NO20121261L - SURVIVIN-DERIVATED PEPTID, PHARMACOYTIC MIXTURE, MULTI-EPITOP VACCINE AND USE OF THE PEPTID FOR THE PREPARATION OF A MEDICINE FOR TREATMENT OF CANCER - Google Patents

Abstract

MHC class 1 restricted peptides derived from the tumor-associated antigen, survivin, where the peptides are capable of binding class 1 HLA molecules with a high affinity, are capable of triggering IFN-v producing cells in a PBL. population of a cancer patient and is capable of in situ detection of cytotoxic T cells in a tumor tissue, therapeutic and diagnostic compositions comprising the peptide and applications thereof.

Description

FIELD OF THE INVENTION

The present invention relates to new survivin-derived peptides and their use for diagnostic and therapeutic purposes, especially in cancer. In particular, the new peptides are MHC class 1-restricted T-cell epitopes that have the ability to trigger cytotoxic T-cell responses in cancer patients, including in situ and ex vivo responses. Specifically, such novel peptides are derived from the apoptosis inhibitor protein survivin, a known tumor-associated antigen (TAA).

TECHNICAL BACKGROUND AND STATE OF THE ART

The process by which the mammalian immune system recognizes and responds to non-self or foreign materials is complex. An important part of the system is the T-cell response. This response requires T cells to recognize and react with complexes of cell surface molecules, so-called human leukocyte antigens (HLA), which make up the human major histocompatibility complex (MHC), and peptides. The peptides are derived from larger molecules that are processed by the cells, which also present the HLA/MHC molecule. The interaction between T cells and complexes of HLA/peptide is limited in that it requires a T cell that is specific for a particular combination of an HLA molecule and a peptide. If a particular T cell is not present, there will be no T cell response, even if the partner complex is present. Likewise, there is no response if the specific complex is absent but the T cell is present.

The mechanism by which T cells recognize abnormal cellular conditions has also been implicated in cancer. WO92/20356 describes e.g. a family of genes that are processed into peptides that, in turn, are expressed on cell surfaces and can lead to lysis of the tumor cells by specific CTLs. These genes are called the MAGE family and are thought to code for tumor rejection antigen precursors or 'TRAP' molecules, and the peptides derived from them are called 'tumour rejection antigens' or ' TRAs".

In WO 94/05304 nonapeptides are described which bind to the HLA-A1 molecule. The reference describes that given the known specificity of particular peptides for particular HLA molecules, one should expect that a particular peptide binds to one HLA molecule, but not to others. This is important because different individuals have different HLA phenotypes. Although identification of a particular peptide as a partner for a particular HLA molecule has diagnostic and therapeutic implications, these are therefore only relevant for individuals with this particular HLA phenotype.

Several peptides presented by MHC molecules have been characterized, and it has been found that some of these may contain post-translational modifications that possibly have an impact on the functionality of the HLA-peptide complex. Consequently, a number of studies have found a connection between changes in the phosphorylation pattern and malignant transformation. Furthermore, it has been shown that phosphorylation could have a neutral, negative or even positive effect on peptide binding to class 1 HMC and that phosphopeptide-specific CTL, which discriminated between the phosphorylated and the non-phosphorylated versions of the peptide, could be generated, which showing that such CTL are most likely part of the class 1 MHC-restricted CTL repertoire. Recently, it has been shown that phosphorylated peptides are indeed naturally processed and presented by MHC class 1 molecules in vivo. Moreover, it has been established that phosphorylated peptides are present in extracts of isolated class 1 molecules from several different EBV-transformed B cells.

It is thus well known that peptide epitopes derived from tumor-associated antigens (TAA) can be recognized as antigens by cytotoxic T-lymphocytes (CTL) in the context of MHC molecules (1). However, although it is generally accepted that most, if not all, tumors are antigenic, only a few are truly immunogenic in the sense that tumor progression is satisfactorily controlled by the immune system.

To overcome this limitation, several immunotherapeutic trials have been initiated, such as vaccinations with TAA-derived peptides. For melanoma, the tumor for which the largest number of CTL-defined TAAs have been characterized, robust CTL responses to antigens have been induced by vaccination, and some patients experienced a complete remission of their disease (2,3). However, most of the peptide epitopes used in these vaccination trials are melanocyte-specific, and these peptides cannot be used for tumors of non-melanocyte origin. Furthermore, expression of these TAAs is heterogeneous among tumors from different patients and can even vary between metastases taken from a (same) patient. Nevertheless, during the last couple of years a number of tumor-specific peptide antigens, which are expressed in a number of different cancers, have been identified, for example HER-2 (4), Muc-1 (5) and telomerase (6) .

It has also been shown that through suitable manipulation, tumor antigens present in tumors can be exposed to the immune system. Studies have shown that the CD8+ CTL arm of the immune response, alone or in combination with CD4+ Th cells, constitutes the primary antitumor effector arm of the adaptive immune response. To date, the focus has mainly been on the CTL arm of the immune response. However, it is becoming clearer that the CD4 T cell response plays an essential role in tumor rejection, especially in the induction phase or in the extension of a CTL response in vivo. Accordingly, the incorporation of class II-restricted tumor antigens into effective tumor vaccination protocols may increase the efficacy of the vaccines.

Apoptosis is a genetic program for cellular suicide, and inhibition of apoptosis has been proposed as an important mechanism involved in carcinogenesis by extending the lifespan of cells that favor the accumulation of transforming mutations (7). Survivin is a newly identified member of the family of inhibitors of apoptosis proteins (IAPs). In a global gene expression analysis of approximately 4 million transcripts, survivin was identified as one of the most important genes that was always upregulated in many types of cancer, but not in normal tissue (8). Solid tumors with overexpression of survivin include lung, colon, breast, pancreatic and prostate cancer as well as haematopoietic malignancies (9). Moreover, a number of melanoma and non-melanoma skin cancers have been reported to be consistently survivin-positive (10,11). Overexpression of survivin in most human cancers suggests a general role for apoptosis inhibition in tumor progression, a notion supported by the observation that in colorectal and bladder cancer, as well as in neuroblastoma, expression of survivin was associated with an unfavorable prognosis. In contrast, survivin is undetectable in normal tissue in adults. These properties qualify survivin as a suitable TAA for both diagnostic and therapeutic purposes.

Thus, during the last decade, a large number of TAAs have been identified that are recognized by CTLs in a major histocompatibility antigen complex (MHC)-restricted manner. As survivin is overexpressed in most human cancers, and inhibition of its function results in increased apoptosis, this protein may serve as a target for therapeutic CTL responses. The protein survivin and its potential diagnostic and therapeutic use are described in (8) and US 6,245,523. Survivin is a 16.5 kDa cytoplasmic protein containing a single BIR and a highly charged coiled-coil carboxy-terminal region instead of a "RING finger", which inhibits apoptosis induced by growth factor (IL-3) depletion when transferred into B- cell precursors. The gene encoding survivin is nearly identical to the sequence of Effector Cell Protease Receptor-1 (EPR-1), but oriented in the opposite direction, suggesting the existence of two separate genes duplicated in a "head -to-head" configuration. Accordingly, survivin can be described as an antisense EPR-1 product. Functionally, inhibition of survivin expression via upregulation of its natural antisense EPR-1 transcript results in massive apoptosis and reduced cell growth.

US 6,245,523 describes the isolation of purified survivin, and it presents nucleic acid molecules encoding the protein survivin and antibodies and other molecules that bind to survivin. US 6,245,523 also describes anti-apoptotic active fragments of the survivin protein and variants thereof in which an amino acid residue has been inserted N- or C-terminal to or within the described survivin sequence. It has been specifically shown that such peptides may contain key functional residues required for apoptosis, i.e. Trp at position 67, Pro at position 73 and Cys at position 84.

Andersen et al. CANCER RESEARCH, vol. 61, No. 3 (2001) 869-872 describes HLA-A2-specific survivin peptides: Surl-SurlO, SurlL2 and SurlM2. Andersen et al. also describes specific CTL responses against said peptides, affinity values for the binding of said peptides to class I HLA molecules and suggests the use of the peptides in vaccination, cancer therapy and cancer diagnostics.

Andersen et al. CANCER RESEARCH, vol. 61, No. 16 (2001) 5964-68 describes HLA-binding survivin peptides: Surl (LTLGEFLKL), Sur9 (ELTLGEFLKL) and SurlM2 and the generation of CTL responses by said peptides. The document also describes MHC class I/peptide complexes that were multimerized, and their use in the detection of survivin-reactive cells in tumor tissue. Andersen et al. propose applications of such survivin peptides for peptide-based anticancer vaccines in combination with conventional cancer therapy.

Schmitz et al. CANCER RESEARCH, vol. 60, No. 17 (2000) 4845-49 describes the HLA-binding peptide ELTLGEFLKL. The article demonstrates the generation of CTL responses by said peptide and suggests the use of the peptide in anticancer vaccines.

Hirohashi et al. CLINICAL CANCER RESEARCH: AN OFFICIAL JOURNAL OF THE AMERICAN ASSOCIATION FOR CANCER RESEARCH, vol. 8, No. 6 (2002) 1731-39 describes an HLA-A24-restricted CTL epitope of survivin-2B (splice variant found in tumors). The article also describes other survivin peptides, e.g. AFLSVKKQF and

QFEELTLGEF.

Fortugno et al. JOURNAL OF CELL SCIENCE, vol. 115, No. 3 (2002) 575-85 describes polyclonal antibodies against survivin epitopes Ala3-Ilel9, Met38-Thr48, Pro47-Phe58 and Cys57-Trp67.

None of the documents in the prior art describe the MHC class 1-restricted epitope according to the present invention, which comprises the sequence FTELTLGEF. Furthermore, the state of the art only describes binding of survivin CTL epitopes to the HLA-A2 allele. However, to develop effective peptide-based vaccine programs it is necessary to also be able to target different HLA molecules, including HLA-A3, HLA-B and HLA-C alleles. Finally, a set of multiple survivin epitope peptides is desirable for use in such vaccines, where each peptide interacts specifically with a different HLA molecule.

The present invention is based on the discovery that MHC class 1-restricted peptides can be derived from the survivin protein, which has the ability to bind to MHC class 1 HLA molecules and thereby trigger both ex vivo and in situ CTL immune responses in patients suffering from a wide spectrum of cancers. These findings strongly suggest that survivin acts as a TRAP molecule, which is converted by cells into peptides that have TRA function. Obviously, these findings open the way to new therapeutic and diagnostic approaches which, due to the fact that survivin appears to be universally expressed by tumor cells, are generally suitable for the control of cancer diseases.

SUMMARY OF THE INVENTION

Accordingly, the invention relates in a first aspect to an MHC class 1-restricted epitope peptide derived from survivin, said epitope has the sequence FTELTLGEF as specified by SEQ ID NO: 36 and has at least one of the following characteristics: (i) is able to bind to the class 1 HLA-A1 molecule with an affinity as measured by the amount of the peptide capable of effecting half maximal recovery of the class 1 HLA molecule (Cso value), which is a maximum of 50 µM as determined by an HLA stabilization assay; (ii) is capable of eliciting IFN-γ-producing cells in a PBL population from a cancer patient at a frequency of at least 1 per IO<4>PBL as determined by an ELISPOT assay, and/or (iii) is capable of in situ detection in a tumor tissue of CTVs reactive with the epitope peptide.

Preferably, the peptide according to the invention has at least two, preferably all three of these properties.

In further aspects, the invention provides a pharmaceutical composition comprising the peptide as defined above and a composition for ex vivo or in situ diagnosis of the presence in a cancer patient of survivin-reactive T cells among PBLs or in tumor tissue, wherein this composition comprises a peptide as defined above.

An important aspect of the invention relates to the use of the peptide as defined above for the production of a drug for the treatment of cancer. Finally, the invention provides a multi-epitope vaccine comprising an MHC class I-restricted epitope peptide according to the invention.

DETAILED DESCRIPTION OF THE INVENTION

The novel MHC class I-restricted peptide of the invention is characterized by having at least one of several properties, one of which is the ability to bind the class I HLA-Al molecule to which it is restricted with an affinity which, when measured as the amount of the peptide that can effect half maximal recovery of the class I HLA molecule (C50 value) in an HLA stabilization assay is a maximum of 50 uM. This HLA stabilization assay is performed as described previously (12, 13), and it is based on stabilization of the HLA molecule after insertion of the peptide in the peptide transporter-deficient cell line T2. Then, correctly folded stable heavy chains of HLA are immunoprecipitated using conformation-dependent antibodies, and peptide binding is quantified.

This assay provides a simple method for screening candidate peptides for their ability to bind to a given HLA allele molecule at the above affinity. In preferred embodiments, the peptide of the invention is one that has a C 50 value of at most 30 µM, such as a C 50 value of at most 20 µM.

As mentioned above, the HLA system represents the major human histocompatibility complex (MHC) system. In general, MHC systems control a number of characteristics: transplant antigens, thymus-dependent immune responses, certain complement factors, and predisposition to certain diseases. More specifically, the MHC codes for three different types of molecules, i.e. class I, II and III molecules, which determine the more general characteristics of the MHC. Of these molecules, the class I molecules are so-called HLA-A, HLA-B and HLA-C molecules which are presented on the surface of most nucleated cells and platelets.

The peptides according to the invention are derived from the known sequence of survivin, e.g. the sequence described in US 6,245,523.

Therefore, a simple approach to identify the peptides of the invention would include the following steps: selecting a particular HLA molecule, e.g. one that occurs with great frequency in a given population, to perform a comparative array analysis as described above to identify "anchor residue motifs" in the survivin protein, to isolate or construct peptides of a suitable size that comprise one or more of the identified anchor residues and testing the resulting peptides for (I) ability to bind to the particular HLA molecule using the HLA stabilization assay as described herein, (ii) ability of the peptides to trigger IFN-γ producing cells in a PBL population from a cancer patient with a frequency of at least 1 per IO<4>PBLs as determined by an ELISPOT assay as described herein and/or (III) the ability of the peptides to detect in situ in a tumor tissue CTVs reactive with the epitope peptides which is tested.

In specific embodiments, the peptide according to the invention is an HLA-Al-restricted peptide having the sequence FTELTLGEF (SEQ ID NO:36)).

Preferably, the peptide according to the invention comprises a maximum of 20 amino acid residues, such as a maximum of 10 amino acid residues. In specific embodiments, the peptide is a nonapeptide, a decapeptide, or an undecapeptide.

The peptide according to the invention is, as mentioned above, derived from a survivin protein or a fragment thereof. The survivin protein from which the peptide may be derived is survivin protein from any animal species in which the protein is expressed. In preferred embodiments, the survivin starter protein is derived from a mammalian species, including a rodent species, rabbit and a primate species, such as, for example, human. Based on the sequence of the selected survivin protein, the protein of the invention is derived by any suitable chemical or enzymatic treatment of the survivin starting material resulting in a peptide of an appropriate size as indicated above, or it may be synthesized by any conventional peptide synthesis procedure known to those of ordinary skill in the art.

A significant feature of the peptide according to the invention is its ability to recognize or trigger IFN-γ-producing responder T cells, i.e. cytotoxic T cells (CTLs) that specifically recognize the particular peptide in a PBL population of tumor cells from a cancer patient (target cells). This activity can be easily determined by subjecting PBLs or tumor cells from a patient to an ELISPOT assay as described in reference (16) and in the following examples. Before performing the assay, it may be advantageous to stimulate the PBL population or the tumor cells to be analyzed by leaving the cells in contact with the peptide to be tested. Preferably, the peptide has the ability to trigger or recognize IFN-γ producing T cells at a frequency of at least 1 per 10<4>PBLs as determined by an ELISPOT assay as used herein.

The ELISPOT assay represents a solid tool for monitoring survivin peptide-specific T-cell responses. Although ELISPOT reactivity has been shown to correlate in most cases with the capacity of CLLs to lyse the target cells, however, conclusive evidence for this assumption can only be provided directly. Such direct evidence is provided here, since it was shown (see Example 2) that survivin reactive cells isolated by means of HLA/peptide complexes possess the functional capacity to lyse target cells. In addition, it has been shown that the isolated CTVs which specifically recognize a peptide according to the invention had the ability to lyse HLA-matched tumor cells of other origin, for example melanomas and breast cancer. This finding strongly suggests that cancer cells generally convert and present the same endogenous survivin peptide. Therefore, a main implication of the findings here is that the peptides according to the invention are expressed and complexed with HLA molecules on a number of cancer cells of different histological origins. This makes these cancer cells susceptible to destruction by CTLs and highlights the potential utility of survivin immunization to control the growth of various neoplasms. The occurrence of spontaneous CTL responses in PBLs and tumor cells to HLA-restricted survivin-derived peptide epitopes from patients suffering from three unrelated cancer types, i.e. breast cancer, melanoma and CLL, further strengthens the universal immunotherapeutic potential of this tumor antigen.

