WO1993005658A1 - Targeted cytotoxic effector cells and methods for their production and use - Google Patents

Abstract

The present invention is directed to the targeting of cytotoxic immunological effector cells, such as human macrophages and lymphokine activated killer (LAK) cells, to tumors, such as colon adenocarcinomas, their metastases, and other diseased tissues, which selectively express specific cell surface antigens. The targeting of the effector cells is accomplished by binding to them antibodies which are specific for the selectively expressed antigens of the tumors or other diseased tissues. The invention is also directed to the process of producing these targeted, cytotoxic effector cells, and to the targeted, cytotoxic effector cells themselves. The invention is also directed to uses for these targeted effector cells including their in vivo use to suppress the growth of, to kill, and their in vivo and in vitro use to diagnostically image tumor and other diseased animal cells in humans and other animals.

Description

Title of the Invention

Targeted Cytotoxic Effector Cells And Methods For Their Production And Use

I. Cross Reference to Related Applications This application is a continuation-in-part (CIP) of United States Patent

Application 07/765,227 which was filed on September 25, 1991 and which is related to United States Patent Applications Serial Nos. 07/203,198 filed on June 7, 1988 and 07/130,777 (now abandoned) which was filed on December 9, 1987. II. Rights of the United States Federal Government

in this Invention

This invention was made with United States government support under National Cancer Institute Grant Numbers CA-57584 and CA-35711. The United States federal government has certain rights in this invention. III. Field of the Invention

The present invention is directed to the targeting of cytotoxic, immunological, effector cells, such as human macrophages, natural killer (NK) cells, eosinophils, basophils, neutrophils, tumor inf iltrating lymphocytes (TIL) and lymphokine activated killer (LAK) cells, to tumors, such as colon adenocarcinomas and their metastases, and other diseased tissues, which selectively express specific cell surface antigens. IV. Background of the Invention A. Colorectal Cancer and Hepatic Metastases

Some tumors and other diseased tissues selectively express constitutive antigens which are not expressed by normal animal tissues. An example is colorectal adenocarcinoma and its SF-25 constitutive antigen.

Colorectal cancer is one of the most common malignancies in both men and women in the Western world. More than 150,000 new cases will be diagnosed in 1991 in the United States alone (Boring et al., Cancer Statistics 41:19-36 (1991)). Despite major advances in general patient care and surgical therapy, the mortality rate associated with this disease has not changed significantly over the last forty years (Fleischer et al., JAMA 267:580-586 (1989)). Indeed, about 60,000 patients die of this disease each year in this country principally because of advanced disease or recurrence (Cancer Facts & Figures - 1990, American Cancer Society, Inc., Atlanta, Ga. (1990)).

Sixty percent of the patients with advanced colon adenocarcinoma will develop hepatic metastases (Weiss, L., J. Pathol. 150: 195-203 (1986)). Although 70-80% of these patients will present with operable tumors at the time of diagnosis, even complete surgical resection is often unable to permit long term survival due to the presence of occult disease or hepatic micrometastases. Numerous post operative adjuvant treatment regimens have failed to reduce the incidences of hepatic metastases and tumor recurrence (Grem, J.L., Semin. Oncol. 18 (Suppl. 1):17-26 (1991); Buyse et al. , JAMA 259:3571-3578 (1988); Mayer, R.J. N. Engl. J. Med. 322:399-401 (1990); Wolmark etal., J. Natl. Cancer Inst. 80:30-36 (1988)). The death rate will undoubtedly remain the same until improved methods for the treatment of hepatic involvement become available. Thus, new, useful clinical treatment regimens are needed. B. The SF-25 Antigen

A repertoire of MAbs against transformed cells of human endodermal origin has been developed (Wilson et al., Proc Natl. Acad. Sci. 55:3140-3144 (1988)). One such murine monoclonal antibody, SF-25, recognizes a 125 Kd cell surface glycoprotein designated the SF-25 antigen (Takahashi et al., Cancer Res. 48:6513-6519 (1988)). A hybridoma which secretes the SF-25 monoclonal antibody that selectively binds to this SF-25 antigen was deposited under the provisions of the Budapest Treaty at the American Type Culture Collection, 12301 Parklawn Drive, Rockville, Maryland, and has been given the ATCC designation HB 9599. The SF-25 antigen and the SF-25 hybridoma cell line have been described in U.S. Patent application Serial No. 07/203,198 which was filed on June 7, 1988 and is herein incorporated by reference.

Murine MAb SF-25 recognizes a 125 KDa cell surface antigen, the SF- 25 antigen, which is highly expressed in human colon adenocarcinomas, their hepatic metastases, and other primary tumors of endodermal origin. Murine MAb SF-25 has a high association constant (K_A = 2.36 x 10⁹/M) and when injected iv. is immunolocalized to human colon adenocarcinoma xenografts in nude mice (Takahashi et al., Cancer Res. 48:6513-6519 (1988); Takahashi et al., Gastroenterol. 96:1311-1329 (1989)). However, the MAb SF-25 by itself does not inhibit tumor cell growth in vitro (unpublished data).

Metabolic and cell surface labeling studies have demonstrated that the SF-25 antigen is a disulfide-bond-Iinked heterodimer which is composed of two subunits termed a and β. The expression of the SF-25 antigen in colon adenocarcinoma tissues is uniform in contrast to the heterogeneous expression of other tumor associated antigens (Atkinson et al. , Cancer Res. 42:4820-4823 (1982); Hand et al. , Id. 43:128-135 (1983)). In addition, it appears that there is no antigenic modulation of the SF-25 antigen in liver metastases (Takahashi et al., submitted to Cancer Res.). The SF-25 antigen and antibodies which recognize this antigen have been extensively described in related U.S. Patent Application Serial Number 07/203,198 which was filed on June 7, 1988, the contents of which are herein incorporated by reference. The SF-25 antigen is a constitutive antigen that is expressed on most if not all tumors of endodermal origin. The SF-25 antigen has been shown by immunohistological staimng to be expressed by the following human tumor types: colon adenocarcinoma; rectal adenocarcinoma; hepatocellular carcinoma; cholangiocellular carcinoma; gastric adenocarcinoma; breast adenocarcinoma; pancreatic adenocarcinoma; bladder adenocarcinoma; squamous cell carcinoma of the lung; adenocarcinoma of the lung; small cell carcinoma of the lung; large cell carcinoma of the lung; kidney carcinoma; ovary adenocarcinoma; cervix carcinoma; endometrial adenocarcinoma; choriocarcinoma; leukemia; lymphoma; and malignant melanoma. Previous studies of the SF-25 antigen's distribution revealed that a number of normal tissues were found to be negative by immunohistological staining including: esophagus; stomach; small intestine; colon; liver; bile duct; spleen; adrenal gland; lung; thyroid; skin; skeletal muscle; myocardium; connective tissue; brain; and spinal cord. Positive staining was present in a subpopulation of proximal tubular cells of the kidney. Weak staining was also observed in normal islet cells of the pancreas.

The SF-25 antigen is localized on the tumor cell surface and antibody binding to the SF-25 antigen does not induce internalization therefrom. The SF-25 antigen is not shed from the cell when it is examined by radioimmunoassay in culture supernatant and flow cytometric analysis. Furthermore, MAb SF-25 has a high association constant (K_A = 1.36 x 10⁸/M) and is able to immunolocalize to human colon adenocarcinomas established in the livers of nude mice (Takahashi et al., Gastroenterology 95:1317-1329 (1989)). The high number of antibody binding sites per cell (2.5 x 10⁵/colon adenocarcinoma cell) suggest that the SF-25 MAb will be bound to the tumor cells in a high density. Taken together, these properties suggest that the SF-25 MAb may be effective as an immunotherapeutic reagent (Schlom et al., in Monoclonal Antibodies in Cancer: Advances in Diagnosis and Treatment, (Roth, J.A. Ed.), Futura Publishing Company, Mount Kisco, NY, 1-65 (1986); Oldham, R.K., in Biological Response Modifiers and Cancer Therapy, (Chlao, J.K., Ed.) Marcel Dekker, Inc. New York, 3-16 (1988)).

C. Monoclonal Antibodies and the Treatment of Cancer Patients

Possible anti-tumor mechanisms mediated by MAbs include: 1) induction of tumor cytotoxicity by effector cells such as NK-cells and macrophages (Ravetch et al. , Ann. Rev. Immunol 9:457-492 (1991)); 2) activation of complement and induction of complement-mediated cytotoxicity (Frank, M. M., N. Engl J. Med. 316:1525-1530 (1987)); 3) interference with cell growth or differentiation by binding growth factors or receptors on the surface of tumor cells (Sporn et al. , Nature 313:145-141 (1985)); Rodeck et al. , Cancer Res. 47:3692-3696 (1987)); 4) induction of anti-idiotypic antibodies which subsequently have been used as novel vaccines against tumors (Wettendorff et al. , Proc. Nad. Acad. Sci. USA 86:3181-3191 (1989)); and 5) delivery of cytotoxic agents such as drugs, toxins, and radionucleotides to the tumor cells (Vitetta et al. , Science 238: 1098-1104 (1987); Waldmann, T. A., Science 252:1651-1662 (1991); Dillman, R.O., Ann. Int. Med. 111:592-603 (1989)).

