-4,117, for 16 hours at 4°C. Slides were washed three times in buffer and further incubated with rabbit antifluorescein for 1 hour and, after washing in buffer, immersed in DAKO EnVision+™ peroxidase-conjugated goat antirabbit immunoglobulin secondary antibody (DAKO Corporation, Carpenteria, CA) for 30 minutes. The sections were then washed in buffer, developed using the DAB kit (Sigma Chemicals), counterstained using hematoxylin and analyzed by brightfield microscopy. The results show expression of PSCA in the tumor cells of prostate adenocarcinoma (A, B), bladder transitional carcinoma (C) and pancreatic ductal adenocarcinoma (D). These results indicate that PSCA is expressed in human cancers and that antibodies directed against this antigen are useful as diagnostic reagents (Figure 17).

These results indicate that PSCA is a target for diagnostic, prognostic and therapeutic applications in cancer.

Throughout this application, various website data content, publications, patent applications and patents are referenced. (Web pages are referenced by their Uniform Resource Locator or URL, addresses on the world wide web.) Descriptions of each of these references are hereby incorporated by reference here in their entirety.

The present invention shall not be limited in scope by the embodiments described herein, which are intended as simple illustrations of individual aspects of the invention and any that are functionally equivalent are within the scope of the invention. Various modifications to the models and methods according to the invention, in addition to those described here, will be clear to those skilled in the art from the preceding description and teachings and are accordingly intended to fall within the scope of the invention. Such modifications or other embodiments may be made without departing from the true scope and spirit of the invention.

Claims (12)

PATENT CLAIMS 1. Monoclonal antibody or an antigen-binding fragment thereof, characterized in that it comprises an antigen binding site that binds specifically to a PSCA protein comprising the amino acid sequence SEQ ID NO: 2 and where the monoclonal antibody comprises the amino acid sequence of the VH region of SEQ ID NO: 35 and the VL region of SEQ ID NO: 37. 2. Antibody or antigen-binding fragment according to claim 1, characterized by the fact that the monoclonal antibody is produced by the hybridoma deposited under American Type Culture Collection (ATCC) Accession No. PTA-6699. 3. Antibody or fragment according to claim 1 or 2, characterized in that the fragment is an F ab, F(ab')2, Fv or scFv fragment. 4. Antibody or fragment according to any one of claims 1, 2 or 3, characterized in that the antibody or fragment is coupled to a detectable marker. 5. Hybridoma, characterized in that it produces the monoclonal antibody according to claim 1 or 2. 6. Hybridoma according to claim 5, characterized in that the hybridoma is deposited under American Type Culture Collection (ATCC) Accession No. PTA-6699. 7. Polynucleotide, characterized in that it codes for a sequence comprising a light chain variable region and a heavy chain variable region of the antibody according to any one of claims 1, 2 or 3. 8. Polynucleotide according to claim 7, characterized in that it codes for the light chain and the heavy chain of the antibody according to any one of claims 1, 2 or 3. 9. Vector, characterized in that it comprises the polynucleotide according to claim 7 or claim 8. 10. Cell, characterized in that it is transfected with the vector according to claim 9, 11. Composition, characterized in that it comprises the antibody, or F ab, F(ab')2, Fv or scFv fragment thereof according to any one of claims 1 to 4. 12. Method for producing an antibody according to claim 1 or an antigen-binding fragment thereof, characterized in that it comprises: cultivation of a cell, where: (a) the cell has been transfected with a polynucleotide encoding a sequence comprising a light chain variable region of the antibody according to claim 1 and a polynucleotide encoding a sequence comprising a heavy chain variable region of the antibody according to claim 1; or (b) the cell of claim 10; whereby the antibody or an antigen-binding fragment thereof is produced.