In accordance with this, the peptide according to the invention in another preferred embodiment has the ability to trigger IFN-γ-producing cells in a PBL population from a patient who has a cancer in which survivin is expressed, including a hematopoietic malignancy including chronic lymphocytic leukemia and chronic myeloid leukaemia, melanoma, breast cancer, cervix cancer, ovarian cancer, lung cancer, colon cancer, pancreatic cancer and prostate cancer. Specifically, the peptide according to the invention has the ability to trigger an immune response in the form of a T cell that has a cytotoxic effect against survivin-expressing cells of a cancer cell line, including a cell line selected from the breast cancer cell line MCF-7 and the melaoma cell line FM3.

In addition to their capacity to trigger immune responses in PBL populations and cancer cell lines, it was shown that the peptides according to the invention also have the ability to trigger cytolytic immune responses in situ, i.e. in solid tumor tissue. This was shown by providing HLA-peptide complexes, for example by multimensing and by being provided with a detectable label, and using such complexes in immunohistochemical stainings to detect in a tumor tissue CTVs reactive with the epitope peptide according to the invention. Consequently, a further significant feature of the peptide according to the invention is that it is capable of in situ detection in a tumor tissue of CTVs that are reactive with the epitope peptide.

It is expected that the peptides of the invention, in addition to their capacity to bind to HLA molecules resulting in the presentation of complexes of HLA and peptides on cell surfaces, where these complexes in turn act as epitopes or targets for cytolytic T cells , can trigger other types of immune responses, such as B responses that result in the production of antibodies against the complexes and/or a delayed type hypersensitivity (DTH, Delayed Type Hypersensitivity) reaction. The latter type of immune response is defined as a redness and a noticeable induration at the injection site for the peptide according to the invention.

It is well known that the various HLA molecules are of different occurrence in the large population populations. Accordingly, there is a need to identify peptide epitopes that are restricted to more HLA class I molecules in order to expand the patient cohort that can be treated according to the methods of the present invention. The characterization of multiple survivin epitopes with different HLA restriction elements expands the clinical potential of this target antigen in two important ways: (i) It increases the number of patients suitable for immunotherapy based on survivin-derived peptides. The HLA-A2 antigen is expressed by about 50% of Caucasian and Asian populations, the HLA-A1 and HLA-A3 antigens are both expressed by about 25% of Caucasians and 5% of Asians, while the HLA-All antigen is expressed by about 15% of Caucasians and 30% of Asians. Although these numbers cannot be summed due to co-expression, a combination of peptides limited by a plurality of these will certainly include most cancer patients. (ii) The joint targeting of several restriction elements in each patient is likely to reduce the risk of loss of immunity due to HLA allele loss. Loss of a single HLA allele constitutes a significant part of MHC changes described for cancer cells, while complete loss of class I expression is a rather rare event. Therefore, through the identification of survivin epitopes restricted to different HLA alleles, it is now possible to target more than one HLA molecule simultaneously in patients with allelic overlap.

Accordingly, based on the disclosure of the present invention, one skilled in the art will be able to develop highly immunogenic multi-epitope vaccines. Preferably, such vaccines should be designed to promote a simultaneous release of the most suitable survivin-derived peptides, possibly in combination with other suitable peptides and/or adjuvants as described below.

Another aspect of the invention thus provides a multi-epitope vaccine comprising an MHC class I-restricted epitope peptide derived from survivin, said epitope having the sequence FTELTLGEF as specified by SEQ ID NO: 36 and having at least one of the following characteristics: (i) is capable of binding the class I HLA-Al molecule to which it is restricted with an affinity as measured by the amount of the peptide capable of effecting half maximal recovery of the class I HLA molecule (Cso value), as maximally is 50 µM as determined by an HLA stabilization assay; (ii) is capable of eliciting IFN-γ-producing cells in a PBL population from a cancer patient at a frequency of at least 1 per IO<4>PBLs as determined by an ELISPOT assay, and/or (iii ) is capable of in situ detection in a tumor tissue of CTVs reactive with the epitope peptide.

In preferred embodiments, the multi-epitope vaccine includes a combination of survivin-derived peptide epitopes depending on the tissue type of the particular patient. Selection of peptides that potentially have the ability to bind to a specific HLA molecule can be done by comparing known sequences that bind to a given specific HLA molecule in order to reveal the dominance of a few related amino acids at specific positions in the peptides . Such prominent amino acid residues are also referred to in this document as anchor residues or anchor residue motifs. By following such a relatively simple procedure based on known sequence data that can be found in available databases, peptides can be derived from the survivin protein molecule likely to bind to the particular HLA molecule. Representative examples of such assays for a variety of HLA molecules are provided in the table below: Accordingly, as an example, nonapeptides potentially having the ability to bind to HLA-A1 will have one of the following sequences: Xaa-T-D-Xaa- Xaa-Xaa-L-Xaa-Y, Xaa-T-E-Xaa-Xaa-Xaa-L-Xaa-Y; Xaa-S-D-Xaa-Xaa-Xaa-L-Xaa-Y or Xaa-S-E-Xaa-Xaa-Xaa-L-Xaa-Y (where Xaa indicates any amino acid residue). In a similar manner, sequences potentially capable of binding to any other HLA molecule can be constructed.

Furthermore, as previously described, there has been a growing interest in eliciting tumor-specific T-helper cell immunity, ie, vaccinating with class II MHC-restricted epitopes, despite the fact that tumors generally do not express class II MHC. This is based on the new finding that the induction and effectiveness of the vaccine-induced antitumor response in many cases requires the participation of tumor-specific CD4-positive "IV cells. An important driving factor for the development of vaccines with a more complex composition is therefore the desire to designate multiple tumor antigens, for example, by designing vaccines that comprise or encode a collection of precisely selected CTL and Th cell epitopes.

Obviously, multi-epitope vaccines represent an effective method for increasing immunity against epitopes derived from several different antigens without the need for the introduction (gene coding) of potentially harmful proteins such as oncoproteins. Such vaccines also allow selective induction of immunity against subdominant and cryptic T-cell epitopes, which may be particularly important in the case of tumor-associated autoantigens for which tolerance may occur

the epitopes that are significantly presented in normal tissues. Furthermore, antigen-presenting cells may fail to present certain epitopes expressed in tumor cells due to functional differences between the immunoproteasomes of antigen-presenting cells and the "constitutive" proteasomes present in most tumor cells. In the case of peptide-based vaccines, such epitopes can be administered in an "MHC-ready" form, enabling presentation through exogenous insertion independent of antigen uptake and transformation by the host's antigen-presenting cells.

It is obvious that the findings of the present invention provide the basis for both therapeutic and diagnostic applications of the survivin-derived peptides.

An important aspect of the invention thus provides a pharmaceutical composition comprising an MHC class I-restricted epitope-peptide derived from survivin, wherein said epitope has the sequence FTELTLGEF as specified by SEQ ID NO: 36 and has at least one of the following characteristics: (i) is capable of binding to the class I HLA-Al molecule to which it is restricted with an affinity as measured by the amount of peptide capable of effecting half-maximal recovery of the class I HLA molecule (Cso value), which is a maximum of 50 uM as determined by an HLA stability assay; (ii) is capable of triggering IFN-γ-producing cells in a PBL population from a cancer patient at a frequency of at least 1 per 10<4>PBLs as determined by an ELISPOT assay, and/or (lii) is capable of in situ detection in a tumor tissue of CTLs that are reactive with the epitope peptide.

In a further aspect, the present invention provides a pharmaceutical composition comprising one or more of the peptides according to the invention alone or in suitable combination with other proteins or peptide fragments.

In particular, the invention provides a pharmaceutical composition comprising, in addition to the epitope having the sequence FTELTLGEF, a combination of two or more MHC class I-restricted epitope peptides derived from survivin, each interacting specifically with another HLA molecule and having at least one of the following characteristics: (i) is capable of binding to the class I HLA molecule to which it is restricted with an affinity as measured by the amount of the peptide capable of effecting half maximal recovery of the class I HLA molecule (C50 value); , which is a maximum of 50 (xM as determined by an HLA stabilization assay; (ii) is capable of eliciting IFN-γ-producing cells in a PBL population from a cancer patient at a frequency of at least 1 per 10<4>PBL 'is as determined by an ELISPOT assay, and/or (iii) is capable of in situ detection in a tumor tissue of CTLs reactive with the epitope peptide.

In particular embodiments, the pharmaceutical composition is one in which each epitope peptide specifically interacts with an MHC class I HLA-A molecule, an MHC class I HLA-B molecule, or an MHC class I HLA-C molecule.

In further embodiments, said MHC class I HLA-A molecule is an HLA-A1, HLA-A2, HLA-A3, HLA-A9, HLA-A10, HLA-A11, HLA-Awl9, HLAA23(9), HLA-A24 (9), HLA-A25(10), HLA-A26(10), HLA-A28, HLA-A29(wl9), HLA-A30(wl9), HLA-A31(wl9), HLA-A32(wl9), HLA-Aw33(wl9), HLA-Aw34(10), HLA-Aw36, HLA-Aw43, HLA-Aw66(10), HLA-Aw68(28), HLA-A69(28) molecule.

In still further embodiments, said MHC Class I HLA-B molecule includes any of the following: HLA-B5, HLA-B7, HLA-B8, HLA-B12, HLA-B13, HLA-B14, HLA-B15, HLA-B16, HLA-B17, HLA-B18, HLA-B21, HLA-Bw22, HLA-B27, HLA-B35, HLA-B37, HLA-B38, HLA-B39, HLA-B40, HLA-Bw41, HLA-Bw42, HLA- B44, HLA-B45, HLA-Bw46 and HLA-Bw47.

In other embodiments, said MHC class I HLA-C molecule includes any of the following: HLA-Cwl, HLA-Cw2, HLA-Cw3, HLA-Cw4, HLA-Cw5, HLA-Cw6, HLA-Cw7 and HLA-Cwl6.

According to some embodiments, the pharmaceutical composition comprises an epitope peptide which is a native sequence of survivin from a mammalian species.

In further embodiments, the pharmaceutical composition comprises an epitope peptide selected from the group consisting of SEQ ID NO: 1, SEQ ID NO: 2, SEQ ID NO: 3, SEQ ID NO: 6, SEQ ID NO: 7, SEQ ID NO: 27, SEQ ID NO: 45 and SEQ ID NO: 66. In other embodiments, the pharmaceutical composition comprises an epitope peptide selected from the group consisting of: SEQ ID NO: 4, SEQ ID NO: 5, SEQ ID NO: 8, SEQ ID NO: ID NO: 9, SEQ ID NO: 36, SEQ ID NO: 38, SEQ ID NO: 39, SEQ ID NO: 58.

In preferred embodiments, said epitope peptides are capable of triggering IFN-γ-producing cells in a PBL population from a cancer patient at a frequency of at least 10 per 10<4>PBLs. In further preferred embodiments, said epitope peptides are capable of triggering IFN-γ-producing cells in a PBL population from a patient having a cancer, in which survivin is expressed.

Furthermore, it can be advantageous to carry out post-translational modifications of the peptides according to the invention. It has been shown that exposure of breast carcinoma MCF-7 or cervical carcinoma HeLa cells to anticancer agents including Adriamycin, Taxol or UVB resulted in a 4-5-fold increase in survivin expression. Changes in survivin levels after anticancer treatment did not involve modulation of survivin mRNA expression and were independent of de novo gene transcription. Conversely, inhibition of survivin phosphorylation at Thr<34> with the cyclin-dependent kinase inhibitor flavopiridol resulted in loss of survivin expression, and non-phosphorylatable survivin Thr<34->Ala exhibited accelerated clearance compared to wild-type survivin. Sequential removal of survivin phosphorylation at Thr<34> increased tumor cell apoptosis induced by anticancer agents independent of p53 and suppressed tumor growth without toxicity in a breast cancer xenograft model in vivo. These data suggest that Thr<34->phosphorylation crucially regulates survivin levels in tumor cells and that sequential removal of Thr34 kinase activity can remove the survivin viability checkpoint and increase apoptosis in tumor cells. Consequently, it is considered that the survivin-derived peptides according to the invention comprise phosphorylated peptides. Native survivin phosphopeptide antigens can be identified by scanning for the presence of MHC peptide binding motifs around the phosphorylation site Thr34. Thus, possible survivin-derived phosphopeptide sequences include TPERMAEAGF, a putative HLA-B35 and/or HLA-B7 and/or an HLA-B51 restricted peptide antigen. Additional native phosphopeptides encompassed include: HLA-A2: CACTPERMA and CTPERMAEA; HLA-A3: FLEGCACTP; HLA-B7/HLA-B35/HLA-B51: WPFLEGCACT (phosphorylated Thr residue is marked in bold).

Further embodiments of the invention provide pharmaceutical compositions further comprising an immunogenic protein or peptide fragment selected from a protein or peptide fragment that does not belong to or is derived from the survivin protein family.

In specific embodiments, such other proteins or peptide fragments include, but are not limited to, proteins involved in the regulation of

cell apoptosis or peptide fragments thereof. Suitable examples of such proteins may be selected from the Bcl-2 protein family, for example the Bcl-2 protein, the Bcl-w protein, the Mcl-1 protein, the Bcl-XL protein and peptide fragments derived from any of the proteins. Other known apoptosis inhibitors include members of

the inhibitor of apoptosis protein (IAP) family, such as X-IAP, C-IAP1 and C-IAP2. These proteins are all relatively widely expressed, while the inhibitor of apoptosis polypeptide ML-IAP has a rather selective expression, and is predominantly detected in melanomas. Fragments of ML-IAP which have the ability to trigger a specific T-cell response, i.e. a cytotoxic T-cell response or a T-helper cell response, can therefore optionally be included in the mixture according to the present invention. Useful peptide fragments of ML-IAP include ML-IAP245(RLQEERTCKV)(SEQ ID NO:75), ML-IAP28o(QLCPICRAPV)(SEQ ID NO:76),

ML-IAP90(RLASFYDWPL)(SEQ ID NO:77), ML-IAP154(LLRSKGRDFV)(SEQ ID

NO:78), ML-IAP230(VLEPPGARDV)(SEQ ID NO:79), ML-IAP98(PLTAEVPPEL)(SEQ ID NO:80), ML-IAP34(SLGSPVLGL)(SEQ ID NO:81), ML-IAP54 (QILGQLRPL)(SEQ ID NO:82), ML-IAP99(LTAEVPPEL)(SEQ ID NO:83), ML-IAP83(GMGSEELRL)(SEQ ID NO:84) and ML-IAP20o(ELPTPRREV)(SEQ ID NO: 85).

Moreover, the composition according to the present invention can be provided as a multi-epitope vaccine comprising class I-restricted epitopes and/or class II-restricted epitopes as defined above.

Example of a currently preferred multi-epitope vaccine includes "tailored" combinations of survivin-derived peptide epitopes depending on the tissue type of the given patient, for example an individual carrying HLA-A2, HLA-A3 and HLA-B35 phenotypes can be vaccinated with a vaccine comprising surlM2, sur9, surl8K10, sur46Y9, sur51Y9. In addition, the pharmaceutical mixture according to the invention can advantageously comprise at least one further immunogenic protein or peptide fragment thereof selected from a protein or peptide fragment which does not belong to or is derived from the survivin protein. In specific embodiments, the immunogenic protein or peptide fragment thereof is derived from the Bcl-2 protein family as described above. Another immunogenic Bcl-2 derived peptide is an HLA-A2 restricted peptide having a sequence selected from the following: Bdi72, Bcl80, Bcl2o8 and Bcl2i4.