Cells with cytotoxic potential that bear receptors for the Fc fragment of IgG (FcγR) may bind and lyse target cells in the presence of antibody (antibody-dependent cell-mediated cytotoxicity; ADCC) (Kay et al. , J. Immunol. 118:2058-2066 (1977); Lubeck et al. , Cell Immunol. 111:101-111 (1988)). ADCC requires the simultaneous binding of the Fab fragment of the antibody to its antigen and the binding of the Fc fragment to FcγR expressed on the effector cells. Macrophages express the three types of the FcγR which have been identified in human cells (FcγRI, II and III). FcγRI is found only on macrophages and is important for ADCC. NK-cells only express low affinity FcγR type III which will initiate ADCC by NK-cells upon binding to antibody (Ravetch et al. , Ann. Rev. Immunol. 9:451-492 (1991); Unkeless et al. , Id. 6:251-281 (1988); Adams et al. , Id. 2:283-318 (1984); Perussia et al. , J. Exp. Med. 170:13-86 (1989); Vivier et al. , J. Immunol 146:206-210 (1991)).

The initial use of unmodified murine monoclonal antibodies (MAb) to treat humans with cancer has been disappointing. Only 23 partial and 3 complete remissions have been reported among 185 patients in 25 clinical trials. This is partially due to the fact that most mouse MAbs are not cytocidal against neoplastic cells in humans because these MAbs do not participate in human-complement or cell-mediated cytotoxicity. (Waldmann, T.A., Science 252:1657-1662 (1991); Catane et al., J. Med. Sci. 24:471 (1988)).

It has been proposed that a tumor specific antibody may be useful for targeting cytotoxic effector cells to tumor ceils. However, the improved in vivo immunoiocalization of effector cells by antibody has never been clearly demonstrated (Nelson, H., Cancer Cells 3:163 (1991)).

A chimeric MAb (c-SF-25 MAb) that has the Fc fragment of human

IgG1 and the Fab fragment of the murine SF-25 MAb is mentioned in Takahashi et al., Antibody Immunocon. Radiopharm. 3:86 (1990).

D. The SCID Mouse

SCID mice have no functional T and B-cells and will not reject xenografts of human lymphocytes or human tumors (Bosma et al., Nature 301:521 (1983); and Id., Annu. Rev. Immunol 9:323 (1991)). Therefore, the use of SCID mice is viewed as a possible model to explore in vivo immune responses of human lymphocytes which are difficult to study in man. Repopulation of human T or B-cells and propagation of human tumors including infiltrating lymphocytes in SCID mice have been demonstrated (Mosler et αl., Nature 555:256 (1988); McCune et al. , Science 241:1632 (1988); Bankert et al. , Curr. Top. Microbiol. Immunol 152:201 (1989); Pfeffer et al., Ibid. 152:212 (1989); and Simpson et al. , Immunol Rev. 124:91 (1991)). SCID mice have also been used to examine the growth of human lymphoma cells following exposure to cytokine-activated human killer cells in vitro (Schmidt-Wolf et al. , J. Exp. Med. 174:139 (1991)). However, the use of the SCID mouse model to test the anti-tumor effects of exogenous human effector cells by intravenous (iv.) administration has not been established. V. Summary of the Invention

The arming of the effector cells is accomplished by binding to them antibodies which are specific for the selectively expressed antigens of the tumors or other diseased tissues. The invention is also directed to the process of producing these targeted effector cells, and to the targeted effector cells themselves. The invention is further directed to uses for these targeted effector cells including their in vivo use to suppress the growth of, to kill, and their in vivo and in vitro use to diagnostically image tumor and other diseased animal cells in humans and other animals.

The present invention provides a method of producing a targeted, cytotoxic, effector cell, comprising treating said cytotoxic effector cell with a) a cytokine and b) an antibody or an effective derivative or fragment thereof, the antibody being to a constitutive antigen of a tumor or other diseased tissue, wherein the treatment with the antibody or effective fragment or derivative thereof in the presence of a conjugating reagent, whereby the antibody or effective derivative or fragment thereof is bound to the cytotoxic effector cell.

In the method of producing the targeted, cytotoxic effector cell of the present invention, the effector cell may be treated with the cytokine prior to, or simultaneously with, or after treatment of the effector cell with an antibody or effective derivative or fragment thereof.

The present invention also provides a targeted, cytotoxic, effector cell, comprising a stable complex between a cytotoxic effector cell and an antibody, effective fragment, or derivative thereof, wherein said antibody, effective fragment, or derivative thereof preserves its binding ability towards its antigen after it has been bound to said cytotoxic effector cell. The present invention also provides a method of suppressing the growth of a tumor or other diseased tissue, comprising administering to an animal an effective single dose, repeated doses, or an infusion of the targeted cytotoxic effector cell.

The present invention also provides a method of killing tumor cells, comprising administering to an animal an effective single dose, repeated doses, or an infusion of the targeted cytotoxic effector cell.

The present invention also provides a method of preventing the development of tumors, comprising administering to an animal an effective single dose, repeated doses, or an infusion of the targeted cytotoxic effector cell.

The present invention also provides a method of imaging cells of tumors or other diseased tissues in vivo, comprising: administering an effective dose of detectably labeled, targeted, cytotoxic, effector cells to an animal; and measuring the distribution of said detectably labeled, targeted, effector cells in said animal.

The present invention further provides an in vitro method of imaging cells of tumors or other diseased tissues, comprising: administering an effective concentration of detectably labeled, targeted, cytotoxic effector cells to a tissue removed from an animal; and measuring the distribution of said detectably labeled, targeted, effector cells in said tissue.

The inventors have also developed a hepatic metastatic model of human colon adenocarcinoma in severe combined immune deficiency (SCID) mice which is useful for testing the effects of anti-tumor agents against colon adenocarcinoma cells. The SCID mouse model is useful because: 1) the blood supply to the tumor cells grown in the liver is substantially better than that to tumor cells grown in previously described models that used subcutaneous tumor xenografts; and 2) the SCID mouse lacks both T and B cells and therefore will accept xenografts of normal as well as tumor human tissues (Proc. Curr. TopMicrobiol Immunol 152: 1-263 (1989); Bosma et al., Ann. Rev. Immunol 9:323-350 (1991); McCune et al ., Science 241: 1632-1639 (1988); Mosier et al., Nature 335:256-259 (1988)). Inventors demonstrated that the targeted human effector cells of the present invention which had c-SF-25 MAbs bound to them, immunolocalized in vivo to hepatic metastases of human colon adenocarcinoma cells in these SCID mice. Inventors further demonstrated that these c-SF-25 MAb targeted human effector cells had a dramatic and potent antitumor effect which produced dramatic reductions in hepatic tumor size as well as the elimination of hepatic tumors in some mice.

The advantages of the present invention include the following. The cells of the present invention inhibit the in vivo growth of human-derived tumors. The prior technology was only performed in vitro. The cells of the present invention inhibit tumor growth in vivo in the liver. The present invention allows for the delivery to a tumor or other diseased tissue of a variety of effector cells, including cells which may not express Fc receptors on their cell surfaces. The properties of the SF-25 Fab2' allow for the targeting of effector cells to a variety of human tumors. These are all advantages of the present invention which did not exist in the prior technology.

VI. Brief Description of The Figures

Figure 1 shows in situ autoradiography of human PBLs labeled with ³H-uridine.

Figure 2 shows that iv. injected PBLs were detectable in the peripheral blood of SCID mice (Figure 2A) shortly after their injection but that only a small fraction of the PBLs was found in the peripheral circulation three hours later (Figure 2B).

Figure 3 shows the biodistribution of human PBLs after their iv. administration to SCID mice in various organs as a function of time. These biodistribution data are plotted as the specific uptake of PBLs in each organ (cpm/g; Figure 3A) and also as total radioactivity per organ (cpm/organ; Figure 3B). Figure 4 shows the number of c-SF-25 MAb which were bound per human LAK cell as a function of time after the MAbs were cross-linked to the

LAK cells by the modified PEG method of this invention (open circles) or after they were merely preincubated with the LAK cells (closed circles). The number of c-SF-25 MAb bound per human LAK cell after cross-linking by this PEG method was examined using ¹²⁵I-labeled c-SF-25-MAb. The time course of dissociation was examined by incubating these LAK cells in antibody-free medium at 37°C for 1-48 hrs. Human LAK cells were efficiently cross-linked with c-SF-25 MAb by this PEG technique and approximately 10 times more antibodies were cross-linked to LAK cells than by mere preincubation with the c-SF-25 MAb.

Figure 5 shows that in the presence of c-SF-25 MAb, purified human NK-cells and macrophages (Figure 5B) mediated strong ADCC against human colon-adenocarcinoma derived LS 180 tumor cells as measured in a four hour 5¹Cr-reIease assay. This ADCC was substantially greater than that mediated by a mixed cellular population of human PBLs in the presence of murine SF- 25 MAb (Figure 5A).

Figure 6 shows that human PBLs armed with c-SF-25 MAb mediated cytotoxicity which is similar to ADCC (in the presence of 200-2,000 ng/ml of c-SF-25 MAb). Normal human PBLs did not produce cytotoxicity against the LS 180 tumor cells.

Figure 7 shows the cytotoxicity produced by induced human LAK cells at various E:T ratios compared to that produced by PBLs alone and PBLs in the presence of 20 μg/ml of c-SF-25 MAb as measured in a four hour ⁵¹Cr- release assay.

Figure 8. Biodistribution of human LAK cells in SCID mice. C.B-17 SCID mice were purchased from Taconic Farms (Germantown, NY). AH animal experiments were approved by the Committee on Research, Animal Care Protocols Review Group and were carried out according to Massachusetts General Hospital's guidelines. Mononuclear cells were isolated from healthy volunteers by gradient density centrifugation and were incubated in a plastic plate coated with fibronectin to prepare adherent cell-free human PBLs. LAK cells were generated by incubating these PBLs with 100U/ml of recombinant human IL-2 (Shionogi Pharmaceutical Co., Osaka, Japan) for 48 hours at 37°C. Their biodistribution was examined by injecting iv. human LAK cells radiolabeled with ³H-uridine and human IL-2 (500 U/mouse) into SCID mice. The radioactivity of each organ was examined after the tissues had been digested by Solvable (Dupont) and was expressed as % of injected dose.

Figure 9. SCID mouse/human colon cancer model. (Fig. 9A): Hepatic metastatic tumors established in SCID mice demonstrate tumor growth at 5 weeks after the injection of LS180 human colon adenocarcinoma cells into the portal vein. All mice developed large "cannon ball-like" tumors in their livers as indicated by the arrows. (Figure 9B): Survival curves of tumor- bearing mice from two independent experiments of 9 (open circles) and 10 animals (closed circles), respectively. In both experiments, all animals died within 7 weeks after tumor cell injection.

Figure 10. Cytotoxicity produced by human PBLs and LAK cells cross-linked with c-SF-25 MAb compared to that produced by human PBLs or LAK cells alone or at different effector/target cell ratios (E/T ratios).

Figure 11. Tumor growth after SF-25-LAK treatment. Tumors were carefully separated from surrounding normal liver following formaldehyde fixation and were evaluated by an independent investigator. The vertical axis denotes the weight of hepatic metastases (Hepatic Mets. Wt.). Each column represents the tumor weight from each mouse sacrificed 5 weeks after tumor cell injection. The statistical significance of tumor weight was examined by the two-sided Mann-Whitney U-test. (Fig. 11A): The effect of SF-25-LAK cells on the weight of hepatic metastases. Statistical significance: treatment with SF-25-LAK cells, LAK cells alone or LAK cells preincubated with c-SF- 25 MAb (preincubated LAK) vs. untreated mice (P < 0.01, P > 0.1 and P > 0.1, respectively); treatment with LAK cells alone or preincubated LAK cells vs. SF-25-LAK cells (P < 0.05 and P > 0.1, respectively); treatment with LAK cells alone vs. preincubated LAK cells (P > 0.1). Note that eight out of thirteen mice treated with SF-25-LAK cells were free of detectable hepatic metastases (P < 0.005: vs. untreated controls as determined by the chi-square test with Yates' correction). (Fig. 11B): The effect of human LAK cells cross-linked with nonspecific human IgG1 (human IgG1-LAK) on the weight of hepatic mestastases. Statistical significance: treatment with SF-25- LAK cells, LAK cells alone or untreated mice vs. treatment with human IgG1- LAK cells (P < 0.005, P > 0.1 and P > 0.1, respectively); treatment with LAK cells alone vs. untreated mice vs. treatment with SF-25-LAK (P < 0.05, respectively); treatment with LAK cells alone vs. untreated mice (P > 0.1). (Fig. 11C): Multiple Injections of SF-25-LAK cells on the weight of hepatic metastases. Statistical significance: single treatment or untreated mice vs. multiple treatments with SF-25-LAK cells (P > 0.1 and P < 0.005, respectively); single treatment with SF-25-LAK cells vs. untreated mice (P < 0.01).

Figure 12. The expression of SF-25 antigen in LS 180 tumor cells after SF-25-LAK treatment. (Figure 12A): Histological examination of the SF-25 antigen. The cryostat sections of hepatic metastases derived from treated (single or multiple injections of SF-25-LAK cells) or untreated SCID mice were incubated with ¹²⁵I-Iabeled SF-25 MAb and were autoradiographed as described before. The SF-25 antigen was highly expressed in both SF-25- LAK treated and untreated tumors as shown by the dense and homogenous radioactive spots. This reaction is specific as shown by its inhibition by "cold" unlabeled specific antibody and lack of inhibition by a nonrelevant MAb (B2TT: anti-tetanus toxoid MAb). (Figure 12B): Flow cytometric analysis of the SF-25 antigen. A single cell suspension of LS 180 cells was prepared from hepatic tumors derived from single or multiple SF-25-LAK treated and untreated mice. The expression of the SF-25 antigen at the cellular level was examined by flow cytometry using FITC-labeled SF-25 MAb. FITC-labeled nonrelevant MAb (B2TT) served as a negative control. The SF-25 antigen was uniformly expressed at high density in the majority of the cell populations from both untreated and treated mice.

Figure 13. Cytotolysis of LS 180 Tumor Cells by LAK Cells. A single cell suspension of LS 180 cell lines was prepared from the hepatic tumors derived from mice that have been injected with PBS (untreated) or SF- 25-LAK cells (single and multiple treatment). Cytolysis of these cells induced by human LAK cells at different effector to target ratios (E/T ratio) were compared to the original LS 180 cells. All LS 180 cell lines demonstrated similar sensitivity to human LAK cells as shown in Figure 13.

Figure 14. Complement-mediated cytotoxicity by c-SF-25 MAb was assayed using ⁵¹Cr-labeled LS 180 cells as target cells. Anti-LS 180 serum was produced in rabbits by injecting LS 180 cells with Freund's complete adjuvant and was used as a positive control. Complement-mediated cytolysis was induced by anti-LS 180 rabbit serum, but not by c-SF-25 as demonstrated in Figure 14.

VII. Definitions

In order to provide a clear and consistent understanding of the specification and claims, including the scope to be given such terms, the following definitions are provided. Any terms which are not specifically defined in this or other sections of this patent application have the ordinary meaning they would have when used by one of skill in the art to which this invention applies.

As used herein, a cytotoxic effector cell means a cell of the immune system which can injure or destroy invading microorganisms, tumors or other diseased tissue cells. This term is meant to include natural killer (NK) cells, activated NK cells, neutrophils, cytotoxic T-cells, eosinophils, basophils, B- cells, macrophages and lymphokine-activated killer (LAK) cells among other cell types.

As used herein, a targeted, cytotoxic, effector cell means a cytotoxic effector cell, as previously defined, to which antibodies, or effective fragments, or derivatives thereof, have been bound to form a stable complex, wherein said antibody, effective fragment, or derivative thereof, preserves its selective binding ability towards its antigen, after it has been bound to said cytotoxic effector cells. As used herein, the term armed cytotoxic effector cell means a cytotoxic effector cell, as previously defined, to which antibodies, or effective fragments or derivatives thereof, have been bound to form a stable complex, wherein said antibody, effective fragment, or derivative thereof, preserves its selective binding ability towards its antigen, after it has been bound to said cytotoxic effector cells.

As used herein, the term arming means the process wherein antibodies, or effective fragments or derivatives thereof, are bound to a cytotoxic effector cell, as previously defined, to form a stable complex, wherein said antibody, effective fragment or derivative thereof, preserves its selective binding ability towards its antigen, after it has been bound to said cytotoxic effector cell.

As used herein, an effective derivative or fragment of an antibody means a derivative or fragment of an antibody which is still capable of selectively binding to the same molecule(s) as that which the whole antibody binds to.

As used herein, a constitutive antigen means an antigen that is produced by the majority or all of the cells of a particular tumor type or disease type.

As used herein, a conjugating reagent means a chemical which can be used to bind the antibody, or an effective derivative or fragment thereof, to a cytotoxic effector cell. An example of a conjugating reagent is polyethylene glycol 8000.

As used herein, the term animal includes a human being.

As used herein, the term tumor includes a primary tumor, a metastases, metastatic tumor, a micrometastatic tumor or a micrometastases.

The term imaging means the visualization or location of cells, or tumors, or other diseased tissues which express constitutive antigens and which bind detectably labeled, targeted, cytotoxic, effector cells to them.

As used herein, a detectable label is an atom or molecule which is attached to the targeted, cytotoxic, effector cell or constituent thereof, and which is used in imaging cells or tumors or other diseased tissues. Examples of such labels include, but are not limited to, radϊoisotopic labels, non- radioactive isotopic labels, chemiluminescent labels, fluorescent labels and enzyme labels.

VIII. Description of the Preferred Embodiments

The present invention derives from the discovery that cytotoxic, immunological, effector cells, such as human NK-cells, macrophages and LAK cells, can be targeted to tumors, such as colon adenocarcinomas, their metastases, and other diseased tissues, which selectively express constitutive antigens. The targeting of the cytotoxic effector cells is accomplished by binding to them antibodies which are specific for the selectivity expressed antigens of the tumors or other diseased tissues. The present invention also derives from the discovery of a method to produce these targeted effector cells and methods to use them.

Among the cytotoxic effector cells which can be used in the present invention are cytotoxic T-cells, neutrophils, eosinophils, basophils, B-cells, macrophages, natural killer (NK) cells, activated NK cells, and lymphokine activated killer (LAK) cells, which list is not inclusive. Among the most preferred embodiments of the present invention are human NK cells, macrophages and LAK cells.

Among the types of antibody which can be used in the present invention are a polyclonal antibody, a monoclonal antibody, a chimeric monoclonal antibody, a humanized chimeric monoclonal antibody, and a mouse-human chimeric antibody, which list is not inclusive. Among the preferred embodiments is a mouse-human chimeric antibody.