Because the peptides according to the invention are relatively small molecules, in such mixtures it may be necessary to combine the peptides with various substances such as adjuvants to produce vaccines, immunogenic mixtures etc. Adjuvants are, broadly defined, substances that promote immune responses. Often the best choice of adjuvant is Freund's complete or incomplete adjuvant, or killed fl. pertussis/s organisms, used for example in combination with alum-precipitated antigen. A general discussion of adjuvants can be found in Goding, Monoclonal Antibodies: Principles and Practice (2nd ed., 1986) at pages 61-63. However, Goding writes that when the relevant antigen has a low molecular weight or is weakly immunogenic, coupling to an immunogenic carrier is recommended. Examples of such carrier molecules include keyhole limpet haemocyanin, bovmt serum albumin, ovalbumin and poultry immunoglobulin. Various saponin extracts have also been suggested to be useful as adjuvants in immunogenic compositions. Recently, it has been proposed to use granulocyte-macrophage colony-stimulating factor (GM-CSF), a well-known cytokine, as an adjuvant (WO 97/28816).

Accordingly, the invention comprises a therapeutic mixture which further comprises any substance including any of the above mentioned or combinations thereof. It is also contemplated that the antigen, i.e. the peptide according to the invention, and the adjuvant can be administered separately in any suitable order.

The choice of antigen in the pharmaceutical mixture according to the invention will depend on parameters that can be determined by the person skilled in the art. As previously mentioned, each of the various peptides according to the invention is presented on the cell surface by a specific HLA molecule. As such, if an individual to be treated is typed with respect to HLA phenotype, a peptide(s) known to bind to the particular HLA molecule is selected.

Alternatively, the appropriate antigen is selected based on the occurrence of the various HLA phenotypes in a given population. As an example, HLA-A2 is the most frequently occurring phenotype in the Caucasian population, and therefore a mixture containing a survivin-derived peptide that binds to HLA-A2 will be active in a large portion of this population. However, the mixture according to the invention can also contain a combination of two or more survivin-derived peptides where each of them interacts specifically with another HLA molecule to cover a larger part of the target population. As examples, therefore, the pharmaceutical composition may contain a combination of a peptide restricted to an HLA-A molecule and a peptide restricted to an HLA-B molecule, for example including the HLA-A and HLA-B molecules corresponding to the occurrence of the HLA phenotypes in the target population, such as HLA-A2 and HLA-B35. Moreover, the mixture may comprise a peptide restricted to an HLA-C molecule.

It is believed that useful immunogenic compositions according to the invention, in addition to a survivin-derived peptide as defined here, may comprise an immunologically effective amount of the survivin protein as such as defined here, or an immunogenic fragment thereof.

The amount of the immunogenic peptide according to the invention in the pharmaceutical composition may vary depending on the particular application. But a single dose of the immunogen is preferably somewhere between approx. 10 ug and approx. 5000 ug, more advantageously between approx. 50 ug and approx. 2500 ug, for example between approx. 100 ug and approx. 1000 ug. Routes of administration include intradermal, subcutaneous and intravenous administration, implantation in the form of a "time release" formulation, etc. All known forms of administration are covered here. This also includes all conventional dosage forms which in the art are considered suitable formulations of injectable immunogenic peptide mixtures, such as lyophilized forms and solutions, suspensions or emulsion forms which, if necessary, contain conventional pharmaceutically acceptable carriers, solvents, preservatives, adjuvants, buffer components, etc.

The immunoprotective effect of the composition according to the invention can be determined using different approaches. Examples of this are given in the following examples. A further example of how a CTL response elicited by the immunogenic composition can be determined is given in WO 97/28816, supra. A successful immune response can also be determined by the occurrence of DTH reactions after immunization and/or detection of antibodies that specifically recognize the peptide(s) in the vaccine mixture.

In preferred embodiments, the pharmaceutical composition according to the invention is an immunogenic composition or vaccine that has the ability to trigger an immune response to a cancer disease. As used herein, the term "immunogenic composition or vaccine" refers to a composition that elicits at least one type of immune response directed against cancer cells. Such an immune response can therefore be of any of the types mentioned above: A CTL response in which CTVs are generated which are able to recognize the HLA/peptide complex presented on cell surfaces, and which results in cell lysis, i.e. .in the vaccinated individual, the vaccine triggers the production of effector T cells that have a cytotoxic effect against the cancer cells, a B cell response that leads to the production of anticancer antibodies and/or a DHT-type immune response.

In useful embodiments, an immune response directed against a cancer is elicited by administering the peptide of the invention either by loading MHC class I molecules onto antigen-presenting cells (APCs) from the patient, by isolating PBLs from the patient and incubating the cells with the peptide prior to injection of the cells back into the patient or by isolating precursor APCs from the patient and differentiating the cells into professional APCs using cytokines and antigen before injecting the cells back into the patient. Thus, in one embodiment of the present invention, a method for treating cancer patients is one in which the peptide is administered by presenting the peptide to the patient's antigen-presenting cells (APCs) ex vivo followed by injection of the treated APCs back into the patient. There are at least two alternative ways of doing this. One alternative is to isolate APCs from the cancer patient and incubate (charge) the MHC class I molecules with the peptide. "Loading" the MHC class I molecules means incubating the APCs with the peptide so that the APCs with the MHC class I molecules specific for the peptide will bind to the peptide and therefore become able to present it to the T cells. The APCs are then re-injected into the patient. Another alternative method is based on the new discoveries made in the field of dendritic cell biology. In this case, monocytes (which are dendritic precursor cells) are isolated from the patient and differentiated in vitro into professional APCs (or dendritic cells) using cytokines and antigen. This is described in Examples 3 and 5, where adherent PBLs (which are mainly monocytes) are cultured in vitro together with GM-CSF, IL-4 and TNF-α. The in vitro generated DCs are then pulsed with the peptide and injected into the patient.

Due to the fact that survivin is found to be expressed in most cancers, it is very likely that the vaccines of the invention can be provided to control any type of cancer in which survivin is expressed. Thus, the vaccine mixture according to the invention is immunologically active against, for example, a hematopoietic malignancy including chronic lymphocytic leukemia and chronic myeloid leukemia, melanoma, breast cancer, cervical cancer, ovarian cancer, lung cancer, colon cancer, pancreatic cancer and prostate cancer.

From the above description, the person skilled in the art will easily understand that the peptides according to the invention are applicable as cancer diagnostic tools, particularly applicable, as the peptides are derived from survivin which is expressed in all types of cancer. Therefore, the peptides according to the invention provide the basis for developing universally applicable diagnostic and prognostic procedures when it comes to cancer diseases. In other applicable embodiments, the mixture according to the invention is a mixture for ex vivo or in situ diagnosis of the occurrence in a cancer patient, for example based on the detection of survivin-reactive T cells among PBLs or in tumor tissue.

In one aspect, the invention provides a complex of a peptide of the invention and a class I HLA molecule or a fragment of such a molecule, which is useful as a diagnostic reagent as described supra. The complex is formed by any conventional method including those described in the following examples. Such a complex can be monomeric or multimeric.

The present invention provides aids to relieve or cure a cancer. Accordingly, it is a further aspect of the invention to use the peptides as defined above for the production of a drug for the treatment of cancer. An even further aspect of the present invention relates to the use of the peptide as defined above for the production of a drug for the treatment of cancer. Preferably a cancer associated with expression of survivin, including as examples: a hematopoietic malignancy including chronic lymphocytic leukemia and chronic myeloid leukemia, melanoma, breast cancer, cervical cancer, ovarian cancer, lung cancer, colon cancer, pancreatic cancer and prostate cancer. The application comprises administering to a patient suffering from the disease, an effective amount of the pharmaceutical composition according to the invention, a molecule which has the ability to bind specifically to a peptide according to the invention and/or a molecule which has the ability to block the binding of such a molecule.

In some cases, it will be appropriate to combine the application of the invention with a conventional cancer treatment, such as radiotherapy or chemotherapy.

The invention will now be described in greater detail in the non-limiting examples and figures below, therein

Fig. 1 illustrates the T cell response as measured by an ELISPOT in patient CLL1 to no peptide, Sur1 (LTLGEFLKL, SEQ ID NO:10) peptide and Sur9 (ELTLGEFLKL, SEQ ID NO:3) peptide. PBLs were stimulated once with peptide before plating at 6x10<5> cells per well in duplicate. The average number of spots per peptide was calculated by

use a CCD scanning device and a computer system,

Fig. 2 illustrates T cell response as measured by an ELISPOT in patient CLL1 to no peptide, the peptide analog SurlL2 (LLLGEFLKL, SEQ ID NO:4) and the peptide analog SurlM2 (LMLGEFLKL, SEQ ID NO:5). PBLs were stimulated once with peptide before plating at 10<4> cells per well in duplicate. The average number of spots per peptide was calculated using a CCD scanning device and a computer system,

Fig. 3 shows responses, measured in an ELISPOT in patient CLL2 and CLL3, to no peptide (black bar), the Surl(LTLGEFLKL, SEQ ID NO:10) peptide (gray bar), Sur9(ELTLGEFLKL, SEQ ID NO:3 ) peptide (white column), the analogous peptide SurlL2 (LLLGEFLKL, SEQ ID NO:4) (light gray column) and the analogous peptide SurlM2 (LMLGEFLKL, SEQ ID NO:5) (dark gray column). Each experiment was performed with 10<5> cells per well in duplicate, and the average number of spots was calculated, Fig. 4 represents T cells that were isolated from tumor-infiltrated lymph nodes from patients Meil, Mel2 and Mel3, stimulated once in vitro and analyzed in an ELISPOT assay for response to no peptide (black bar), the peptides Sur1 (LTLGEFLKL, SEQ ID NO: 10) (gray bar) and Sur9 (ELTLGEFLKL, SEQ ID NO:3) (white bar). Each experiment was performed in duplicate with 10<5> cells per well. In each experiment, two wells without the addition of peptide were also included. The average number of spots per peptide was calculated for each patient, Fig. 5 shows functional activity of survivin-specific CTLs. CTLs were isolated from a melanoma-infiltrated lymph node using survivin-coated magnetic beads. (A) Specific lysis of melanoma cell lines; the HLA-A2-positive FM3 (triangle) and the HLA-A2-negative FM45 (square). (B) Specific lysis of breast cancer cell lines; the

HLA-A2-positive MCF-7 (triangle) and the HLA-A2-negative BT-20 (square),

Fig. 6 shows the frequency of survivin-reactive CTVs in PBL from breast cancer patients. Reactivity was examined in three breast cancer patients (top, middle and bottom, respectively) by ELISPOT. For each patient, the assays were performed in the absence of peptide, in the presence of surl peptide, in the presence of sur9 and in the presence of the modified surlM2 peptide. 1 x IO<4>effector cells per well were used. The graph depicts the quantification of reactive cells, gray columns represent the average number of IFN-γ-producing cells, Fig. 7 illustrates HLA-35 binding of survivin-derived peptides and analysis of the peptide-mediated recovery of HLA-B35 molecules by survivin- derived peptides. Lysates of metabolically labeled T2-B35 cells were incubated at 4°C in the presence of 50, 5, 0.5, 0.05 and 0.005 mM peptide. The recovery of HLA-B35 was analyzed in a complex binding assay and quantified after IEF gel electrophoresis using ImageGauge phosphorimager software (FUJI photofilm Co., Ltd., Japan). The C50 value is the concentration of the peptide required to achieve half of maximal binding to HLA-B35, Fig. 8 shows spontaneous T-cell responses observed in PBLs from cancer patients. A) Number of IFNγ spot-forming cells measured in ELISPOT assay without peptide (white bars), with sur51-59 (black bars) or sur46-54 (grey bars), among in vitro stimulated PBLs from patient CLL5 (IO<5 >cells/well), HEM12 (10<5>cells/well) and HEM8 (5xl0<4>cells/well). B) Number of spot-forming cells among 1.7x10<5>PBLs from HEM12 cultured for 10 days with peptide-pulsed mature autologous dendritic cells. The columns represent the average of two measurements, Fig. 9 shows spontaneous T-cell responses to natural and modified survivin peptides in melanoma patients. A) Number of spot-forming cells measured by ELISPOT assay against sur51-59 and sur51Y9 from patient FM25 in PBLs (4xl0<3>cells/well) and TILs (7xl0<4>cells/well) as well as TIL' is from FM45 (IO<4>cells/well). B) Number of spot-forming cells measured by ELISPOT assay against sur46 and sur46Y9 measured in TILs from FM74 (5xl0<3>cells/well). The columns represent the mean of two measurements with the non-specific IFNγ release subtracted, Fig. 10 illustrates binding affinity of survivin-derived peptides to HLA-A1. The bands of class I MHC heavy chains were quantified on a Phosphorimager. The amount of stabilized HLA-A1 heavy chains is directly related to the binding affinity for the added peptide. The peptide-mediated recovery of HLA-A1 (arbitrary units) induced by 40, 4, 0.4, 0.04 uM of Sur93-101 (line), Sur93T2 (square), Sur49-58 (circle) or Influenza A, PB1 591 -599 (triangle), Fig. 11 shows spontaneous responses to HLA-Al-restricted peptides. Spontaneous T-cell responses to survivin-derived peptides measured by ELISPOT assay. The mean number of peptide-specific IFNγ spots formed in response to Sur92-101, Sur38Y9, Sur47Y10 and Sur93T2 among 5xl0<4>in vitro stimulated PBL or TIL from melanoma patients. The peptide-specific responses shown were observed among analyzes of 6 PBL samples and 3 TIL samples from melanoma (Med patients. Non-specific IFNγ spots are subtracted. Bars: extent of duplicates, Fig. 12 shows spontaneous responses against HLA-A11-restricted peptides Spontaneous T-cell responses to survivin-derived peptides measured by the ELISPOT assay The mean number of peptide-specific IFNγ spots formed in response to Sur53-62 among 5xl0<4>in vitro stimulated PBL or TIL from cancer patients. The peptide-specific responses shown were observed among analyzes of 5 melanoma (Mel) patients (5 PBL, 1 TIL) and 2 CLL (CLL) patients (PBL). Non-specific IFNγ spots have been subtracted. Bars: extent of duplicates, Fig. 13 illustrates spontaneous responses to HI_A-A3-restricted peptides Spontaneous T cell responses to survivin-derived peptides as measured by the ELISPOT assay The mean number of peptide-specific IFNγ spots formed in response to Surl8K10 among 5xl0<4> in vitr o stimulated PBL or TIL from melanoma patients. The peptide-specific responses shown were observed among analyzes of 23 PBL samples and TIL samples from melanoma (Mel) patients. Non-specific IFNγ spots are subtracted. Bars: extent of duplicates, Fig. 14 illustrates spontaneous responses to HLA-A2-restricted peptides. Spontaneous T cell responses to survivin-derived peptides measured by the ELISPOT assay. The average number of peptide-specific IFNγ spots formed in response to the llmer peptide, Surl8-28 among 5xl0<4> in vitro stimulated PBL from cancer patients. The peptide-specific responses shown were observed among analyzes of 10 PBL samples from 2 melanoma (Mel), 6 CLL (CLL) and 2 breast carcinoma (MC) patients. Non-specific IFNγ spots are subtracted. Bars: extent of duplicates, Fig. 15 illustrates spontaneous T cell responses to survivin-derived peptides measured by the ELISPOT assay. The mean number of peptide-specific IFNγ spots formed in response to sur6-14 (LPPAWQPFL) among IO<5>in vitro stimulated PBL from five melanoma patients (mel25, mel26, mel3, mel6, mel39), two CLL patients (CLL1, CLL54) and two breast cancer patients (breastll, breastl5). Non-specific IFNy spots are subtracted, Fig. 16 illustrates the laboratory values for stable detection of LDH, cholinesterase, creatinine, hemoglobin, leukocytes and platelets after vaccination treatment of four patients (ARW, • KN, - WWE, ■ GB) Fig. 17 shows kinetic analysis of immunity for survivin peptides determined by IFNγ ELISPOT. PBMCs were obtained before the first DC vaccination and three months after it. The number of IFNγ spot-forming cells greater than background is described.

The following table lists amino acid sequences for peptides used here and their respective SEQ ID NOs:

EXAMPLE 1

Identification of a cytotoxic T-lymphocyte response to the apoptosis inhibitor protein survivin in cancer patients

Summary

Using CTL epitopes derived from survivin, a study of specific T-cell reactivity against such antigens in peripheral blood from patients with chronic lymphocytic leukemia (CLL) and in tumor-infiltrated lymph nodes from melanoma patients has been carried out by ELISPOT analysis. CTL responses to survivin-derived peptide epitopes were detected in three of six melanoma patients and in three of four CLL patients. No T cell reactivity was detected in PBL from six healthy controls. Survivin-derived peptides can therefore serve as important and widely applicable targets for anticancer immunotherapeutic strategies.