Among the effective antibody derivatives and antibody fragments which can be used in the present invention are Fab fragments of murine monoclonal antibodies, Fc fragments of human immunoglobulin, F(ab)₂ fragments, Fv fragments, and single chain antibody binding proteins, which list is not inclusive. Among the most preferred embodiments of the present invention are Fab fragments, and F(ab')₂ fragments. Among the cytokines which can be used in the present invention are interieukin-1 (IL-1), interleukin-2 (IL-2), interIeukin-3 (IL-3), interleukin-4 (IL-4), interleukin-5 (IL-5), interleukin-6 (IL-6), interleukin-7 (IL-7), α- interferon, β-interferon, γ-interferon, macrophage colony stimulating factor, granulocyte-macrophage colony stimulating factor, natural killer cell stimulating factor, and macrophage activating factor, which list is not inclusive. Among the most preferred embodiments of the present invention is interleukin-2.

Among the conjugating reagents which can be used in the present invention are polyethylene glycol (PEG) 200, PEG 400, PEG 600, PEG 1500, PEG 4000, PEG 6000, mixtures of polyethylene glycols of various molecular weights, biotin-N-hydroxysuccinimϊde, and N-succimidil-3-(2- pyridyIdϊthio)propϊonate (SPDP), which list is not inclusive. Among the most preferred embodiments of the present invention is polyethylene glycol 8000.

An effective dose range for using the cells of the present invention is from 10⁶ to 10¹² cells per dose.

Among the routes of administration which can be used in the present invention are the intravenous, intraarterial, intramuscular, intraperitoneal, intrapericardϊal, intradermal, transdermal, intrape.vic, intrapharyngeal, intranasal, intrapleural, intravaginal, mtravesicular, intrasplenic, intrathecal, ϊntraurethal, intraureteral, inrraprostatic, intrapulmonary, intrarenal, intrascrotal, intraspinal, intrauterine, rectal, oral, subcutaneous and intrarachidian which list is not inclusive. Among the most preferred embodiments which can be used in the present invention are intravenous, intramuscular, intraperitoneal and subcutaneous routes.

Among the diseases that can be imaged or treated by the armed cytotoxic effector cells of the present invention are colon adenocarcinoma, hepatocellular carcinoma, cholangiocellular carcinoma, gastric adenocarcinoma, rectal adenocarcinoma, breast adenocarcinoma, pancreatic adenocarcinoma, bladder adenocarcinoma, squamous cell carcinoma of the lungs, adenocarcinoma of the lungs, large cell carcinoma of the lungs, small cell carcinoma of the lungs, Iymphoproliferative disease, myeloprolϊferative disease, lymphoma, leukemia, kidney carcinoma, ovary adenocarcinoma, cervical carcinoma, uterine endometrial adenocarcinoma, liver hepatoma, choriocarcinoma, malignant melanoma, including the primary tumors, metastases and micrometastases of these diseases, which list is not all inclusive. Among the most preferred diseases for treatment with the present invention are colon adenocarcinoma, and hepatocellular carcinoma.

Among the antigens to which the antibodies, fragments or derivatives thereof, which are coupled to the armed, cytotoxic effector cells of the present invention can specifically bind are the SF-25 antigen, the XF-8 antigen, the AF-20 antigen, the carcinoembryonic (CEA) antigen, the K-314 antigen, the V-215 antigen, the CA19-9 antigen, the CO29.11 antigen, the DU-PAN-2 antigen, the TAG-72 antigen, and the LEA antigen, which list is not inclusive. Among the preferred embodiments are the SF-25, XF-8 and AF-20 antigens. Among the most preferred embodiments is the SF-25 antigen. A. Diagnostic Uses of Targeted, Cytotoxic Effector Cells

The targeted, cytotoxic, effector cells of the present invention are particularly suited for in vivo and in vitro imaging of certain tumors and other diseased tissues

Antibodies, or fragments thereof, may be labeled using any of a variety of labels and methods of labeling. Examples of types of labels which can be used in the present invention include, but are not limited to, enzyme labels, radioisotopic labels, non-radioactive isotopic labels, fluorescent labels, toxin labels, chemiluminescent labels, and nuclear magnetic resonance contrasting agents.

Examples of suitable enzyme labels include malate dehydrogenase, staphylococcal nuclease, delta-5-steroid isomerase, yeast-alcohol dehydrogenase, alpha-glycerol phosphate dehydrogenase, triose phosphate isomerase, peroxidase, alkaline phosphatase, asparaginase, glucose oxidase, beta-galactosidase, ribonuclease, urease, catalase, glucose-6-phosphate dehydrogenase, glucoamylase, acetylcholine esterase, etc. Examples of suitable radioisotopic labels include ³H, ¹¹¹In, ¹²⁵I, ¹³¹I, 3²P, ³⁵S, ¹⁴C, ⁵¹Cr, ⁵⁷To, ⁵⁸Co, ⁵⁹Fe, ⁷⁵Se, ¹⁵²Eu, ⁹⁰Y, ⁶⁷Cu, ²¹⁷Ci, ²¹¹At, 2¹²Pb, ⁴⁷Sc, ¹⁰⁹Pd, ¹⁸⁶Re, ^99mTc, ⁶⁷Ga, ²¹²Bi, ⁷⁷Br, ¹⁵³Sm, ³²P, ^{11 1}Ag, etc. 1¹¹In is a preferred isotope. Its use may have substantial advantages since its avoids the problem of dehalogenation of the ¹²⁵I or ¹³¹I-labeled monoclonal antibody by the liver. In addition, this radionucleotide has a more favorable gamma emission energy for imaging (Perkins et al., Eur. J. Nucl Med. 10:296-301 (1985); Carasquillo etal., J. Nucl Med. 28:281-281 (1987)). For example, ^{11 1}In coupled to monoclonal antibodies with 1-(P- isothiocyanatobenzyO-DPTA have shown little uptake in non-tumorous tissues, particularly the liver, and therefore enhanced specificity of tumor localization (Esteban et al. , J. Nucl Med. 28:861-810 (198/)).

Examples of suitable non-radioactive isotopic labels include ¹⁵⁷Gd, 5⁵Mn, ¹⁶²Dy, ⁵²Tr, ⁵⁶Fe, etc.

Examples of suitable fluorescent labels include an ¹⁵²Eu label, a fluorescein label, an isothiocyanate label, a rhodamine label, a phycoerythrin label, aphycocyanin label, an allophycocyanin label, an o-phthaldehyde label, a fluorescamine label, etc.

Examples of suitable toxin labels include diphtheria toxin, ricin, and cholera toxin, etc.

Examples of chemiluminescent labels include a luminal label, an isoluminal label, an aromatic acridinium ester label, an imidazole label, an acridinium salt label, an oxalate ester label, a Iuciferin label, a luciferase label, an aequorin label, etc.

Examples of nuclear magnetic resonance contrasting agents include heavy metal nuclei such as Gd, Mn, iron, etc.

Those of ordinary skill in the art will know of other suitable labels which may be employed in accordance with the present invention. The binding of these labels to the targeted cytotoxic, effector cells or the antibodies, derivates, or fragments thereof, can be accomplished using standard techniques commonly known to those of ordinary skill in the art. Typical techniques are described by Kennedy et al (Gin. Chim. Ada 70: 1-31 (1976)), and Schurs et al. (Clin. Chim. Acta 81:1-40 (1977)). Coupling techniques mentioned in the latter are the glutaraldehyde method, the periodate method, the dimaleimide method, the m-maleimidobenzyl-N-hydroxy- succinimide ester method, all of which methods are incorporated by reference herein.

The detection of cells which express a targeted antigen may be accomplished by the use of in vivo imaging techniques, in which the labeled, targeted, cytotoxic, effector cells are administered to a patient or other animal, and the presence of tumors or other diseased tissues expressing the antigen is detected without the prior removal of any tissue sample. Such in vivo detection procedures have the advantage of being less invasive than other detection methods, and are, moreover, capable of detecting ώe presence of antigen- expressing cells in tissue which cannot be easily removed from the patient.

The targeted, cytotoxic, effector cells of the present invention are also particularly suited for use in vitro to detect cells which express the targeted antigen in body tissue, fluids (such as blood, lymph, etc.), stools, or cellular extracts. In such methods, the targeted, cytotoxic, effector cells may be utilized in a liquid phase or bound to a solid-phase carrier.

Alternatively, the detection of cells which express the targeted antigen may be accomplished by removing a sample of tissue from a patient or other animal and then treating the isolated sample with any of the suitably labeled, targeted, cytotoxic effector cells of the present invention. Preferably, such in vitro detection is accomplished by removing a histological specimen from a patient or other animal, and providing the labeled, targeted, cytotoxic effector cells of the present invention to such specimen by applying them or by overlaying them onto a sample of tissue. Through the use of such a procedure, it is possible to determine not only the presence of the targeted antigen, but also the distribution of the antigen on the examined tissue. Using the present invention, those of ordinary skill in the art will readily perceive that any of a wide variety of histological methods (such as staining procedures) can be modified in order to achieve such in vitro detection. The detection of the targeted antigen can be improved through the use of carriers. Well-known carriers include glass, polystyrene, polypropylene, polyethylene, dextran, nylon, amylases, natural and modified celluloses, polyacrylamides, agaroses, and magnetite. The nature of the carrier can be either soluble, to some extent, or insoluble for the purposes of the present invention. The support material may have virtually any possible structural configuration so long as the coupled, labeled, targeted, cytotoxic, effector cell is capable of binding to the targeted antigen. Thus, the support configuration may be spherical, as in a bead, or cylindrical, as in the inside surface of a test tube, or the external surface of a rod. Alternatively, the surface may be flat such as a sheet, test strip, etc. Those skilled in the art will note many other suitable carriers for biding the labeled, targeted, cytotoxic effector cell, or will be able to ascertain the same by use of routine experimentation.

B. Diagnostic Uses of Targeted, Cytotoxic Effector Cells In vitro or in vivo detection methods may be used in the diagnosis of certain cancers such as colon adenocarcinoma, or other diseases which express a constitutive antigen. Additionally, such detection methods may be used to assist in the determination of the stage of a malignancy or other disease, or to determine whether an individual possesses malignant or other lesions which may be obscured (or whose detection may be complicated) by the close association of normal tissue.

One especially preferred use for the targeted cytotoxic effector cells of the present invention is as an aid in the diagnosis of colon cancer in patients who present with symptoms of inflammatory bowel diseases, and in particular, ulcerative colitis or intestinal polyps. Using the methods of the prior art, the early diagnosis and detection of colon cancer in individuals suffering from such inflammatory bowel disease is often complicated, or masked, by the symptoms of bowel disease. Thus, concern that an occult colorectal carcinoma may be present in an individual suffering from inflammatory bowel disease may result in a recommendation that such individuals submit to a colectomy. Because the targeted cells of the present invention are capable of identifying colorectal carcinomas, they can be used to determine the presence of otherwise occult lesions. Thus, their use in the diagnosis of the cause and severity of inflammatory bowel disease and colorectal carcinoma is capable of preventing unwarranted colectomies, and is, therefore, highly desirable.

As used herein, an effective amount of targeted, cytotoxic effector cells is one capable of achieving the desired diagnostic discrimination. The amount of such cells which are typically used in a diagnostic test are generally between 10⁶ to 10¹². C. Therapeutic Uses of Targeted, Cytotoxic Effector Cells

In addition to providing a method for diagnosing and imaging cancers and other diseased tissues, the present invention also provides a means for preventing the onset of such cancers, and for treating animals with cancer, including humans. The discovery that the SF-25 antigen is expressed as a constitutive antigen on cancer cells, of endodermal origin and the invention of targeted, cytotoxic effector cells capable of binding to this and other constitutive antigens provides a means for preventing and treating these cancers.

The ability to conjugate toxins with antibodies, or fragments of antibodies, and to the cells of the present invention provides an additional method for treating tumors and other diseased tissues (Dillman, R.O., Ann. Int. Med. 111:592-603 (1989)). In this embodiment, antibodies, or fragments of antibodies which are capable of recognizing the SF-25 antigen, or other constitutive antigens, are conjugated with cytotoxic molecules and cytotoxic effector cells and administered to a patient suspected of or having a tumor. Alternatively, the toxin can be conjugated directly to the cytotoxic effector cell independently of the binding of the antibody to the cell. When such a toxin- derivatized targeted cytotoxic, effector cell binds to a cancer or other diseased cell, the toxin moiety will cause the death of the cancer or diseased cell. Any of a variety of toxin molecules may be employed to produce such toxin-conjugated, targeted, cytotoxic effector cells. Examples of suitable cytotoxic molecules include: ricin; diphtheria, pseudomonas, and cholera toxins; TNF, etc. Toxins conjugated to antibodies or other ligands are known in the art (see, for example, Olsnes et al., Immunol Today 10:291-295 (1989)).

Additional types of therapeutic moieties including, but not limited to, radionuclides and cytotoxic drugs and other agents, can be conjugated to the targeted, cytotoxic, effector cells of the present invention to treat cancer patients. Examples of radionuclides which can be coupled to the cells of the present invention and delivered in vivo to sites of antigen include ²¹²Bi, ¹³¹I, 1⁸⁶Re, ¹⁸⁸Re, ⁹⁰Y, ⁶⁷Cu, ¹⁵³Sm, ^114mIn, ³²P, ^{11 1}Ag, ²¹¹At, ²¹²Bi, ²¹²Pb, ¹²⁵I, and ⁷⁷Br, which list is not intended to be exhaustive. The radionuclides exert their cytotoxic effect by locally irradiating the cells, leading to various intracel-ular lesions, as is known in the art of radiotherapy.

Additional types of therapeutic moieties which can be used in the present invention include cytotoxic drugs which interfere with critical cellular processes including DNA, RNA, and protein synthesis. For a fuller exposition of these classes of drugs which are known in the art, and their mechanisms of action, see Calabresi, P., et al. , Chemotherapy of Neoplastic Diseases (pp. 1202-1208) and Antineoplastic Agents (pp. 1209-1263), both in Goodman and Gilman's THE PHARMACOLOGICAL BASIS OF THERAPEUTICS (Gilman, A.G., et al. , eds.), 8th Edition, Pergamon Press, New York, NY (1990).

The targeted cytotoxic effector cells of the present invention can be advantageously utilized in combination with lymphokines or hemopoietic growth factors, etc., which serve to increase the number or activity of effector cells.

As would be understood by one of ordinary skill in the art, such compositions may contain salts, buffers, adjuvants, or other substances which are desirable for improving the efficacy of the composition. Adjuvants are substances that can be used to specifically augment a specific immune response. Normally, the adjuvant and the composition are mixed prior to presentation to the immune system, or presented separately, but into the same site of the animal being immunized. Adjuvants can be loosely divided into several groups based upon their composition. These groups include oil adjuvants (for example, Freund's complete and incomplete), mineral salts (for example, AlK(SO₄)₂ , AINa(SO₄)₂ , AINH₄(SO₄), silica, kaolin, and carbon), polynucleotides (for example, poly IC and poly AU acids), and certain natural substances (for example, wax D from Mycobacterium tuberculosis, as well as substances found in Corynebacterium parvum, or Bordetella pertussis, and members of the genus Brucella). Among those substances particularly useful as adjuvants are the saponins such as, for example, Quil A. (Superfos A/S, Denmark). Examples of materials suitable for use in vaccine compositions are provided in Remington's Pharmaceutical Sciences (Osol, A., Ed., Mack Publishing Co., Easton, PA, pp. 1324-1341 (1980)).

Treatment of an individual with a tumor bearing the constitutive antigen recognized by the targeted cytotoxic effector cells of this invention comprises parenterally administering a single dose, multiple doses or an infusion of these cells to a patient or other animal. The effective dose is a function of the individual, targeted, cytotoxic, effector cells, the presence and nature of a conjugated therapeutic agent (see above), the patient and his clinical status, and can vary from about 1 ng/kg body weight to about 1 g/kg body weight.

In addition to pharmacologically active compounds, these cell preparations may contain suitable pharmaceutically acceptable carriers. The therapeutic compositions of the present invention can be administered parenterally by injection. Preferably the preparations for parenteral administration contain from about 0.01 to 99 percent, preferably from about

25 to 75 percent of the cells of the this invention.

The pharmaceutical preparations of the present invention are manufactured in a manner which is itself known.

Compositions for parenteral administration include sterile aqueous or non-aqueous solutions, suspensions, and emulsions.

Suitable formulations for parenteral administration include aqueous solutions of the active cells. In addition, suspensions of the active cells may be administered. Aqueous injection suspensions may contain substances which increase the viscosity of the suspension and include, for example, sodium carboxymethyl cellulose, sorbitol, and/or dextran. Optionally, the suspension may also contain stabilizers.

Examples of non-aqueous solvents are propylene glycol, polyethylene glycol, vegetable oils such as olive oil, and injectable organic esters such as ethyl oleate. Carriers or occlusive dressings can be used to increase skin permeability and enhance absorption.

According to the present invention, an "effective amount" of a therapeutic composition is one which is sufficient to achieve a desired biological effect. Generally, the dosage needed to provide an effective amount of the composition, and which can be adjusted by one of ordinary skill in the art, will vary depending upon such factors as the animal's or patients age, condition, sex, and extent of disease, if any, and other variables.

The cells of the invention can be administered by either single or multiple dosages or infusion of an effective amount. Effective doses of the cells of the present invention can vary from 10⁶ to 10¹² cells per dose.

Having now generally described the invention, the same will be more readily understood through reference to the following methods and examples, which are provided by way of illustration, and are not intended to be limiting of the present invention, unless specified.

D. Production of Targeted Cytotoxic Effector Cells

1. Production of SF-25 Monoclonal Antibodies

The production of monoclonal antibodies to the SF-25 antigen has been extensively described in U.S. Patent Application Serial No. 07/203, 198 which was filed on June 7, 1988. The contents of this patent application are incorporated herein by reference. 2. Production of the Chimeric SF-25 Monoclonal Antibodies

A human-mouse, chimeric immunoglobulin gene was constructed by joining L and H chain V genes isolated from SF-25 secreting hybridoma cells to human Kappa and gamma 1 C region genes. This construct was then transfected into Sp 2/0 myeloma cells to produce the humanized antibody. The production of the chimeric SF-25 MAbs was extensively described in related United States Patent Application entitled "Chimerized SF-25 Antibodies With Specificity For The Human Tumor SF-25 Antigen, And Methods For Their Production And Use" U.S. Patent Application Serial No. 07/765,612, and which is herein incorporated by reference. 3. Purification of Human PBLs and Induction of LAK Cells

Venous blood was drawn from healthy volunteers using heparin (200 U/ml final concentration) and was overlain on Ficoll-Paque (Pharmacia LKB Biotechnology, Inc., Piscataway, NJ). Mononuclear cells were isolated by centrifugation (1,200 r.p.m. for 30 minutes at 25°C) and were washed three times with RPMI 1640 medium. Mononuclear cells (5 x 10⁶/ml) were incubated on a plastic plate coated with fibronectin (Freundlish et al. , J. Immun. Meth. 62:31-37 (1983)) for 40 minutes at 37°C in a CO₂ incubator. Non-adherent cells were collected and used as human PBLs. Human LAK cells were obtained by incubating these human PBLs with recombinant human interleukin-2 (rIL-2; 1-10,000 U/ml) for 48 hours to 30 days at 37°C.