Introduction

The survivin protein was examined for the presence of HLA-A<*>0201 (HLA-A2)-binding peptide motifs, and after successful identification, the peptides were used to test for specific T-cell reactivity in leukemia and melanoma patients by ELISPOT assay. In both patient cohorts, CTL responses to two survivin-derived peptide epitopes were detected, whereas no T-cell reactivity could be detected in the healthy controls. These data suggest that survivin represents a widely expressed tumor antigen that is recognized by autologous T cells.

Material and other methods

Patients and normal controls

Peripheral venous blood samples from 4 patients diagnosed with CLL (called CLL1-4) and blood samples from 6 normal individuals were drawn into heparinized tubes. PBLs were isolated using Lymphoprep separation and frozen in fetal calf serum (FCS) with 10% dimethyl sulfoxide. In addition, T-lymphocytes from tumor-infiltrated lymph nodes were collected from 6 melanoma patients (called mel 1-6). Freshly removed lymph nodes were cut into small fragments, crushed to release cells for culture, and cryopreserved. PBLs were available from 4 of the melanoma patients. All the included subjects were HLA-A2 positive as determined by FACS analysis using the HLA-A2-specific antibody BB7.2. The antibody was purified from hybridoma supernatant. Patient samples were obtained from Copenhagen County Hospital in Herlev, Denmark. Informed consent was obtained from the patients before any of these measurements.

Survivin-derived peptides

All peptides were obtained from Research Genetics (Huntsville, AL, USA) and supplied with

>90% purity verified by HPLC and MS analysis. The peptides used are listed in Table 1.

Complex binding assay for peptide binding to class I MHC molecules

Complex binding assays for binding of the synthetic peptides to class I MHC molecules metabolically labeled with [35S]-methionine were performed as described (12,13). The complex binding assay is based on stabilization of the class I molecules after loading with peptide in the peptide transporter-deficient cell line T2. Then, correctly folded, stable heavy chains of MHC were immunoprecipitated using conformation-dependent antibodies. After IEF electrophoresis, the gels were exposed to phosphorlmager screens, and peptide binding was quantified using the Imagequant Phosphorlmager program (Molecular Dynamics, Sunnyvale,

ABOUT).

Antigen stimulation of PBUs

To extend the sensitivity of the ELISPOT assay, PBLs were stimulated once in vitro before analysis (14,15). Fresh and previously frozen PBLs gave similar results in the ELISPOT assay. On day 0, PBLs or crushed lymph nodes were thawed and seeded in 2 ml/well at a concentration of 2x10<6> cells in 24-well plates (Nunc, Denmark) in AIM medium (Life Technologies, Roskilde, Denmark), 5% heat-inactivated human serum and 2 mM L-glutamine in the presence of 10 µM of peptide. In each experiment, one well without peptide was included. Two days later, 300 IU/ml recombinant interleukin-2 (IL-2) (Chiron, Ratingen, Germany) was added to the cultures. The cultured cells were tested for reactivity by ELISPOT assay on day 12.

ELISPOT assay

The ELISPOT assay used to quantify peptide-epitope-specific interferon-γ-releasing effectors was performed as in (16). Briefly, 96-well plates with nitrocellulose bottoms (MultiScreen MAIP N45, Millipore, Hedehusene, Denmark) were coated with anti-IFN-γ antibody (1-D1K, Mabtech, Nacka, Sweden). The wells were washed, blocked with AIM V medium and cells were added in duplicates at various cell concentrations. Peptides were then added to each well and the plates were incubated overnight. The following day, the medium was discarded and the wells were washed before the addition of biotinylated secondary antibody (7-B6-1-Biotin, Mabtech). Plates were incubated for 2 h, washed and Avidin enzyme conjugate (AP-Avidin, Calbiochem, Life Technologies) added to each well. Plates were incubated at RT for one hour and the enzyme substrate NBT/BCIP (Gibco, Life Technologies) was added to each well and incubated at room temperature for 5-10 min. The reaction was terminated by washing with tap water until dark violet spots appeared. Spots were counted using the Alphalmager system (Alpha, Innotech, San Leandro, CA. USA) and the peptide-specific CTL frequency (or number) could be calculated from the number of spot-forming cells. The assays were all performed in duplicate for each peptide antigen.

Results

Binding of survivin-derived peptides to HLA-A2

The amino acid sequence of the survivin protein was screened for the most likely HLA-A2 nona and decamer peptide epitopes using the major HLA-A2-specific anchor residues (17). Ten survivin-derived peptides were synthesized and examined for binding to HLA-A2. An epitope from HIV-1 pol476-484 (ILKEPVHGV, SEQ ID NO: 11) (Table 1) was used as a positive control. The peptide concentration required for half maximal recovery of class I MHC (C50 value) was 0.7 µM for the positive control. For comparison, the peptide with the designation Sur9 (ELTLGEFLKL, SEQ ID NO:3) bound with an affinity of C50 = 10 µM. The peptides respectively designated Sur6 (FLKLDRERA, SEQ ID NO:1) and Sur8 (TLPPAWQPFL, SEQ ID NO:2) bound to HLA-A2 at C50= 30 µM, while Surl (LTLGEFLKL, SEQ ID NO: 10) and Sur3 ( KVRRAIEQL, SEQ ID NO: 13) bound more weakly (C50 > 100 uM). Five of the peptides examined (Sur2, Sur4, Sur5, Sur7 and Sur10) did not bind to HLA-A2.

Since Surl is a weak HLA-A2 binder, two analogous peptides, designated respectively SurlL2 and SurlM2, in which a better anchor residue (leucine or methionine) replaced the natural threonine in position 2, were synthesized. Both of these peptides bind with almost as high affinity to HLA-A2 as the positive control (C50= 1 uM).

CTL response to survivin in CLL patients

PBL/s from four HLA-A2-positive CLL patients were stimulated once in vitro before examination in the ELISPOT assay. This procedure was chosen to extend the sensitivity of ELISPOT. All ten survivin-derived peptides mentioned above were included in the first line of experiments. Responses were detected to Sur1 and Sur9, and only data for these peptides are given in the figures. Figure 1 shows CTL reactivity to Surl and Sur9 as determined in patient CLL1. Each spot represents a peptide-reactive, IFN-γ-producing cell. The average number of spots per peptide was calculated using a CCD scanning device and a computer system. Fifty-two Sur9 peptide-specific spots (after subtraction of spots without added peptide) per 6xl0<5> were detected in patient CLL1 (Fig. 1). No response was detected to the weak HLA-A2-binding peptide Surl, but the patient responded strongly to the strong HLA-A2-binding peptide analogue SurlM2 (35 peptide-specific spots per 10<4> cells) (fig.2). No response was detected to the other strong HLA-A2-binding peptide analogue SurlL2 in this patient (Fig. 2). Patient CLL2 responded strongly to Sur9 (128 peptide-specific spots per IO<5> cells) and weakly to Surl (22 peptide-specific spots per IO<5> cells) (fig.3). The response to the SurlL2 analogue was only slightly increased compared to the natural epitope, while the patient responded similarly strongly to the SurlM2 peptide as to the decamer peptide Sur9. In patient CLL3, a weak response to Sur9 was observed (fig.3). No response to Surl or the modified Surl peptide was observed in the patient. No survivin responses were detected in the last patient, CLL4 (data not shown). PBLs from 6 healthy HLA-A2 positive controls were analyzed to investigate whether a response to survivin could be detected in healthy individuals. No response was observed in any of the controls to any of the survivin-derived peptides.

CTL response to survivin in melanoma patients

T-lymphocytes isolated from tumor-infiltrated lymph nodes of HLA-A2-positive melanoma patients were examined. The newly removed lymph node was cut into small fragments and crushed to release cells for culture. Cells were stimulated once with peptide in vitro before examination in the ELISPOT assay. Survivin-specific T cells were detected in three of the six analyzed patients. A strong Sur9 response was detected in patients Mel2 and Mel3. A weaker response to the Surl peptide was also detected in these patients (Fig. 4). In Meil, the response to the weakly binding peptide Surl was stronger than the response to the stronger HLA-A2 binder Sur9 (Fig.4). No response was detected in the tumor-infiltrated lymph nodes from the last three melanoma patients (Mel4-6). PBLs from two of the survivin-responding patients, Meil and Mel2, and from two of the non-responding patients, Mel4 and Mel5, were examined. No response could be detected to either Sur9 or Sur1 in PBLs from any of these patients (data not shown).

EXAMPLE 2

Spontaneous cytotoxic T-cell responses to survivin-derived MHC class I-restricted T-cell epitopes in situ and ex vivo in cancer patients Abstract

Spontaneous cytotoxic T-cell responses to survivin-derived MHC class I-restricted T-cell epitopes were demonstrated in situ as well as ex vivo in breast cancer, leukemia and melanoma patients. Moreover, survivin reactive T cells isolated using magnetic beads coated with MHC/peptide complexes were cytotoxic to HLA-matched tumors from various tissue types. Being a universal tumor antigen, survivin may serve as a broadly applicable target for anticancer immunotherapy.

Material and methods

Construction of HLA-peptide complexes for T-cell staining and T-cell sorting.

A recognition site for enzymatic biotinylation using biotin-protein ligase (BirA) in fusion with the 5' end of the extracellular domain of HLA A<*>0201 (residues 1-275) was expressed in E. coli BL21 (DE3). The recombinant protein was purified by size (Sephadex G25, Pharmacia) and ion exchange (mono-Q, Pharmacia) chromatography from inclusion bodies solubilized in 8 M urea. HLA A<*>0201 was folded in vitro by dilution in the presence of the modified survivin peptide SurlM2 (LMLGEFLKL, SEQ ID NO:5) or the MAA peptide gpl00154-163, and then biotinylated as described previously (35, 36).

After gel filtration on a Pharmacia Sephadex G25 column to remove unbound biotin, the protein was multimerized with streptavidin-FITC-conjugated dextran molecules (kindly provided by L. Winther, DAKO, Denmark) to generate multivalent HLA-dextran preparations for immunohistochemistry. The HLA A<*>0201 construct was a kind gift from Dr. Mark M. Davis (Dept. of Microbiology and Immunology, Stanford University, Palo Alto, CA). Cell separation was performed as previously described (37). Briefly, 5 x 10<6>streptavidin-conjugated magnetic beads (Dynal, Oslo, Norway) were washed twice in 200 µl cold PBS, 0.5 µg peptide/A<*>0201 monomers were added and the mixture incubated for 15 min at room temperature. After two washes, these beads were mixed with PBLs at a ratio of 1:10 and then incubated for 1 hour followed by precipitation of bead-bound cells in a magnetic field. The precipitation step was repeated once.

Immunohistochemical stainings

For staining with FITC-conjugated, multimeric peptide/MHC complexes, tissue sections were dried overnight and then fixed in cold acetone for 5 min. All incubation steps were performed at room temperature and in the dark: (i) 45 min. of the primary antibody (diluted 1:100), (ii) CY 3-conjugated goat anti-mouse (diluted 1:500, code 115-165-100, Jackson ImmunoResearch, obtained from Dianova, Hamburg, Germany) for 45 min. and finally (iii) the multimers for 75 min. Between each step, the slides were washed twice for 10 min. in PBS/BSA 0.1%. The slides were placed in "vectashield" and stored in a refrigerator until they were observed under the confocal microscope.

Cytotoxicity Assay

Conventional [51Cr] release assays for CTL-mediated cytotoxicity were performed as described in (13). Target cells were autologous EBV-transformed B cell lines, the HLA-A2-positive breast cancer cell line MCF-7 (available from ATCC), the HLA-A2-positive melanoma cell line FM3 (38), the HLA-A2-negative breast cancer cell line BT-20 (available from from ATCC) and the HLA-A2-negative melanoma cell line FM45 (38). All cancer cell lines expressed survivin, examined by RT-PCR (data not shown).

ELISPOT assay

The ELISPOT assay was used to quantify peptide epitope-specific IFN-γ-releasing effector cells and has been described previously (39). Briefly, nitrocellulose-bottomed 96-well plates (MultiScreen MAIP N45, Millipore) were coated with an anti-IFN-γ antibody (1-DIK, Mabtech, Sweden), and non-specific binding was blocked using AIM V (GibcoBRL, Life Technologies Inc., Gaithersburg, MD, USA). Lymphocytes were added at various cell concentrations together with the specific peptides and T2 cells and incubated overnight at 37°C. After two washes, the biotinylated detection antibody (7-B6-1-Biotin, Mabtech) was added. Specific binding was visualized using alkaline phosphatase-avidin together with the respective substrate (GibcoBRL). The reaction was terminated by the appearance of dark violet spots, which were quantified using the Alphalmager system (Alpha Innotech, San Leandro, CA, USA). The peptides used in the ELISPOT were Surl, Sur9 and the Surl analog peptide SurlM2 as described in Example 1.

Results

In situ staining of HLA-A2/survivin-reactive T cells

In Example 1, two survivin-derived epitopes recognized by T cells in leukemia and melanoma, i.e. Surl, were identified. The weak binding affinity of Surl to HLA-A2 was significantly improved by replacing threonine at position 2 with a better anchor residue (methionine; SurlM2). This measure made it possible to construct stable HLA-A2/peptide complexes. These complexes were multimerized using dextran molecules, which were conjugated with streptavidin and FITC. Multimerized MHC complexes were used to stain acetone-fixed frozen material. Using a confocal laser microscope, SurlM2/HLA-A<*>0201-reactive CTVs could be easily detected in situ in the tumor microenvironment. We described such cells in the primary tumor and in the sentinel lymph node in a patient with stage III melanoma as well as in a primary breast cancer lesion. To ensure the specificity of the staining, a series of negative controls was carried out. Neither the use of peptide/HLA-dextran multimers with peptides derived from the melanoma differentiation antigen gp100 on the same tumor nor SurlM2/HLA-dextran multimers in the case of a tumor sample taken from an HLA-A2-negative donor resulted in a positive staining.

Isolated survivin-reactive CTUs (?) clear tumor cell lines of other origins

To characterize the functional capacity of survivin-reactive CTVs, these cells were isolated using magnetic beads coated with HLA-A2/SurlM2 complexes (36). A recently resected melanoma-infiltrated lymph node was cut into small fragments and crushed to release cells for culture. Cells were stimulated once with peptide in vitro before isolation. One day after isolation, IL-2 was added, and on day 5 the capacity of these cells to kill tumor cells was tested either by ELISPOT or in standard 51 Cr binding assays. First, by means of ELISPOT analysis it was possible to establish that CTVs isolated using the modified SurlM2/HLA-A2 complex also responded to the native Surl peptide (data not shown). Second, the cytotoxicity of the survivin-reactive CTLs against the HLA-A2-positive melanoma cell line FM 3 (Fig. 5A) and the HLA-A2-positive breast cancer cell line MCF-7 (Fig. 5B) was tested. The isolated T cells efficiently lysed both HLA- A<*>0201 cell lines. In contrast, no cytotoxicity was observed against the HLA-A2-negative melanoma cell line FM45 (Fig. 5A) or the HLA-A2-negative breast cancer cell line BT-20 (Fig. 5B).

Survivin reactivity measured in PBL with ELISPOT

The occurrence of survivm-reactive T cells in PBLs from ten HLA-A2-positive breast cancer patients was investigated with ELISPOT. Before analysis, PBLs were stimulated once in vitro to extend the sensitivity of the assay. Reactivity to the following survivin peptides was examined: Surl, Sur9 and SurlM2. Survivin-specific T cells were detected in six of the ten HLA-A2-positive breast cancer patients. Representative examples are given in fig.6. In PBLs from two patients, a response to Surl and the modified analogue Sur 1M2 was detected, but not to Sur9 (Fig. 6, top, middle), in three patients a response to Sur9 was detected, but not to Surl and SurlM2 (Fig.6, bottom), and one patient responded only to SurlM2. In contrast, no survivin responses were detected in PBLs from 20 healthy HLA-A2-positive donors. Likewise, PBLs from fourteen HLA-A2-positive melanoma patients were examined. Survivin responses were present in seven of these patients (Table 2). Two patients responded to the Sur9 peptide, three to the SurlM2 peptide, one to both Surl and SurlM2 and one to all three peptides. In Example 1, T cell responses to survivin in 3 chronic lymphocytic leukemia (CLL) patients were tested (Table 2; CLL1, CLL2, CLL3). These studies were extended using PBLs from three other CLL patients. Strikingly, all patients produced a T-cell response to at least one survivin epitope (Table 2, CLL5, CLL6, CLL7). Also, PBLs from a patient suffering from chronic myeloid leukemia (CML) were investigated. In this patient, a response to all three peptides was identified (data not shown). The data are summarized in table 2.

a) Frequency of reactive cells per IO4; 14 patients examined.

b) Frequency of reactive cells per IO<4>; 10 patients examined.

c) Frequency of reactive cells per IO<5>; 7 patients examined.