4. Binding of the Antibodies to the

Cytotoxic Effector Cells

Effector cells were armed with murine or chimeric SF-25 MAb or F(ab')₂ fragments of SF-25 MAb using PEG 8000 as follows. Mononuclear cells were incubated with 2 mg/ml of murine or chimeric SF-25 MAb or F(ab')₂ fragments thereof for 30 minutes at 4°C. An equal volume of 30% PEG 8000 was added to this cell mixture and it was then incubated for an additional 90 minutes at 4°C. The cells were extensively washed with PBS.

Human effector cells were armed with 1-3 X 10⁶ molecules of chimeric SF-25 (c-SF-25) MAb/cell. In addition, other methods are also available to arm effector cells with SF-25 MAb such as biotmylation of said cells with biotin-N-hydroxysuccinimide followed by conjugation with avidin linked MAbs (Lo et al., Nature 310:792-794 (1984)).

E. In Vitro Binding of Targeted Effector Cells to LS 180 Tumor Cells The ability of human mononuclear cells armed with SF-25 MAbs to bind to LS 180 tumor cells via the Fab fragment was measured by two different methods: in situ autoradiography and a cytotoxicity assay.

1. Autoradiography

The binding of effector cells armed with chimeric and murine SF-25 MAbs as well as F(ab')₂ fragments of chimeric SF-25 MAb were examined by autoradiography as follows. SF-25 MAb armed, radiolabeled, effector cells were added to LS 180 cells grown in a 2-chamber LabTek Chamber Slide (Nunc). After a 1 hour incubation at 4ºC, the slides were washed extensively with PBS and fixed with 4% formaldehyde. Autoradiography was performed as before (Takahashi et al. , Gastroenterology 96:1317-1329 (1989); Takahashi et al., Hepatology 9:625-634 (1989)). In brief, the slides were stained with eosin and then coated with Kodak autoradiography emulsion type NTB 3 for ⁵¹Cr, or NTB 2 for ³H. The slides were then exposed at 4°C for 1 to 10 days in the dark and were developed using Kodak Dektol, D19. The cells were counterstained with hematoxylin. 2. In Vitro Cytotoxicity Assay

The cytotoxicity of SF-25 MAb targeted effector cells was measured using ⁵¹Cr-labeled LS 180 cells as follows. Confluent LS 180 cells were harvested with EDTA/Versene buffer and radiolabeled by incubating 1 X 10⁶ cells with 100 μCi of sodium ⁵¹Cr chromate for 30 minutes at 37°C. After washing, the concentration of the radiolabeled LS 180 cells was adjusted to 1 X 10⁵/ml of RPMI 1640.

One hundred μl of LS180 cells and 100 μl of various concentrations of effector cells (human PBLs or LAK cells) were pipetted into 96-well U bottom plates. For ADCC assays 50 μl of c-SF-25 MAb (100 μg/ml in RPMI 1640) was added to each well. Plates were incubated in a CO₂ chamber at 37°C for 4 hours. After centrifugation of the U bottom plates at 1,500 rpm for 15 minutes, 200 μl of culture supernatant was collected and the radioactivity was determined by a gamma well counter. The percent specific lysis was determined by the following formula: - x 100 = % specific lysis.

The spontaneous release of ⁵¹Cr was measured after incubation of LS

180 cells alone with RPMI 1640. The total releasable counts were determined after incubation of LS 180 cells in 1.0 N HCl. In all experiments, results were considered valid only if the measure of spontaneous release was less than 10% of the total releasable counts.

Production of SCID Mice with Human Colon Adenocarcinoma Hepatic Metastases

Hepatic tumors were established using four to five week old female athymic, C.B-17 SCID mice maintained under specific-pathogen-free- conditions. The mice were anesthetized with 0.4 ml of 2 % chloral hydrate ip. and were placed in the decubitus position. Their spleens were exposed by transverse incisions which were made in the left flanks through the skin and peritoneum. The mice were injected with 1.0 x 10⁶ LS 180 cells and 20 μg of anti-asialo GMl rabbit serum (Kasai et al. , Eur. J. Immunol 10:175-180 (1980)) in 0.25 ml PBS into the portal vein via the splenic hilus using a 27- gauge needle. The portal vein was ligated one minute later and the spleen was removed to prevent bleeding and subsequent dissemination of tumor cells into the abdominal cavity. G. Testing Targeted Human Cytotoxic Effector Cells in SCID

Mice with Hepatic Metastases.

1. Human PBLs as the Targeted Cells

³H-Iabeled human PBLs armed with c-SF-25 were injected into SCID mice 20 days after the mice had been injected with LS180 tumor cells. These mice had established human-adenocarcinoma-derived hepatic metastases. Control mice were injected with ³H-labeIed normal human PBLs. All mice were sacrificed 1, 3, 6 or 24 hours after the PBL injections. The presence of effector cells in the murine tumors were detected by in situ autoradiography as previously described. 2. Human LAK Cells as the Targeted Cells c-SF-25 MAb armed human LAK cells were injected iv. into SCID mice (10⁷ cells/mouse) 5 days after tumor cell injection. The same number of human LAK cells alone were injected into a group of control mice. A second group of control mice received no effector cell treatment. Hepatic tumor growth was examined at 35 days in all three groups. Example 1

Radiolabeling Of Effector Cells

Radiolabeling of human effector cells was necessary to examine the biodistribution and targeting of these cells. Radiolabeling of human PBLs was compared using five different radionucleotides: ³H-uridine; ³H-thymidine; 1¹¹Inoxyquinoline (oxine); ¹²⁵I-iodine and ⁵¹Cr-chromate (Mantovani et al., Int. J. Cancer. 23:18-27 (1979); Danpure et al. , Brit. J. Radiology 54:591- 601 (1981); Soule et al. , Int. J. Cancer 29:331-344 (1982); Miller et al., Cell Immunol 14:284-302 (1984)). The specific activity and viability of human PBLs after radiolabeling was determined first. The specific activity of lymphocytes labeled with ³H-thymidine was very low (Table 1). Although 1¹¹In-oxine and ¹²⁵I-iodine produced higher specific activity, the spontaneous release of radioactivity was unacceptably high by PBLs incubated at 37°C in the culture medium for ^{11 1}In-oxine and ¹²⁵-iodine compared to ³H-uridine. Therefore, ³H-uridine and ⁵¹Cr-chromate were used to radiolabel human effector cells in the following experiments.

Most human PBLs labeled with ³H-uridine were covered with black silver grains (Figure 1) demonstrating that such cells could be detected by autoradiography. Example 2

Development of a Hepatic Metastatic Model of Human Colon Adenocarcinoma in SCID Mice.

Hepatic metastases of human colon adenocarcinoma cells were established by injecting human colon adenocarcinoma derived LS 180 cells into the portal vein of C.B-17 SCID mice as previously described. Large "cannon ball-like" hepatic metastases were established at five weeks. Local extrahepatic abdominal tumors were also present at the site of tumor cell injection. All mice died five to seven weeks after tumor cell injection from hepatic dysfunction caused by extensive parenchymal involvement.

Example 3

Survival and Biodistribution of

Human PBLs in SCID Mice.

The survivability of human PBLs in SCID mice was determined by flow cytometric analysis using an antibody to a human leucocyte antigen (HLe- 1/CD45) that is present on all human leucocytes (Rowe et al. , Gin. Exp. Immunol. 45:290-298 (1981)). Blood was collected from the mice by axillary bleeding at various times after the iv. administration of the PBLs. Human PBLs rapidly disappeared from the circulation of the SCID mice (Figure 2).

3H-uridine labeled human PBLs were injected iv. into SCID mice without hepatic tumors to determine their tissue distribution and survivability following clearance from the circulation. The specific uptake of human PBLs in each organ was defined as the cpm per gram of tissue (Figure 3A). The biodistribution among various organs was defined by total radioactivity per organ (Figure 3B). Although the majority of PBLs were detected in the lung one hour after injection, high uptake in the liver and spleen was observed following clearance from the lungs. A similar pattern of biodistribution of injected PBLs was observed in SCID mice with established hepatic metastases of human colon adenocarcinoma cells (data not presented). Since human effector cells were minimally present in the liver after one week, targeted cells were injected on a weekly basis in all subsequent experiments.

Example 4

Binding of c-SF-25 MAb to Human PBLs

or Human LAK Cells

The binding of c-SF-25 MAb to human PBLs was performed with polyethylene glycol (PEG) as modified from Jones et al. (J. Immunol. 125:926-933 (1980)), and as described in sections VIII; D., 3 and 4 herein. Two different PEG preparations (MW 8,000 and 20,000) were used. Using ¹²⁵I-labeled c-SF-25 MAb the total number of antibody molecules bound to the PBL cells was determined to be about 10⁶ molecules of c-SF-25 MAb per PBL cell (Table 2).

These PBL cells were incubated at 37°C in antibody free culture medium to determine the stability of the MAb binding. At 48 hours post MAb treatment, 2-3 x 10⁴ MAb molecules/cell were still associated with PEG- treated human PBLs. In contrast, antibodies dissociated from PBLs if there was no PEG treatment (PEG-; Table 2). Therefore PEG 8000 was used to bind c-SF-25 MAb to effector cells in the remaining experiments.

Similarly, the number of c-SF-25 MAbs bound per human LAK cell was 10 times higher and lasted significantly longer when the MAbs were cross-Iinked to the LAK cells by the PEG method than by mere preincubation with the LAK cells (Figure 4).

Example 5

Functional Analysis of Human Effector Cells Isolated, purified, populations of human NK-cells and macrophages induced substantial ADCC against human colon adenocarcinoma derived LS 180 cells in the presence of c-SF-25 MAb (Figure 5B) that was greater than that induced by a mixed cellular population of PBLs (Figure 5A).

Macrophage-depleted human PBLs to which were armed with c-SF-25 MAbs exhibited strong cytotoxicity against LS 180 tumor cells as measured in a 4-hour ⁵¹Cr release assay using an E:T ratio of 50:1 (Figure 6). The specific lysis was very similar to the ADCC exhibited by human PBLs in the presence of 200-2,000 ng/ml of c-SF-25 MAb (Figure 6). Normal human PBLs did not show cytotoxicity in the absence of antibody in the medium. These data demonstrate that the Fab portion of c-SF-25 MAb is available to bind to LS 180 tumor cells after their attachment to effector cells by PEG 8000 treatment.

SCID mice bearing hepatic tumors were injected iv. with 10 X 10⁶ human PBLs alone or human PBLs armed with c-SF-25 MAb. The mice were sacrificed 3 hours later and the hepatic tumors were examined for the presence of human PBLs using an anti-human CD45 antibody and immunofluorescent staining. The armed human PBLs were detected in the hepatic tumors while the PBLs injected alone were not detected in the hepatic tumors. Example 6

Induction of Human LAK Cells

LAK cells were induced by incubating macrophage depleted human PBLs with 100 U/ml of rIL-2 for 48 hours (Ortaldo et al. , J. Exp. Med. 164:1193-1205 (1986); Phillips et al., Id. 164:814-825 (1986); Rosenberg et al., New Engl. J. Med. 516:889-897 (1987)). These induced human LAK cells demonstrated strong cytotoxicity against LS 180 tumor cells at different E:T ratios (Figure 7, left panel). Although unstimulated PBLs did not show cytotoxicity (Figure 7, middle panel) they produced ADCC in the presence of c-SF-25 MAb (Figure 7, right panel)

As shown by in situ autoradiography, human LAK-cells armed with c-

SF-25 MAb as well as murine SF-25 and F(ab')₂ fragments of c-SF-25 MAb bound to LS 180 cells. In contrast, LAK cells alone did not bind to LS 180 cells under these conditions. These data indicate the importance of the Fab fragment of SF-25 MAb for targeting human LAK cells.

Example 7

In vivo Effect of Human LAK

Cells Targeted With c-SF-25 MAb

All untreated SCID mice which had been injected with LS 180 tumor cells as before, developed hepatic and local abdominal tumors 5 weeks later.

Large multiple tumors were present throughout their livers.

In contrast, SCID mice had substantially inhibited tumor growth if they had been administered 1 x 10⁷ LAK cells armed with 1-3 x 10⁶ molecules of c-SF-25 MAb. In some cases these latter mice had no detectable hepatic tumors. SCID mice which were injected with human LAK cells alone (1 x

10⁷ cells/mouse) developed hepatic tumors just like the untreated controls.

These data demonstrate a potent inhibitory effect by targeted human LAK cells on human colon adenocarcinoma metastatic growth to the liver.

Example 8

Inhibition of Human Colon Cancer Growth By Antibody-Directed Human LAK Cells in SCID Mice

The short term survival and biodistribution of human LAK cells radiolabeled with ³H-uridine was examined in SCID mice after their iv. injection. As demonstrated in Fig. 8, human LAK cells were detectable for up to 14 days and the majority of cells resided in the liver after being temporarily trapped in the lung.

Human colon cancer has been successfully established in the liver of SCID mice (SCID/human colon cancer model) by injecting LS 180 human colon adenocarcinoma ceils into the portal vein via the spleen (Takahashi et al. , Gastroenterology 96: 1317 (1989)). Fig. 9A is a representative example of hepatic tumors at 5 weeks after the injection of LS 180 cells. Although this animal model does not represent all the steps of the metastatic cascade, it simulates the metastatic growth of tumor cells in the liver after their vascular spread. All mice eventually die from extensive hepatic parenchymal involvement of tumors within 6-7 weeks after tumor cell injection as shown in Fig. 9B.

In order to assess if SF-25 MAb would augment the in vivo anti-tumor effect of human LAK cells without interfering with the murine immune system in this animal model, a novel approach was established. We have cross-linked SF-25 MAb to human LAK cells using polyethylene glycol (PEG) according to a modified method of Jones et al. , J. Immunol 125:926 (1980). This PEG technique is not toxic to effector cells and cross-links the MAb to the cell surface of lymphocytes in a random orientation. In this regard, we have employed a recombinant chimeric antibody construct of SF-25 MAb (c-SF-25 MAb) that has the variable region of murine SF-25 MAb and the constant region of human IgG1. The major population of LAK cells expresses a Fc receptor for IgG (FcγR type III: CD16) Ortaldo et al. , J. Exp. Med. 164:1193 (1986); Phillips et al. , Ibid. 164:814 (1986); and Roberts et al. , Cancer Res. 47:4366 (1987)). Human IgG1 binds to this FcγR (Unkeless, J.C., J.Clin. Invest. 83:355 (1989); and Ravetch et al. , Annu. Rev. Immunol. 9:451 (1991)). Therefore, the Fab fragment of c-SF-25 MAb is more likely to be exposed in an outward orientation by the binding of the Fc fragment to FcγR on the cell surface of LAK cells. PEG treatment enhances the binding of MAb to LAK cells.

The cross-linking of human LAK cells with c-SF-25 MAb was performed as follows. LAK cells were incubated with 2 mg/ml of c-SF-25 MAb in RPMI 1640 for 30 min at 4°C. An equal volume of 30% PEG 8000 (Sigma) in RPMI 1640 was added to this cell mixture and incubated for an additional 90 min at 4°C. Finally, the cells were washed with PBS three times. As a control, human LAK cells were preincubated with the same concentration of c-SF-25 MAb (2mg/ml) for 90 min at 4°C and then were washed three times.

Ten times more antibodies were cross-linked to LAK cells by PEG treatment than by preincubation with the c-SF-25 MAb alone (approximately 0.1 and 0.01 pico mole of MAb per 10⁷ LAK cells, respectively). Since NK- cells in human PBLs express FcγR type III and initiate antibody-dependent cell-mediated cytotoxicity (ADCC) upon binding to the antibody (Vivier et al. , J. Immunol. 146:206 (1991)), the availability of Fab fragments to combine with target cells was examined functionally using the previously described cytotoxicity assay of macrophage-depleted human peripheral blood lymphocytes (PBLs). As demonstrated in panel A of Fig. 10, human PBLs cross-Iinked with c-SF-25 MAb (SF-25-PBL) exhibited strong cytotoxicity against LS 180 cells. In contrast, human PBLs alone did not do so under the same conditions (panel B of Fig. 10). ADCC requires simultaneous binding of the Fab fragment of antibody to antigen and the Fc fragment to FcγR (Id.). Therefore, these results confirm that the Fab portion of c-SF-25 MAb is functionally available to bind to LS 180 tumor cells following cross-linking of the antibody via the Fc fragment.

Human LAK cells demonstrated strong cytotoxicity (panel D of

Fig. 10). LAK cells cross-linked with c-SF-25 MAb (SF-25-LAK) also exhibited strong cytotoxicity, but no greater than that of LAK cells alone

(panel C of Fig. 10). However, when these SF-25-LAK cells were radiolabeled with ³H-uridine and were injected into the SCID/human colon cancer model, an increased tumor uptake was observed. Table 4 demonstrates the uptake of SF-25-LAK cells into LS 180 hepatic tumors as expressed by percent injected dose per gram of tumor tissue. The increased tumor uptake of SF-25-LAK was statistically significant and was about 3 times higher than that of LAK cells alone (Table 4).

Table 4. Tumor uptake of LAK cells cross-linked with c-SF-25 MAb.

*Human LAK cells were radiolabeled with ³H-uridine and iv. injected into SCID mice bearing LS 180 hepatic metastases. Tumor uptake was determined by measuring radioactivity in tumors dissected from

surrounding normal liver and expressed as % injected dose per gram of tumor tissue. Four to six animals were used to determine each value.