EXAMPLE 3

HLA-B35-restricted immune responses to survivin-derived peptides in cancer patients

Summary

In this study, two survivin-derived epitopes that are restricted to HLA-B35 were identified and characterized. Specific T-cell reactivity against both of these epitopes was present in peripheral blood from patients with various hematopoietic malignancies and melanoma. Substitutions of anchor residues at the C-terminus improved the recognition of tumor-infiltrating lymphocytes from melanoma patients. Furthermore, spontaneous cytotoxic T-cell responses to survivin were shown in situ in a primary melanoma lesion. These epitopes expand the applicability of future vaccine strategies based on survivin peptides in relation to malignancies as well as the HLA profile of the patients involved.

In Examples 1 and 2, HLA-A2-restricted, survivin-derived T-cell epitopes were studied. Since HLA-A2 is only expressed in approximately 30% of the Caucasian population (63), peptide epitopes restricted to other HLA class I molecules must be identified to increase the fraction of treatable patients. In this study, two novel survivin T-cell epitopes restricted to HLA-B35, which are expressed in 9% of the Caucasian population (63), were identified, and spontaneous immune responses to these survivin peptides were detected in patients with various hematopoietic malignancies and melanoma.

Material and methods

Patients

Peripheral venous blood samples from cancer patients were collected, PBLs were isolated using Lymphoprep separation (Department of Clinical Immunology, University Hospital, Copenhagen) and frozen in FCS with 10% DMSO. Ten HLA-B35-positive patients were selected for further analysis. These patients suffered from melanoma, CLL, follicular lymphoma (FL), diffuse large B-cell lymphomas (DLBCL) and multiple myeloma (MM), respectively. At the time of collection of the blood samples, the patients had not received medical treatment during the previous four months. In addition, tumor infiltrating lymphocytes (TIL) were isolated from lymph nodes collected from three of the melanoma patients and frozen in FCS with 10% DMSO.

Peptides

Seven synthetic survivin-derived peptides were used in this study: Sur6-14, Surll-19, Sur34-43, Sur46-54, Sur51-59, Sur46Y9, Sur51Y9, and an EBV-derived peptide, EBNA3A 457-466 (63). All peptides were obtained from Research Genetics

(Huntsville, AL) and supplied with >90% purity, verified by HPLC and MC analysis. The peptides are listed in Table 3 below.

Complex binding assay for peptide binding to MHC class I molecules

The complex binding assay described in Examples 1 and 2 was used to measure the binding affinity of the synthetic peptides to the HLA-B35 molecules, which were metabolically labeled with [S35]methionine. Briefly, the assay is based on peptide-mediated stabilization of empty HLA molecules released, after cell lysis, from the TAP-deficient cell line 12, stably transfected with HLA-B35 (kindly provided by Dr J. Haurum, Symphogen ApS, Lyngby, Denmark). Stably folded HLA molecules were immunoprecipitated using the conformation-dependent mAb W6/32. The HLA molecules were separated by IEF electrophoresis, gels were exposed to phosphorimager screens (Imaging plate, FUJI photo film Co., LTD., Japan), analyzed and the amount of correctly folded HLA molecules was quantified using ImageGauge phosphorimager software (FUJI photo film Co., LTD., Japan).

Antigen stimulation of PBL' is

To extend the sensitivity of the ELISPOT assay, lymphocytes were stimulated once in vitro with peptide before analysis (14, 15). PBLs or TILs were thawed and stimulated with 50 µM of the individual peptide epitopes in 96-well plates for 2 h at 26 °C

(5xl0<5->10<6> cells per peptide) and pooled for a further 10 days of culture at 37°C in x-vivo with 5% human serum (HS), in 24-well plates (Nunc, Roskilde, Denmark ) with 2xl0<6> cells per well. On the second day of incubation, 40 µg/ml IL-2 (Apodan A/S, Denmark) was added. On day 10, the cultured cells were tested for reactivity by ELISPOT assay.

The ELISPOT assay

The ELISPOT assay used to quantify peptide-specific, IFN-γ-releasing effector cells in PBLs or TILs collected from cancer patients was performed as described in Example 1. Briefly, nitrocellulose-bottomed 96-well plates (MultiScreen MAIP N45; Millipore, Hedehusene, Denmark) coated with mAb against human IFN-γ, 7.5 ug/ml (1-DIK; Mabtech, Nacka, Sweden). The wells were washed and blocked in x-vivo (x-vivo 15TM BioWhittacker, Molecular Applications Aps, Denmark) and cells were added in duplicates at different concentrations. For antigen presentation, 10<4>T2-B35 cells, with and without 10 µM peptide, were added per well. Plates were incubated overnight, cells discarded and wells washed before addition of biotinylated secondary antibody (7-B6-l-Biotin; Mabtech). The plates were incubated for 2 hours at room temperature, washed, and avidin-alkaline phosphatase conjugate was added (AP-Avidin, Calbiochem, Life Technologies, Inc.). After an hour's incubation at room temperature, the enzyme substrate nitroblue tetrazolium/5-bromo-4-chloro-3-indolyl phosphate (Code No.K0598, DakoCytomation Norden A/S) was added, and dark violet spots appeared in 3-7 min. This reaction was terminated by washing with tap water. Spots were counted using the Alpha Imager System (Alpha Innotech, San Leandro, CA), and the frequency of peptide-specific T cells was calculated from the number of spot-forming cells.

All assays were performed in duplicate for each peptide antigen, and lymphocytes cultured in the same well were tested in the same cell numbers with and without peptide to measure the number of peptide-specific cells in the culture.

Maturation of dendritic cells (DCs)

Adherent cells were isolated from PBLs after 2 h of culture. These were cultured for a further 10 days in RPMI 1640 (GibcoTM Invitrogen corporation, UK) with 10%

FCS. 800 ng/ml GM-CSF (PreproTech, London, UK) and 40 ng/ml IL-4 (Prepro-Tech) were added every 3 days. On day 10, DCs were matured for 24 h by adding 50 ng/ml TN F-a (PreproTech). After maturation, DCs were released and pulsed with 20 µM peptide in the presence of 3 µg/ml 32-microglobulin for 2 hours at 26°C.

Isolation of peptide-specific T cells

Antigen-specific T cells were isolated using sur51Y9/HLA-B35-coated magnetic beads as described in Example 2. Biotinylated monomers of HLA-B35 with sur51Y9 (obtained from Prolmmune, Oxford, UK) were coupled to streptavidin-coated magnetic beads ( Dynabeads M-280, Dynal A/S, Oslo, Norway) by incubating 2.5 µg monomers with 5 x IO<6>beads in 40 µl PBS for 20 min. at room temperature. The magnetic complexes were washed three times in PBS using a magnetic device (Dynal A/S, Oslo, Norway) and then mixed with PBLs at a ratio of 1:10 in PBS with 5% BSA and rotated very gently in one hour. Antigen-specific CD8<+->T cells associated with the magnetic complexes were gently washed two or three times. Isolated cells were resuspended several times in x-vivo supplemented with 5% human serum and incubated for 2 h before the magnetic beads were released and removed from the cell suspension. The isolated antigen-specific CD8<+->T cells were used in ELISPOT assay to analyze the cross-reactivity between the natural and the modified peptide.

TCR clonotype mapping by denaturing gradient gel electrophoresis (DGGE)

DGGE clonotype mapping of the human TCR BV regions 1-24 has been described in detail (66). Briefly, RNA was isolated using the Purescript Isolation Kit (Gentra Systems Inc. MN) and transcribed cDNA was amplified by PCR using primers for the variable regions of the TCR beta chains in combination with a common constant region primer. The computer program MELT87 was used to ensure that the amplified DNA molecules were suitable for DGGE analysis provided that a 50 bp GC-rich sequence (GC-clamp) was ligated to the 5' end of the constant region primer. DGGE analysis was done in 6% polyacrylamide gels containing a gradient of urea and formamide from 20% to 80%. Electrophoresis was performed at 160 V for 4.5 h in 1 x TAE buffer at a constant temperature of 54 °C.

Immunohistochemical stainings

Multimerized peptide/HLA complexes were used to identify antigen-specific T cells in situ in tumor lesions of cancer patients using the procedure described in Example 2. Biotinylated sur51Y9/HLA-B35 monomers were supplied by Proimmune limited, Oxford, UK. The biotinylated monomers of sur51Y9/HLA-B35 were multimerized with streptavidin-FITC-conjugated dextran molecules (kindly provided by L. Winther, DAKO, Glostrup, Denmark) to generate multivalent HLA-dextran compounds for immunohistochemistry. Tissue pieces were dried overnight and then fixed in cold acetone for 5 min. All incubation steps were performed in the dark at room temperature: (a) 45 min with the primary antibody (diluted 1:100) (b) Cy 3-conjugated goat anti-mouse antibody (diluted 1:500; code 115-165-100, Jackson ImmunoResearch, obtained from Dianova, Hamburg, Germany) for 45 min, and finally (c) the multimers for 75 min. Between each step, the slides were placed in vectashield and stored in a refrigerator until they were observed under the confocal microscope (Leica).

Results

Identification of HLA-B35-binding survivin-derived peptides

The amino acid sequence of survivin was screened for nonameric and decameric peptides with anchor residues consistent with the peptide binding motif in HLA-B35 (67). Five

peptides were selected that contained proline as the N-terminal anchor in position 2 and phenylalanine, leucine, isoleucine or tyrosine as C-terminal anchor residues (Table 3). Complex binding assay revealed two peptides, sur51-59 (EPDLAQCFF, SEQ ID NO:7) and sur46-54 (CPTENEPDL, SEQ ID NO:6), which had the ability to effectively stabilize HLA-B35. In addition, two peptides, sur34-43 (TPERMAEAGF, SEQ ID NO:20) and sur6-14 (LPPAWQPFL, SEQ ID NO:18) showed a weak stabilization, while the remaining peptide did not stabilize HLA-B35 at all. The peptide concentration required for half-maximal recovery of HLA-B35 (C50) was estimated to be 13 µM for sur51-59 and 20 µM for sur46-54. In comparison, the positive control epitope C24 from EBNA3A458-466 (YPLHEQHQM, SEQ ID NO:21) had an estimated C50 value of 0.8 µM.

To increase the binding affinity for sur46-54 and sur51-59, the amino acid at the C-terminus was replaced with tyrosine, a better anchor residue (67). The recovery of HLA-B35 mediated by the modified peptides was analyzed by complex binding assay, and C 50 values were estimated at 1.5 µM for sur51Y9 and 4 µM for sur46Y9 (Fig. 7).

Spontaneous immune responses against natural peptide epitopes

Initially, five patients were analyzed for spontaneous immune responses to the four natural HLA-B35 binding peptides sur51-59, sur46-54, sur34-43 and sur6-14. These five patients had different hematopoietic malignancies: HEM8 and HEM18 suffered from MM, HEM12 from FL, HEM9 had DLBCL and CLL5 had CLL.

IFN-γ ELISPOT assays were performed on PBLs after 10 days of in vitro stimulation to detect peptide precursor CTVs. Spontaneous immune responses were detected against two of the natural HLA-B35 binding peptides, sur51-59 and sur46-54. Two patients, HEM12 and CLL5, showed a response to both sur51-59 and sur46-54, while HEM8 only showed a response to sur51-59 (Figure 8A and B). No response could be detected in the two remaining patients, HEM9 and HEM18, and no response could be detected to the very weakly binding peptides sur34-46 and sur6-14 in any of the patients.

An alternative approach to in vitro stimulation was used in patient HEM12, i.e. PBLs were co-cultured with matured autologous dendritic cells pulsed with sur51-59 to stimulate a CTL response in vitro. PBLs from this culture showed strong reactivity towards sur51-59 in ELISPOT (Fig. 8B).

Increased recognition of modified peptides

As described above, peptide modifications to increase HLA-B35 affinity resulted in a 5- to 10-fold higher affinity for HLA-B35 compared to the natural peptides. A group of five melanoma patients was analyzed for spontaneous immune responses to both the natural and the modified peptides by ELISPOT assay. PBL samples were analyzed after in vitro stimulation, while TIL samples were analyzed directly. Spontaneous immune responses were observed in either PBLs or TILs from three of the five patients. FM25 showed reactivity against sur51-59 and sur51Y9 in both PBL and TIL samples (Fig. 9A). FM45 responded only to the modified peptide sur51Y9, with a strong response detectable in TILs. No PBLs were available from this patient (Fig. 9A). FM74 showed a strong response to sur46Y9 in TIL, but no response to the native peptide was detectable (Figure 9B). A weak response to sur46Y9 was also observed in PBLs from FM74 (data not shown).

Cross-reactivity between the natural and the modified peptide

The high affinity of sur51Y9 for HLA-B35 makes it possible to produce stable monomers of HLA-B35 with sur51Y9. After establishing the presence of survivin-reactive T-lymphocytes in tumor-infiltrated lymph nodes and PBLs from various cancer patients, magnetic beads were coated with such HLA-B35/Sur51Y9 complexes and these were used to isolate survivin-reactive T-lymphocytes from PBLs patient CLL5. This patient showed a strong response to sur51-59. Beads were strongly bound to the cell surface of the specific cells, visualized by microscopy (data not shown), allowing precipitation of antigen-specific cells in a magnetic field. The isolated sur51Y9-specific cells responded strongly to sur51-59 (Figure 9), whereas no response could be detected in the remaining PBLs (data not shown). The isolate was analyzed by the RT-PCR/DGGE-based TCR clonotype mapping. This technique allows analysis for T-cell clonality in complex cell populations, even if only a small number of cells are available. These analyzes showed that 8 different clones were isolated (data not shown).

Antigen-specific T cells present in situ in melanoma lesions

Sur51Y9/HLA-B35 monomers were multimerized using dextran molecules conjugated with streptavidin and FITC. Multimerized MHC complexes were used to stain acetone-fixed frozen material using the procedure described in Example 2. Antigen-specific cells were visualized using a confocal laser microscope. Sections of primary melanoma from three patients were analyzed, and Sur51Y9/HLA-B35-reactive CTLs could be easily detected in situ in the tumor microenvironment in one of the patients. Co-staining with a mAb against granzyme B showed that these survivin-specific CTLs released granzyme B and exhibited cytotoxic activity, HLA-B35-negative melanoma patients were used as controls (data not shown).

EXAMPLE 4

Identification of novel survivin-derived CTL epitopes with different HLA-A restriction profiles

Summary

New HLA-A1-, HLA-A2-, HLA-A3- and HLA-A11-restricted survivin epitopes were characterized on the basis of CTL responses in cancer patients. These epitopes significantly increase the number of patients eligible for immunotherapy based on survivin-derived peptides. In addition, the collective targeting of multiple residue elements is likely to reduce the risk of immune failure due to HLA allele loss.

Materials and methods

Patients

Patient samples were received from the University of Wurzburg, Germany and Copenhagen County Hospital in Herlev, Denmark. Informed consent was obtained from the patients before any of these measurements. Tissue typing was performed by the Department of Clinical Immunology, University Hospital, Copenhagen, Denmark. Peripheral blood lymphocytes (PBL) from cancer patients with melanoma, breast cancer and chronic lymphocytic leukemia (CLL) were isolated using Lymphoprep separation and frozen in fetal calf serum (FCS) with 10% dimethylsulfoxide. Furthermore, T-lymphocytes were obtained from primary lesions and from tumor-infiltrated lymph nodes from melanoma patients. Freshly resected tumor tissue was cut into small fragments and crushed to release tumor infiltrating lymphocytes (TIL) for cryopreservation.

Peptides

All peptides were purchased from Invitrogen (Carlsbad, CA, USA) and supplied with >80% purity, verified by HPLC and MS analysis. All peptides that were used are listed in Table 4 under Example 5 below.