**Hours after the iv. injection of ³H-uridine-labeled effector cells.

***Statistical significance was examined by two-sided Student's t-test. Five days following the injection of LS 180 cells into the portal vein, and at a time when micrometastatic disease was present, SF-25-LAK cells, LAK cells alone or LAK cells preincubated with c-SF-25 MAb were injected iv. into SCID mice (10⁷ cells with 500 U IL-2 in 100 μl PBS/mouse). Fig. 11A summarizes the results of these experiments and demonstrates the actual weight of hepatic tumor from each mouse sacrificed at 5 weeks following tumor cell injection. All untreated mice developed LS 180 tumors in their livers (Fig. 11A, panel 1) a single intravenous injection of SF-25-LAK cells substantially inhibited the tumor growth and eight out of thirteen animals were free of detectable hepatic tumors (Fig, 11 A, panel 2). Tumor weight in the treated group was significantly decreased (P < 0.01: vs. untreated controls by two-sided Mann-Whitney U test).

In contrast, human LAK cells alone were less effective (Fig. 11 A, panel 3). Only a small proportion of the animals were free of hepatic metastases (3 out of 13 animals) and the majority of mice developed tumors similar to untreated controls (P > 0.1: vs. untreated). Mice injected with LAK cells preincubated with c-SF-25 MAb had lower hepatic tumor burdens compared to untreated animals (Fig. 11 A, panel 4). However, a statistically significant difference was not observed (P > 0.1: vs. untreated).

A similar experiment was performed using human LAK cells cross- linked with nonspecific monoclonal human IgG1 (Sigma) as a control in order to determine if the anti-tumor effect by SF-25-LAK cells was particularly mediated by SF-25 MAb. As shown in Fig. 11B, panel 5, human IgGl-LAK cells did not produce any increased anti-tumor effect (P > 0.1: vs. LAK cells alone or untreated mice, respectively). The anti-tumor effect of multiple injections was also examined by injecting SF-25-LAK cells once weekly for 3 consecutive weeks. Inhibition of tumor growth was more prominent with multiple injections (P < 0.005: multiple injections vs. untreated mice; and P < 0.01: single injection vs. untreated mice) (Fig. 11C, panel 9).

In order to further confirm the observed in vivo anti-tumor effect, the survival curves were determined by intravenously injecting SF-25-LAK (10⁷ cells) into SCID mice 7 days after the tumor cell injection. All untreated mice died within 7 weeks after tumor cell injection due to massive parenchymal spread of tumor in the liver (median survival: 32± 7 days, n=10). In contrast, treated animals survived significantly longer than untreated controls (median survival: 52 ± 16 days, n=9; P < 0.01 vs. untreated by a two- sided Mann-Whitney U test). Furthermore, single or multiple injections (once a week to a total of four injections) of SF-25-LAK cells, starting even 14 days after the tumor cell injection when there are already macrometastases in the liver, improved the survival rate of the mice. Median survival of the single and multiple injection groups were 47 ± 15 days (n=9) and 59 ± 9 days (n=4), respectively. This improved survival was statistically significant (P < 0.02: single injection vs. untreated mice; and P < 0.005: multiple injections vs. untreated mice).

Under these experimental conditions, however, the in vivo purging of tumor cells by SF-25-LAK was not perfect. Therefore, the expression of SF- 25 antigen in the tumors from mice treated with SF-25-LAK cells was examined in order to investigate the possibility that this treatment may have allowed tumor growth as the result of selection of SF-25 antigen-negative tumor cells. This was not the case, as the SF-25 antigen was, in fact, expressed in SF-25-LAK treated tumors as shown by histological examination (Fig. 12A) and by flow cytometry (Fig. 12B). In addition, LS 180 tumor cells derived from both SF-25-LAK-treated (single and multiple treatments) and untreated mice were similarly lysed by LAK cells in vitro (Figure 13). Therefore, the outgrowth of tumor in treated mice is not the result of the selection of SF-25 negative or LAK-resistant tumor cells but may have resulted from an insufficient dose of SF-25 LAK cells compared to the tumor burden.

The ability of the SF-25 MAb alone to produce direct cytotoxicity, including apoptosis, to tumor cells was studied by examining DNA fragmentation and the release of deoxyuridine from the cells. The ability to inhibit in vitro tumor cell growth by SF-25 MAb alone was studied by examining ³H-thymidine uptake by the cells. However, neither apoptosis nor inhibition of in vitro tumor growth was produced by c-SF-25 MAb alone. In addition, complement-mediated cytotoxicity was not induced by c-SF-25 MAb (Figure 14). Therefore, augmentation of the anti -tumor effects by antibody- directed human LAK cells is not due to the effects of the MAb alone.

These studies demonstrated a potent biological effect of MAb-directed human LAK cells to inhibit the growth of human colon cancer established in the liver of SCID mice. Since no increased cytotoxicity was observed in SF- 25-LAK cells compared with LAK cells alone in vitro (Fig. 10) this augmentation of the anti-tumor effect must be largely due to the specific targeting of LAK cells to the tumor mass rather than the cytotoxicity induced by an ADCC mechanism. This is the first direct demonstration that antibodies which are bound to effector cells are able to positively influence the distribution of iv. injected effector cells to a tumor. Therefore, it is concluded that the targeting of human LAK cells to colon tui.iors is useful for the inhibition of tumor growth in vivo, and that the SCID mouse/human xenogenic chimeric model is useful to study the anti-tumor properties of human effector cells.

While the invention has been described in connection with specific embodiments thereof, it will be understood that it is capable of further modifications and this application is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the invention and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains and as may be applied to the essential features hereinbefore set forth and as follows in the scope of the appended claims.

Claims

Claims

1. A method of producing a targeted, cytotoxic, effector cell, comprising treating said cytotoxic effector cell with a) a cytokine and b) an antibody or an effective derivative or fragment thereof, said antibody being to a constitutive antigen of a tumor or other diseased tissue, wherein said treatment with antibody or effective derivative or fragment thereof is in the presence of a conjugating reagent, whereby- said antibody, or effective derivative or fragment thereof, is bound to said cytotoxic effector cell.

2. The method of claim 1, wherein treatment of said effector cell with said cytokine is prior to treatment of said effector cell with said antibody or effective derivative or fragment thereof.

3. The method of claim 1, wherein treatment of said effector cell with said antibody, or effective derivative of fragment thereof, is prior to treatment of said effector cell with said cytokine.

4. The method of claim 1, wherein the effector cell is simultaneously treated with both the cytokine and the antibody, or effective derivative or fragment thereof, at the same time.

5. The method of claim 1, wherein said conjugating reagent is polyethylene glycol.

6. The method of claim 5, wherein said polyethylene glycol is polyethylene glycol 8000.

7. The method of claim 1, wherein said antibody, or effective derivative or fragment thereof, is a mouse-human chimeric antibody to the SF- 25 antigen.

8. The method of claim 1, wherein said antibody is the SF-25 monoclonal antibody which is secreted by the hybridoma cell line given the ATCC designation HB 9599.

9. The method of claim 1, wherein said antibody, or effective derivative or fragment thereof, binds to the SF-25 antigen.

10. The method of claim 1, wherein said antibody is an antibody which specifically binds to the hapten which is bound by the monoclonal antibody secreted by the hybridoma cell line given the ATCC designation HB 9599.

11. The method of claim 1, wherein said cytokine is recombinantly produced IL-2.

12. A targeted, cytotoxic, effector cell, comprising a stable complex between a cytotoxic effector cell and an antibody, effective fragment or derivative thereof, wherein said antibody, effective fragment or derivative thereof, preserves its binding ability towards its antigen after it has been bound to said cytotoxic effector cell.

13. A method of suppressing the growth of the cells of a tumor or other diseased tissue, comprising administering to an animal an effective single dose, repeated doses, or an infusion of the targeted, cytotoxic, effector cell of claim 12.

14. The method of claim 13, wherein the cells of said tumor or other diseased tissue express the SF-25 antigen.

15. A method of killing tumor cells or cells of other diseased tissues, comprising administering to an animal an effective single dose, repeated doses, or an infusion of the targeted, cytotoxic, effector cell of claim 12.

16. The method of claim 17, wherein said cells of the tumor or other diseased tissue express the SF-25 antigen.

17. A method of preventing the development of tumors or other diseased tissues, comprising administering to an animal an effective single dose, repeated doses, or an infusion of the targeted, cytotoxic, effector cell of claim 12.

18. The method of claim 17, wherein said tumor or other diseased tissue expresses the SF-25 antigen.

19. An in vivo method of imaging the cells of tumors or other diseased tissues, comprising: administering an effective dose of detectably labeled, targeted, cytotoxic, effector cells of claim 12 to an animal; and measuring the distribution of said detectably labeled, targeted, cytotoxic, effector cells in said animal.

20. The method of claim 19, wherein the cells of said tumor or other disease tissue express the SF-25 antigen.

21. The method of claim 19, wherein said detectable label is a radiolabel.

22. An in vitro method of imaging the cells of tumors or other diseased tissues of an animal, comprising: administering an effective concentration of detectably labeled, targeted, cytotoxic effector cells to a tissue which has been removed from an animal; and measuring the distribution of said detectably labeled, targeted, cytotoxic effector cells in said tissue.

23. The method of claim 22, wherein the cells of said tumor or other diseased tissue express the SF-25 antigen.

24. The method of claim 23, wherein said label is a radiolabel.

25. The method of claim 24 wherein said detectable radiolabel is ³H.

26. The method of any of claims 13, 15, 17, 19, or 22 wherein said tumor is a colon adenocarcinoma.

27. The method of any of claims 13, 15, 17, 19, or 22 wherein said tumor is a hepatocellular carcinoma.

28. The method of any of claims 13, 15, 17, 19 or 22 wherein said animal is a human being.