Cell lines

The human T2 cell line is a TAPI- and TAP2-deficient hybrid of the B-LCL.174 and T-LCL CEM cells and therefore expresses only low levels of HLA class I molecules (HLA-A<*>0201 and HLA-B<*>5101) on the cell surface. T2 cells transfected with HLA-A<*>0301 were kindly provided by Dr A McMicheael, IMM, John Radcliffe Hospital, Oxford. T2 cells transfected with HLA-A<*>1101 were kindly provided by Dr M Masucci, MTC, Karolinska Institutet, Stockholm, Sweden. The BM36.1 cell line is also deficient in TAP function and has a similar phenotype to T2 with low expression of HLA class I (HLA-A<*>0101, HLA-B<*>3501) on its surface. The BM36.1 cells were kindly provided by Dr A Ziegler, Humboldt University, Berlin, Germany.

Complex binding assay for peptide binding to MHC class I molecules

The binding affinity of synthetic peptides (Invitrogen, Carlsbad, CA, USA) to HLA-Al, -A2, A3, or Al 1 molecules metabolically labeled with [<35>S]-methionine was measured in the complex binding assay, as described previously ( 12). The analysis is based on peptide-mediated stabilization of empty HLA molecules released by cell lysis from the TAP-poor cell lines. Stably folded HLA molecules were immunoprecipitated using the HLA class I-specific conformation-dependent mAb W6/32 and separated by isoelectric focusing (IEF) gel electrophoresis. The bands for the MHC heavy chains were quantified using the ImageGauge Phosphorimager program (FUJI photo film Co., Carrollton, TX, USA). Band intensity is proportional to the amount of peptide-bound class I NIHC complex recovered during the assay. Consequently, the degree of stabilization of the HLA molecule is directly related to the binding affinity of the added peptide. The peptide concentrations used to analyze the recovery of the HLA molecules were 40, 4, 0.4, 0.04 µM for HLA-A1 and HLA-A11 and 100, 10, 1, 0.1, 0.01 µM for HLA-A2 and HLA-A3. The C50 value was then calculated for each peptide as the sufficient peptide concentration to achieve half of maximal stabilization.

Antigen stimulation of PBL

To extend the sensitivity of the ELISPOT assay, PBL were stimulated once in vitro before analysis. On day 0, PBL or crushed lymph nodes were thawed and plated with 2 x 10<6> cells in 2 ml/well of 24-well plates (Nunc, Roskilde, Denmark) in x-vivo medium (Bio Whittaker, Walkersville, Maryland), 5% heat-inactivated human serum and 2 mM L-glutamine in the presence of 10 µM peptide. Two days later, 20 IU/ml recombinant interleukin-2 (IL-2) (Chiron, Ratingen, Germany) was added to the cultures. The cultured cells were tested for reactivity in ELISPOT on day 10.

ELISPOT assay

The ELISPOT assay was used to quantify peptide epitope-specific interferon-γ-releasing effector cells as previously described (16). Briefly, 96-well plates with nitrocellulose bottoms (MultiScreen MAIP N45, Millipore, Hedehusene, Denmark) were coated with anti-IFN-γ antibody (1-DIK, Mabtech, Nacka, Sweden). The wells were washed, blocked with x-vivo medium and the cells added in duplicates at different cell concentrations. The peptides were then added to each well and the plates were incubated overnight. The following day, the media was discarded and the wells were washed before the addition of biotinylated secondary antibody (7-B6-l-Biotin, Mabtech). The plates were incubated for 2 h, washed, and avidin-alkaline phosphatase conjugate (Calbiochem, Life Technologies, Inc. San Diego, CA, USA) was added to each well. The plates were incubated at room temperature for one hour, washed, and the enzyme substrate NBT/BCIP (DakoCytomation Norden A/S Glostrup, Denmark) was added to each well and incubated at RT for 5-10 min. After the appearance of dark violet spots, the reaction was terminated by washing with tap water. Spots were counted using the ImmunoSpot® Series 2.0 Analyzer (CTL Analyzers, LLC, Cleveland, USA), and the peptide-specific CTL frequency could be calculated from the number of spot-forming cells. All assays were performed in duplicate for each peptide antigen.

Results

Identification of HLA-Al-purified survivin epitopes

Binding of survivin-derived peptides to HLA-A1

The amino acid sequence of the survivin protein was screened for the most likely HLA-A1 nonamer or decamer peptide epitopes using the major HLA-A1 anchor residues, aspartic acid (D), glutamic acid (E) at position 3 and tyrosine (Y), phenylalanine ( F) at the C-terminal. Accordingly, six survivin-derived peptides were synthesized and tested for binding to HLA-A1 (Table 4). In addition, they stayed

two peptides Sur38-46 (MAEAGFIHC)(SEQ ID NO:23) and Sur47-56 (PTENEPDLAQ)

(SEQ ID NO:25) included, despite containing only one of the major anchors, since both were identified as possible good binders by the predictive algorithm of Rammensee et al, available at http:// syfpeithi. bmi-heidelberq. com/. C 50 values were estimated for each peptide as the peptide concentration required to achieve half-maximal stabilization of HLA-A1 (Table 4). Only one of these peptides, Sur92-101 (QFEELTLGEF) (SEQ ID NO:27), bound with nearly the same high affinity as a known positive control epitope from the Influenza A protein basic polymerase 1 (PB1) (VSDGGPNLY), exemplified in Figure 10. Sur93-101 (FEELTLGEF) (SEQ ID NO:24) had a low binding affinity for HLA-A1, while none of the other analyzed peptides bound to HLA-A1 (Table 4). Accordingly, we synthesized a series of analogous peptides in which better anchor residues replaced the natural amino acids. We modified the two peptides Sur38-46 (MAEAGFIHC)

(SEQ ID NO:23) and Sur47-56 (PTENEPDLAQ) (SEQ ID NO:25) by introducing tyrosine (Y) instead of cysteine (C) or glutamine (Q) at the C-terminus, respectively. Both modified peptides bound strongly to HLA-Al (Table 4). In addition, we substituted the amino acids at position 2 with the auxiliary anchors threonine (T) or serine (S) in the two peptides Sur92-101 and Sur93-101. These modifications had no positive effect on the binding of Sur92-101 to HLA-A1. In contrast, tied

Sur93T2 (FTELTLGEF) (SEQ ID NO:36) showed high affinity to HLA-Al (table 4).

Figure 10 illustrates the binding of the natural low affinity peptide Sur93-101, the high affinity modified peptide Sur93T2 and the non-binding peptide Sur49-58 compared to the positive control epitope from influenza. Finally, we modified Surl4-22, Sur34-43, Sur49-58, Sur51-59, Sur92-101 and Sur93-101 with tyrosine (Y) at the C-terminus, but this did not improve the binding affinity of HA-A1 for any of these the peptides (data not shown).

HLA-Al-restricted CTL responses against survivin-derived peptides in cancer patients

PBL from six melanoma patients and TIL from three melanoma patients were analyzed for the presence of CTLs specific to any of the four high-affinity survivin-derived peptides Sur38Y9, Sur47Y10, Sur92-101 and Sur93T2 by ELISPOT. T-cell reactivity against at least one of the survivin-derived peptides was observed in three PBL samples and one TIL sample from a total of nine analyzed patients. As can be seen from Figure 11, PBL from one patient, Mel.Al-3, contained a T cell response against all four peptides, Sur38Y9, Sur47Y10, Sur92-101 and Sur93T2. Mel.Al-2 showed responses against Sur47Y10, Sur92-101 and Sur93T2, while in Mel.Al-1/TIL and Mel.Al-4/PBL responses were observed against Sur47Y10 and Sur93T2 respectively (Figure 11).

In addition, ten melanoma patients were tested for immune reactivity against the natural peptides Sur93-101, Sur38-46 and Sur47-56 using ELISPOT; however, no peptide-specific responses were detected in any of these patients (data not shown).

Identification of HLA-All-purified survivin epitopes

Binding of survivin-derived peptides to HLA-A11

The amino acid sequence of the survivin protein was screened for nonameric and decameric peptides with binding motifs corresponding to the sequence of the HLA-A3 superfamily, including HLA-A3 and HLA-A11. Peptide sequences with the main anchor motifs, leucine (L) at position 2 and lysine (K) at the C-terminus, were selected together with peptide sequences that had related amino acids at these positions in accordance with the predictive algorithm of Ramensee et al (Table 4).

Thirteen peptides were predicted from the survivin protein sequence and analyzed for binding to HLA-A11 and HLA-A3. Three of these peptides, Sur53-62 (DLAQCFFCFK)

(SEQ ID NO:47), Sur54-62 (LAQCFFCFK) (SEQ ID NO:42) and Surll2-120

(KIAKETNNK) (SEQ ID NO:44), bound to HLA-A11 with high affinity, comparable to the viral epitope from EBV nuclear antigen 4 (AVFDRKSDAK) (SEQ ID NO:63). In addition, one peptide, Surll2-121 (KIAKETNNKK) (SEQ ID NO:51), bound weakly to HLA-A11 (Table 4).

HLA-A11-restricted CTL responses to survivin-derived peptides in cancer patients

PBL from five melanoma patients and two CLL patients were tested for T cell reactivity against the four HLA-All binding peptides Sur53-62, Sur54-62, Surll2-120 and Surll2-121. We could detect responses against the survivin-derived peptide Sur53-62 in PBL from two of the melanoma patients, Mel.All-1, Mel.All-2, using ELISPOT (figure 12). Moreover, we could detect Sur53.62-specific T cells among tumor-infiltrating lymphocytes (TIL) from a tumor-infiltrated lymph node in patient Mel.All-2 (Figure 12). In patient Mel.All-1, a strong immune response against the survivin peptide Sur53-62 was observed in five different blood samples taken over a period of two years (data not shown).

Identification of HLA-A3-purified survivin isotopes

Binding of survivin-derived peptides to HLA-A3

The survivin-derived peptides predicted to bind to the HLA-A3 superfamily were additionally analyzed for binding to HLA-A3. Only two of the peptides, Surl 12-120 (KIAKETNNK) (SEQ ID NO:44) and Surll2-121 (KIAKETNNKK) (SEQ ID NO:57), bound HLA-A3 with high affinity, similarly to the viral epitope, Influenza A nucleoprotein 265-273 (ILRGSVAHK) (SEQ ID NO:74)

(Table 4). Furthermore, two peptides, Sur53-62 (DLAQCFFCFK) (SEQ ID NO:47) and Sur95-103 (ELTLGEFLK) (SEQ ID NO:43) (seg) bound weakly to HLA-A3.

Some of the peptides with no detectable binding were modified in an attempt to increase binding affinity for HLA-A3. We therefore synthesized two analogous peptides of Sur54-62 and Surl 13-122 in which a better anchor residue, leucine (L), replaced the natural alanine

(A) at position 2. Sur54L2 (LLQCFFCFK) (SEQ ID NO:56) bound HLA-A3 with high affinity, while Surll3L2 (ILKETNNKKK) (SEQ ID NO:59) only weakly bound

(Table 4). In addition, we synthesized four analogous peptides of Sur5-13, Surl3-22, Surl8-27 and Sur53-61 where the better anchor residue lysine (K) replaced the natural phenylalanine (F) at the C-terminus. Sur5K9 (TLPPAWQPK) (SEQ ID NO:54) and Surl8K10 (RISTFKNWPK) (SEQ ID NO:58) bound to HLA-A3 with high affinity,

while the substitutions had no detectable effect on the binding to HLA-A3 of Surl3K9 (FLKDHRISTK) (SEQ ID NO:57) and Sur53K9 (DLAQCFFCK) (SEQ ID NO:55) compared to the natural analogues.

HLA-A3-restricted CTL responses to survivin-derived peptides in cancer patients

Nine samples from melanoma patients (five PBL and four TIL) were analyzed for immunoreactivity against the two natural high-affinity HLA-3A-binding peptides Surll2-120 and Surll2-121 as well as the two natural weak-binding peptides Sur53-62 and Sur95-103. However, no immune responses against these peptides could be detected by ELISPOT in any of the patients. Subsequently, the same patients were analyzed for spontaneous immunoreactivity against the three high-affinity-modified survivin-derived peptides Sur5K9, Sur18K10 and Sur54L2. CTL reactivity was detected against Sur18K10 in TIL samples from three patients, Mel.A3-1, Mel.A3-2, Mel.A3-3 (Figure 13). No responses were detected against the other two peptides, Sur5K9 and Sur54L2. To further verify these responses, PBL from eighteen other melanoma patients were analyzed for CTL reactivity against Surl8K10. Three responding patients, Mel.A3-4, Mel.A3-5 and Mel.A3-6, were found among these, so that the result was a total of six responding patients among the twenty-seven patients who were analyzed (figure 13).

Identification of a novel HLA-A2-purified survivin epitope

Binding of 11-mer survivin-derived peptides to HLA-A2

The amino acid sequence of the survivin protein was screened for the most likely HLA-A2 11-mer epitopes using the major HLA-A2-specific anchor residues. Six survivin-derived peptides were synthesized and examined for binding to HLA-A2. None of the peptides examined bound with similar high affinity to a known positive control epitope from Epstein-Barr virus BMLF28o-288 peptide (GLCTLVAML) (SEQ ID NO:72) (Table 4). The peptide concentration required for half maximal recovery of HLA-A2 (C 50 value) was 0.9 µM for the positive control. For comparison, the peptides bound Surl8-28 (RISTFKNWPFL)

(SEQ ID NO:67) and Sur86-96 (FLSVKKQFEEL) (SEQ ID NO:69) relate weakly to HLA-

A2 (C50= 69 uM and 72 uM respectively). However, the two known HLA-A2-restricted survivin epitopes similarly bound weakly to HLA-A2, Sur95-104 (ELTLGEFLKL) (SEQ ID NO:43) bound with intermediate affinity (C50= 10 uM), while Sur96- 104 (LTLGEFLKL) (SEQ ID NO: 10) bound only weakly (C50 > 100 µM). The remaining four 11-mer peptides examined (Sur4-14 (PTLPPAWQPFL) (SEQ ID NO:66), Sur54-64 (LAQCFFCFKEL) (SEQ ID NO:68), Sur88-98 (SVKKQFEELTL) (SEQ ID NO: 70) and Surl03-113 (KLDRERAKNKI) (SEQ ID NO:74), did not bind to HLA-A2.

HLA-A2-restricted CTL responses to survivin-derived peptides in cancer patients

PBL from ten cancer patients (two melanoma (Mel), six CLL (CLL) and two breast cancer (MC) patients) were first analyzed to investigate whether the two weakly binding llmer proteins, Surl8-28 and Sur86-96, were presented by HLA-A2 and recognized by the cancer patients' immune system. CTL responses against Surl8-28 were found in PBL from two of the ten patients analyzed (CLL-1, CLL-2, Figure 14), while no response could be detected against Sur86-96 (data not shown). To further verify these Surl8-28-specific responses, PBL from twelve other patients (seven melanoma, one CLL and four breast cancer patients) were analyzed for CTL reactivity against this peptide. Among these, four patients (CLL-3, MC-1, MC-2, Mel.A2-1) had Surl8-28-specific immune reactivity which was detectable by ELISPOT (Figure 14). In total, PBL from six out of twenty-two patients analyzed had a CTL response against Surl8-28.

Identification of HLA-B7 purified survivm epitopes

Binding of survivin-derived peptides to HLA-B7

The amino acid sequence of the survivin protein was screened for peptides of nine to ten amino acids, with anchor residues consistent with the peptide binding motif in HLA-B7. Five peptides were selected and analyzed for their ability to stabilize HLA-B7 in complex binding assay. C 50 values were estimated for each peptide as the peptide concentration required for half-maximal stabilization of HLA-B7 (Table 4). Two survivin-derived peptides, sur6-14 (LPPAWQPFL) (SEQ ID NO: 18) and surl 1-19 (QPFLKDHRI) (SEQ ID NO: 19), weakly stabilized HLA-B7 with C50 values above 100 uM, while sur46 -54 (CPTENEPDL) (SEQ ID NO:6), sur51-59 (EPDLAQCFF) (SEQ ID NO:7) and sur34-43 (TPERMAEAGF) (SEQ ID NO:20) did not bind to HLA-B7 (Table 4 ).

HLA-B7'-restricted CTL responses against survivin-derived peptides in cancer patients

HLA-A7-positive PBL from five melanoma patients (mel25, mel26, mel3, mel6, mel39), two CLL patients (CLL1, CLL54) and two breast cancer patients (breastll, breastl5) were tested for T-cell reactivity against the weak HLA-A7 -binding peptides sur6-14 (LPPAWQPFL) (SEQ ID NO: 18) and surll-19 (QPFLKDHRI) (SEQ ID

NO: 19). We could detect a strong, spontaneous CTL response against the survivin-derived peptide sur6-14 in PBL in a CLL patient and in a breast cancer patient (figure 15). In addition, we could detect a weak response to this peptide in the melanoma patient mel3 (figure 15).

Summary<g>of HLA-allele-purified immune responses to survivin-derived peptides in cancer patients

A series of survivin-derived peptides comprising 9-11 amino acid residues were tested for binding to the following HLA alleles: HLA-A1, HLA-A3, HLA-A11 and HLA-B7 using complex binding assay for peptide binding to MHC class I- molecules described in the previous examples. In addition, several of the peptides were tested for their ability to trigger a CTL immune response using the ELISPOT assay, as also described above.

A summary of the results, including results obtained in the previous examples, is given in the following table 4:

' A response was observed against the peptide Surll2-120 in one lymphoma patient (HEM34) using ELISPOT." Responses were detected against the peptide Sur53-62 in 3 lymphoma patients (HEM9, 11, 34) using ELISPOT. IV A weak response was observed in a melanoma patient (FM-TIL95) using ELISPOT.<v>" A response was observed against Surl 12-121 in a melanoma patient (PM6), most evident in a metastatic lymph node suspension, and weaker in TIL from primary tumor and PBL, using ELISPOT.

<Vl>" A response to the peptide Sur6-14 was observed in a CLL patient (CLL9), and a weaker response was observed in a lymphoma patient using ELISPOT (HEM21)

(data not shown).

EXAMPLE 5

Therapeutic trial procedures using survivin-derived peptides as immunogens

Summary

Five heavily pretreated stage IV melanoma patients were vaccinated with the modified HLA-A2-restricted survivin epitope, namely the surlM2 peptide, presented by autologous dendritic cells in a palliative setting. Four of the patients showed strong T-cell responses to this epitope as measured by ELISPOT assay. Furthermore, in situ peptide/HILA-A2 multimer staining revealed the infiltration of survivin-reactive cells in both visceral and soft tissue metastases. It was striking that no vaccination-associated toxicity was observed. The data show that inducing T-cell responses against survivin is useful, even for late-stage melanoma patients, and that these vaccinations are well tolerated.

Materials and other methods

Patient eligibility and treatment regimen

All clinical treatment regimens were in accordance with the Declaration of Helsinki and all patients gave informed consent before treatment. Patients with stage IV cutaneous or uveal melanoma were eligible if their disease was progressive despite at least two different chemo, immune or chemoimmunotherapy. Patients also had to be 18 years of age or older, express HLAA<*>0201 and suffer from measurable disease validated by cranial, thoracic and abdominal CT examinations. The patients' Karnofsky index had to be 60% or better. No systemic chemotherapy and/or immunotherapy was allowed in the 4-week period before vaccination. Important exclusion criteria were evidence of CNS metastases, active autoimmune or infectious diseases, pregnancy and lactation and significant psychiatric abnormality. Peptide-pulsed dendritic cells were generated as previously described (82). Briefly, PBMCs from liver cachexia were isolated on LymphoprepTM (Nycomed Pharma), frozen in aliquots and stored in liquid nitrogen. One week before vaccination, PBMCs were thawed, washed and cultured in medium containing gentamicin, glutamine and heat-inactivated autologous plasma. On days 1 and 5, IL-4 and GM-CSF were added. To differentiate mature DCs, TNF-γ and prostaglandin E2 were added on day 6. On day 7, when the cells showed phenotypic and morphological characteristics of mature DCs, i.e., an obscured appearance and = 75% CD83 expression, they were pulsed with a modified survivin-derived HLA-A2-restricted survivin96.io4 epitope, LMLGEFLKL (SEQ ID NO: 10) (Clinalfa, Switzerland) 14. Cells were only used for vaccination if microbial tests on samples from cultures taken on days 1 and 5 showed to be sterile.

Patients were vaccinated at 7-day intervals for the first two vaccines followed by 28-day intervals for further vaccinations. A total of 10-20 x IO<6>mature, surviving6-io4-pulsed DCs were resuspended in PBS containing 1% human serum albumin, and injected intradermally in aliquots of 1.5 x IO<6>DCs per injection site into the ventromedial areas of the thighs near the regional lymph nodes. Limbs where draining lymph nodes had been removed and/or irradiated were excluded. Levcapheresis was repeated after 5 vaccinations in the absence of serious deterioration of the patient's health condition or occurrence of CNS metastases.

Measurement of clinical and immunological responses

CT scans were performed before vaccination and every three months thereafter or at severe clinical signs of disease progression. Immunological responses were monitored by ELISPOT assay using PBMCs taken every three months to detect survivin96-io4-specific IFN-γ release. To extend the sensitivity of the ELISPOT assay, PBMCs were stimulated once in vitro at a concentration of 1 x 10 cells per ml in 24-well plates (Nunc, Denmark) in X-vivo medium (Bio Whittaker, Walkersville, Maryland ), supplemented with 5% heat-inactivated human serum and 2 mM L-glutamine in the presence of 10 µM peptide. Two days later, 40 IU/ml recombinant interleukin-2 (IL-2) (Chiron, Ratingen, Germany) was added. After 10 days, the cells were tested for reactivity. For this purpose, 96-well plates with nitrocellulose bottom (MultiScreen MAIP N45, Millipore, Glostrup, Denmark) were coated with an anti-IFN-γ antibody (1-DIK, Mabtech, Sweden). Lymphocytes were added at 10<4>10<5> cells in 200 µl X-vivo medium per well together with 10<4> T2 cells and the relevant peptides at a final concentration of 2 µM. After overnight incubation at 37°C and two washes, the biotinylated detection antibody (7-B6-l-Biotin, Mabtech, Sweden) was added, its specific binding was visualized using alkaline phosphatase-avidin together with the respective substrate (GibcoBRL) . The reaction was terminated by the appearance of dark violet spots, which were quantified using the Alphalmager System (Alpha Innotech, San Leandro, CA, USA).

Survivin96-io4/HLA-A<*>0201-reactive CD8+ T-lymphocytes were also tracked in situ both at the vaccination sites and in visceral, soft tissue or in cutaneous metastases using multimer surviving6-io4/HLA-A<*>0201 -complexes. Vaccination sites were excised 24 h after intradermal injection in all patients, while metastatic lesions were removed only in selected patients, if easily accessible (patients KN and GB), or removed during a curative treatment (patient WW). The staining procedure for multimeric peptide/MHC complexes has been described recently (68). The multimeric survivin96-io4/HLA-A<*>0201 complexes were generated by introducing an enzymatic biotinylation recognition site at the 5' end of the extracellular domain of HLA-A<*>0201 (residues 1-275). The recombinant protein was purified by size exclusion (Sephadex G25, Pharmacia, Erlangen, Germany) and ion exchange chromatography (mono-Q, Pharmacia) and folded in vitro by dilution in the presence of the respective peptides and 82-microglobulin. After gel filtration on a Sephadex G25 column, the protein was multimerized with streptavidin-FITC conjugated with (two) dextran molecules (kindly provided by L. Winther, DAKO, Copenhagen, Denmark) to generate multivalent HLA-dextran complexes. Cryopreserved sections of the respective samples were dried overnight and then fixed in cold acetone for 5 min. All incubation periods were carried out in the dark at room temperature as follows: (i) 45 min. with an anti-CD8 antibody (1:100, clone HIT8a, Pharmingen, San Diego, CA), (ii) Cy3-conjugated goat anti-mouse (diluted 1:500, code 115-165-100, Dianova, Hamburg , Germany) for 45 min and finally (iii) the multimers for 75 min. Between each step, slides were washed twice for 10 min in PBS/BSA 0.1%. Finally, the slides were placed in vectashield and observed under a Leica Confocal microscope (TCS 4D, Leica, Mannheim, Germany).

Results

Patient characteristics and clinical course

Five patients with advanced stage IV melanoma were included, two suffered from uveal melanoma, one from soft tissue melanoma and the remaining two from cutaneous melanoma. Due to the manifestation of symptomatic brain metastasis, one patient was withdrawn from treatment after only two vaccinations. The other four patients received up to 15 vaccinations. One patient died of cardiac arrest in the tumor-free state after surgical resection of residual metastases. Another patient was taken off treatment after 10 vaccinations due to the appearance of visceral metastases (RW). One patient remained in the study after 15 vaccinations. Detailed patient characteristics, previous treatment, number of vaccinations and survival status are summarized in Table 5.

No serious toxicity occurred. Hemoglobin, leukocytes and thrombocytes as well as lactate dehydrogenase, creatinine and cholinesterase were thus not affected by the vaccination treatment (Fig. 16). No signs of systemic or local toxicity were observed at the injection sites. Furthermore, there was no detection of impaired wound healing, hemorrhagic diseases, poor cardiac function, vasculitis or inflammatory bowel syndrome. In one patient (WW), pre-existing liver metastases could be stabilized during the vaccination treatment, but a new adrenal metastasis occurred. Unfortunately, this patient died of cardiac arrest, despite tumor size after curative surgery. A brain metastasis was detected in patient PB only 4 weeks after the start of vaccination. Therefore, this patient had to be excluded from further vaccinations after only two DC injections. The other three patients showed slow progression of metastatic disease without significant deterioration in their general state of health. Remarkably, for patient KN, an overall survival of 13 months (from start of vaccination to death) could be achieved despite a heavy metastatic load and rapid disease progression at the start of vaccination. Patient GB remained on protocol for 14 months after starting vaccination with survivin-peptide pulsed DCs. However, it should be noted that both patients (RW and GB) received additional local treatment for tumor control, either irradiation of subcutaneous tumors (RW) or local chemotherapy (GB).

Survivin-specific CD8+ T-cell responses

To monitor the kinetics of cytotoxic T-cell responses, PBMCs taken before and three months after vaccination were tested for reactivity to the modified survivin96-io4 epitope by ELISPOT for IFN-γ. Before analysis, PBMCs were stimulated once in vitro to extend the sensitivity of this assay. In all four patients tested, an induction of survivin96-io4-reactive T cells was evident (Fig. 17). Analysis for reactivity to other HLA-A<*>0201-restricted survivin peptides, i.e. the unmodified survivin96-io4 and the neighboring Sur9 epitope, showed a T-cell response against these peptides in two of the patients (KN and RW) (data not shown).

The prognostic and clinical value of measurements of tumor-specific T-cell responses in peripheral blood has been repeatedly questioned, therefore we also tested for the presence of survivin96-io4/HLA-A<*>0201-reactive CD8+ T-lymphocytes among tumor-infiltrating lymphocytes in situ by peptide/MHC multimer staining. To validate the method, we first analyzed tissue samples from delayed-type hypersensitivity reactions that occurred at the vaccination site within 24 hours. This analysis confirmed previous observations that intradermal injections of peptide-pulsed DC induce a strong peptide-specific inflammatory cell infiltrate. Next, the peptide/MHC multimer staining procedure was applied to soft tissue and visceral metastases, revealing the presence of survivin96-io4/HLA-A<*>0201-reactive cells in the CD8 + infiltrate. This observation suggests that the vaccination not only induces T cells with the desired specificity, but also provides them with the required homing capacity.

REFERENCE LIST

1. Van den Eynde, B. J. and Boon, T. Tumor antigens recognized by T lymphocytes. Int.3.Chn.Lab.Res., 27: 81-86, 1997. 2. Rosenberg, S. A. Development of cancer immunotherapies based on identification of the genes encoding cancer regression antigens. J.Nati.Cancer Inst., 20;88: 1635-1644, 1996. 3. Marchand, M., van Baren, N., Weynants, P., Brichard, V., Dreno, B., Tessier, M. H., Rankm, E., Parmiani, G., Arienti, F., Humblet, Y., Bourlond, A., Vanwijck, R.,

Lienard, D., Beauduin, M., Dietrich, P. Y., Russo, V., Kerger, J., Masucci, G., Jager, E., De Greve, J., Atzpodien, J., Brasseur, F., Coulie, P. G., van der Bruggen, P., and Boon, T. Tumor regressions observed in patients with metastatic melanoma treated with an antigenic peptide encoded by gene MAGE-3 and presented by HLA- Al. Int.J.Cancer, 80: 219-230, 1999. 4. Brossart, P., Stuhler, G., Flad, T., Stevanovic, S., Rammensee, H. G., Kanz, L., and Brugger, W. Her-2/neuderived peptides are tumor-associated antigens expressed by human renal cell and colon carcinoma lines and are recognized by in vitro induced specific cytotoxic T lymphocytes. Cancer Res., 58: 732-736, 1998. 5. Brossart, P., Heinrich, K. S., Stuhler, G., Behnke, L., Reichardt, V. L., Stevanovic, S., Muhm, A., Rammensee, H. G. , Kanz, L., and Brugger, W. Identification of HLA-A2-restricted T-cell epitopes derived from the MUC1 tumor antigen for broadly applicable vaccine therapies. Blood, 93: 4309-4317, 1999. 6. Vonderheide, R. H., Hahn, W. C, Schultze, J. L, and Nadler, L. M. The telomerase catalytic subunit is a widely expressed tumor-associated antigen recognized by cytotoxic T lymphocytes. Immunity., 10: 673-679, 1999. 7. LaCasse, E. C, Baird, S., Korneluk, R. G., and MacKenzie, A. E. The inhibitors of apoptosis (IAPs) and their emerging role in cancer. Oncogene, 17: 3247-3259, 1998. 8. Altieri, D. C, Marchisio, P. C, and Marchisio, C. Survivin apoptosis: an interloper between cell death and cell proliferation in cancer. Lab Invest, 79: 1327-1333, 1999. 9. Ambrosini, G., Adida, C, Sirugo, G., and Altieri, D.C. Induction of apoptosis and inhibition of cell proliferation by survivin gene targeting. J.Biol.Chem., 273: 11177-11182, 1998. 10. Grossman, D., McNiff, J. M., Li, F., and Altieri, D. C. Expression and targeting of the apoptosis inhibitor, survivin, in human melanoma. J.Invest Dermatol., 113: 1076-1081, 1999. 11. Grossman, D., McNiff, J. M., Li, F., and Altieri, D. C. Expression of the apoptosis inhibitor, survivin, in nonmelanoma skin cancer and gene targeting in a keratinocyte cell line. Lab Invest, 79: 1121-1126, 1999. 12. Andersen, M. H., Sondergaard, I., Zeuthen, J., Elliott, T., and Haurum, J. S. An assay for peptide binding to HLA-Cw<*>0102. Tissue Antigens., 54: 185-190, 1999. 13. Andersen, M. H., Bonfill, J. E., Neisig, A., Arsequell, G., ndergaard, I., Neefjes, J., Zeuthen, J., Elliott, T. ., and Haurum, J. S. Phosphorylated Peptides Can Be Transported by TAP Molecules, Presented by Class I MHC Molecules, and Recognized by Phosphopeptide-Specific CTL. J.Immunol., 163: 3812-3818, 1999. 14. McCutcheon, M., Wehner, N., Wensky, A., Kushner, M., Doan, S., Hsiao, L., Calabresi, P., Ha, T., Tran, T. V., Tate, K. M., Winkelhake, J., and Spack, E. G. A sensitive ELISPOT assay to detect low-frequency human T lymphocytes. J.Immunol.Methods, 210: 149-166, 1997. 15. Pass, H. A., Schwarz, S. L., Wunderlich, J. R., and Rosenberg, S. A. Immunization of patients with melanoma peptide vaccines: immunologic assessment using the ELISPOT assay. Cancer J.Sci.Am., 4: 316-323, 1998. 16. Berke, Z., Andersen, M. H., Pedersen, M., Fugger, L., Zeuthen, J., and Haurum, J. S. Peptides spanning the junctional region of both the abl/bcr and the bcr/abl fusion proteins bind common HLA dass I molecules. Leukemia, 14: 419-426, 2000. 17. Falk, K., Rotzschke, O., Stevanovic, S., Jung, G., and Rammensee, H. G. Allele-specific motifs revealed by sequencing of self-peptides eluted from MHC molecules. Nature, 351: 290-296, 1991. 18. Cornelison, T. L. Human papillomavirus genotype 16 vaccines for cervical cancer prophylaxis and treatment. Curr.Opin.Oncol., 12: 466-473, 2000. 19. Lee, S. P., Chan, A. T., Cheung, S. T., Thomas, W. A., CroomCarter, D., Dawson, C. W., Tsai, C. H., Leung, S. F., Johnson, P. J., and Huang, D. P. CTL control of EBV in naso-pharyngeal carcinoma (NPC): EBV-specific CTL responses in the blood and tumors of NPC patients and the antigen-processing function of the tumor cells. J.Immunol., 165: 573-582, 2000. 20. Swana, H. S., Grossman, D., Anthony, J. N., Weiss, R. M., and Altieri, D. C. Tumor content of the antiapoptosis molecule survivin and recurrence of bladder cancer. N.Engl.J.Med., 341: 452-453, 1999. 21. Salgaller, M. L., Afshar, A., Marincola, F. M., Rivoltini, L., Kawakami, Y., and Rosenberg, S. A. Recognition of multiple epitopes in the human melanoma antigen gplOO by peripheral blood lymphocytes stimulated in vitro with synthetic peptides. Cancer Res., 55: 4972-4979, 1995. 22. Salgaller, M. L., Marincola, F. M., Cormier, J. N., and Rosenberg, S. A. Immunization against epitopes in the human melanoma antigen gplOO following patient immunization with synthetic peptides. Cancer Res., 56: 4749-4757, 1996. 23. Valmon, D., Fonteneau, J. F., Lizana, C. M., Gervois, N., Lienard, D., Rimoldi, D., Jongeneel, V., Jotereau, F. ., Cerottini, J. C, and Romero, P. Enhanced generation of specific tumor-reactive CTL in vitro by selected Melan-A/ MART-1 immunodominant peptide analogues. J. Immunol., 160: 1750-1758, 1998.

24. Pardoll, D. M. Cancer vaccines. Nat. Med., 4: 525-531, 1998.

25. Kugler, A., Stuhler, G., Walden, P., Zoller, G., Zobywalski, A., Brossart, P., Trefzer, U., Ullrich, S., Muller, C. A., Becker, V. , Gross, A. J., Hemmerlein, B., Kanz, L., Muller, G. A., and Ringert, R. H. Regression of human metastatic renal cell carcinoma after vaccination with tumor cell-dendritic cell hybrids. Nat.Med., 6: 332-336, 2000. 26. Becker, J. C, Guldberg, P., Zeuthen, J., Brocker, E. B., and thor Straten, P. Accumulation of identical T cells in melanoma and vitiligo - like leukoderma. J.Invest.Dermatol., 113: 1033-1038, 1999. 27. Rohayem, J., Diestelkoetter, P., Weigle, B., Oehmichen, A., Schmitz, M., Mehlhorn, J., Conrad, K. ., and Rieber, E. P. Antibody response to the tumor-associated inhibitor of apoptosis protein survivin in cancer patients. Cancer Res.2000.Apr.l.;60.(7.):1815.-7., 60: 1815-1817. 28. Adida, C, Haioun, C, Gaulard, P., Lepage, E., Morel, P., Briere, J., Dombret, H., Reyes, F., Diebold, J., Gisselbrecht, C, Salles , G., Altieri, D. C, and Molina, T. J. Prognostic significance of survivin expression in diffuse large B-cell lymphomas. Blood, 96: 1921-1925, 2000. 29. Islam, A., Kageyama, H., Takada, N., Kawamoto, T., Takayasu, H., Isogai, E., Ohira, M., Hashizume, K. ., Kobayashi, H., Kaneko, Y., and Nakagawara, A. High expression of Survivin, mapped to 17q25, is significantly associated with poor prognostic factors and promotes cell survival in human neuroblastoma. Oncogene, 19: 617-623, 2000. 30. Kawasaki, H., Altieri, D. C, Lu, C. D., Toyoda, M., Tenjo, T., and Tanigawa, N. Inhibition of apoptosis by suryivin predicts shorter survival rates in colorectal cancer. Cancer Res., 58: 5071-5074, 1998. 31. Schmitz, M., Diestelkoetter, P., Weigle, B., Schmachtenberg, F., Stevanovic, S., Ockert, D., Rammensee, H. G., and Rieber , E. P. Generation of survivin-specific

CD8+ T effector cells by dendritic cells pulsed with protein or selected peptides. Cancer Res., 60: 4845-4849, 2000. 32. Andersen, M. H., Pedersen, L. O., Becker, J. C, and thor Straten, P. Identification of a Cytotoxic T Lymphocyte Response to the Apoptosis Inhibitor Protein Survivin in Cancer Patients . Cancer Res., 61: 869-872, 2001. 33. Lee, K. H., Panelli, M. C, Kim, C. J., Riker, A. I., Bettinotti, M. P., Roden, M. M., Fetsch, P., Abati, A., Rosenberg, S. A., and Marincola, F. M. Functional dissociation between local and systemic immune response during anti-melanoma peptide vaccination. J.Immunol., 161: 4183-4194, 1998. 34. Rosenberg, S. A., Yang, J. C, Schwartzentruber, D. J., Hwu, P., Marincola, F. M., Topalian, S. L., Restifo, N. P., Dudley, M. E., Schwarz, S. L., Spiess, P. J., Wunderlich, J. R., Parkhurst, M. R., Kawakami, Y., Seipp, C. A., Einhorn, J. H., and White, D. E. Immunologic and therapeutic evaluation of a synthetic peptide vaccine for the treatment of patients with metastatic melanoma. . Nat.Med., 4: 321-327, 1998. 35. Altman, J. D., Moss, P. A., Goulder, P. J. R., Barouch, D. H., McHeyzer Williams, M. G., Bell, J. I., McMichael, A. J., and Davis, M. M. Phenotypic analysis. of antigen-specific T lymphocytes. Science, 274: 94-96, 1996. 36. Schrama, D., Andersen, M. H., Terheyden, P., Schroder, L., Pedersen, L. O., thor Straten, P., and Becker, J. C. Oligoclonal T-Cell Receptor Usage Of Melanocyte Differentiation Antigen-reactive T Cells in Stage IV Melanoma Patients. Cancer Res., 61: 493-496, 2001. 37. Luxembourg, A. T., Borrow, P., Teyton, L., Brunmark, A. B., Peterson, P. A., and Jackson, M. R. Biomagnetic isolation of antigen-specific CD8+ T cells usable in immunotherapy. Nat.Biotechnol., 16: 281-285, 1998. 38. Kirkin, A. F., Reichert Petersen, T., Olsen, A. C, Li, L., thor Straten, P., and Zeuthen, J. Generation of human -melanoma specificT lymphocyte clones defining novel cytolytic targets with panels of newly established melanoma cell lines. Cancer Immunol.Immunother., 41: 71-81, 1995. 39. Scheibenbogen, C, Lee, K. H., Mayer, S., Stevanovic, S., Moebius, U., Herr, W., Rammensee, H. G., and Keilholz , U. A sensitive ELISPOT assay for detection of CD8+ T lymphocytes specific for HLA dass I-binding peptide epitopes derived from influenza proteins in the blood of healthy donors and melanoma patients. Clin.Cancer Res., 3: 221-226, 1997. 40. thor Straten, P., Guldberg, P., Grønbæk, K., Zeuthen, J., and Becker, J. C. in situ T-Cell Responses against Melanoma Compnse High Numbers of Locally Expanded T-Cell Clonotypes. J.Immunol., 163: 443-447, 1999. 41. Kessler, J. H., Beekman, N. J., Bres-Vloemans, S. A., Verdijk, P., van Veelen, P. A., Kloosterman-Joosten, A. M., Vissers, D. C. , ten Bosch, G. J., Kester, M. G., Sijts, A., Wouter, D. J., Ossendorp, F., Offringa, R., and Melief, C. J. Efficient identification of novel HLA-A(<*>)0201-presented cytotoxic T lymphocyte epitopes in the widely expressed tumor antigen PRAME by proteasome-mediated digestion analysis. J.Exp.Med., 193: 73-88, 2001. 42. de Vries, T. J., Fourkour, A., Wobbes, T., Verkroost, G., Ruiter, D. J., and van Muijen, G. N. Heterogeneous expression of immunotherapy candidate proteins gplOO, MART-1, and tyrosinase in human melanoma cell lines and in human melanocytic lesions. Cancer Res., 57: 3223-3229, 1997. 43. Jager, E., Ringhoffer, M., Karbach, J., Arand, M., Oesch, F., and Knuth, A. Inverse relationship of melanocyte differentiation antigen expression in melanoma tissues and CD8+ cytotoxic-T-cell responses: evidence for immunoselection of antigen-loss variants in vivo. Int.J.Cancer, 66: 470-476, 1996. 44. Cormier, J. N., Abati, A., Fetsch, P., Hijazi, Y. M., Rosenberg, S. A., Marincola, F. M., and Topalian, S. L. Comparative analysis of the in vivo expression of tyrosinase, MART-1/Melan-A, and gplOO in metastatic melanoma lesions: implications for immunotherapy. J.Immunother., 21: 27-31, 1998. 45. Riker, A., Cormier, J., Panelli, M., Kammula, U., Wang, E., Abati, A., Fetsch, P., Lee, K. H., Steinberg, S., Rosenberg, S., and Marincola, F. Immune selection after antigen-specific immunotherapy of melanoma. Surgery, 126: 112-120, 1999. 46. Maeurer, M. J., Gollin, S. M., Martin, D., Swaney, W., Bryant, J., Castelli, C, Robbins, P., Parmiani, G., Storkus. , W. J., and Lotze, M. T. Tumor escape from immune recognition: lethal recurrent melanoma in a patient associated with downregulation of the peptide transporter protein TAP-1 and loss of expression of the immunodominant MARTl/Melan-A antigen. J.CIin.Invest, 98: 1633-1641, 1996. 47. Grossman, D., Kim, P. J., Schechner, J. S., and Altieri, D. C. Inhibition of melanoma tumor growth in vivo by survivin targeting. Proc.Natl.Acad.Sci.U.S.A, 98: 635-640, 2001. 48. Tamm, I., Wang, Y., Sausville, E., Scudiero, D. A., Vigna, N., Oltersdorf, T., and Reed, J. C. IAP-family protein survivin inhibits caspase activity and apoptosis induced by Fas (CD95), Bax, caspases, and anticancer drugs. Cancer Res., 58: 5315-5320, 1998. 49. Monzo, M., Rosell, R., Felip, E., Astudillo, J., Sanchez, J. J., Maestre, J., Martin, C, Font, A ., Barnadas, A., and Abad, A. A novel anti-apoptosis gene: Re-expression of survivin messenger RNA as a prognosis marker in nonsmall-cell lung cancers. J.CIin.Oncol., 17: 2100-2104, 1999. 50. Nakagawara, A. Molecular basis of spontaneous regression of neuroblastoma: role of neurotrophic signals and genetic abnormalities. Hum.Cell, 11: 115-124, 1998. 51. Renkvist, N., Castelli, C, Robbins, P. F., and Parmiani, G. A listing of human tumor antigens recognized by T cells. Cancer Immunol Immunother., 50: 3-15, 2001. 52. Melief, C. J., Toes, R. E., Medema, J. P., van der Burg, S. H., Ossendorp, F., and Offringa, R. Strategies for immunotherapy of cancer. Adv.Immunol., 75:235-82.: 235-282, 2000. 53. Gilboa, E. The makings of a tumor rejection antigen. Immunity., 11: 263-270, 1999. 54. Li, F., Ambrosini, G., Chu, E. Y., Plescia, J., Tognin, S., Marchisio, P. C, and Altieri, D. C. Control of apoptosis and mitotic spindle checkpoint by survivin. Nature, 396: 580-584, 1998. 55. Zaffaroni, N. and Daidone, M. G. Survivin expression and resistance to anticancer treatments: perspectives for new therapeutic interventions. Drug Resist.Updat., 5: 65-72, 2002. 56. Shinozawa, I., Inokuchi, K., Wakabayashi, I., and Dan, K. Disturbed expression of the anti-apoptosis gene, survivin, and EPR- 1 in hematological malignancies. Leuk.Res, 24: 965-970, 2000. 57. Granziero, L., Ghia, P., Circosta, P., Gottardi, D., Strola, G., Geuna, M., Montagna, L., Piccoli , P., Chilosi, M., and Cahgaris-Cappio, F. Survivin is expressed on CD40 stimulation and interfaces proliferation and apoptosis in B-cell chronic lymphocytic leukemia. Blood, 97: 2777-2783, 2001. 58. Ambrosini, G., Adida, C, and Altieri, D. C. A novel anti-apoptosis gene, survivin, expressed in cancer and lymphoma. Nat.Med., 3: 917-921, 1997. 59. Altieri, D. C. Validating survivin as a cancer therapeutic target. Nat.Rev.Cancer, 3: 46-54, 2003. 60. Ohe, R. A., Simoes-Wust, A. P., Baumann, B., Leech, S. H., Fabbro, D., Ståhei, R. A., and Zangemeister-Wittke, U A novel antisense oligonucleotide targeting survivin expression induces apoptosis and sensitizes lung cancer cells to chemotherapy. Cancer Res, 60: 2805-2809, 2000. 61. Andersen, M. H. and thor Straten, P. Survivin~a universal tumor antigen. Histol.Histopathol., 17: 669-675, 2002. 62. Andersen, M. H., Pedersen, L. O., Capeller, B., Brocker, E. B., Becker, D. C, and thor, S. P. Spontaneous cytotoxic T-cell responses against survivin -derived MHC dass I-restricted T-cell epitopes in situ as well as ex vivo in cancer patients. Cancer Res, 61: 5964-5968, 2001. 63. Curner, J. R., Kuta, E. G., Turk, E., Earhart, L B., Loomis-Price, L, Janetzki, S., Ferrari, G., Birx, D. L, and Cox, J. H. A panel of MHC dass I restncted viral peptides for use as a quality control for vaccine trial ELISPOT assays. J.Immunol.Methods, 260: 157-172, 2002. 64. Elvin, J., Potter, C, Elliott, T., Cerundolo, V., and Townsend, A. A method to quantify binding of unlabeled peptides to dass In MHC molecules and detect their allele specificity. J Immunol Methods, 158: 161-171, 1993. 65. Ruppert, J., Sidney, J., Celis, E., Kubo, R. T., Grey, H. M., and Sette, A. Prominent role of secondary anchor residues in peptide binding two HLA-A2.1 molecules. Cell, 74: 929-937, 1993. 66. thor Straten, P., Barfoed, A., Seremet, T., Saeterdal, I., Zeuthen, J., and Guldberg, P. Detection and characterization of alpha-beta -T-cell clonality by denaturing gradient gel electrophoresis (DGGE). Biotechniques, 25: 244-250, 1998. 67. Rammensee, H., Bachmann, J., Emmerich, N. P., Bachor, O. A., and Stevanovic, S. SYFPEITHI: database for MHC ligands and peptide motifs. Immunogenetics, 50: 213-219, 1999. 68. Schrama, D., Pedersen Ls, L. O., Keikavoussi, P., Andersen, M. H., Straten, P. P., Brocker, E. B., Kampgen, E., and Becker, J. C. Aggregation of antigen-specific T cells at the inoculation site of mature dendritic cells. 3.Invest Dermatol., 119: 1443-1448, 2002. 69. Mahotka, C, Wenzel, M., Springer, E., Gabbert, H. E., and Gerharz, C. D. Survivin-deltaEx3 and survivin-2B: two novel splice variants of the apoptosis inhibitor survivin with different antiapoptotic properties. Cancer Res., 59: 6097-6102, 1999. 70. Hicklin, D. J., Marincola, F. M., and Ferrone, S. HLA dass I antigen downregulation in human cancers: T-cell immunotherapy revives an old story. Mol.Med.Today, 5: 178-186, 1999. 71. Seliger, B., Cabrera, T., Garrido, F., and Ferrone, S. HLA dass I antigen abnormalities and immune escape by malignant cells. Semin.Cancer Biol., 12: 3-13, 2002. 72. Sette, A., Vitiello, A., Reherman, B., Fowler, P., Nayersina, R., Kast, W. M., Melief, C. J., Oseroff. , C, Yuan, L., Ruppert, J., and The relationship between dass I binding affinity and immunogenicity of potential cytotoxic T cell epitopes. J.Immunol., 153: 5586-5592, 1994. 73. Moudgil, K. D. and Sercarz, E. E. Can antitumor immune responses discriminate between self and nonself? Immunol.Today, 15: 353-355, 1994. 74. Parkhurst, M. R., Salgaller, M. L., Southwood, S., Robbins, P. F., Sette, A., Rosenberg, S. A., and Kawakami, Y. Improved mduction of melanoma- reactive CTL with peptides from the melanoma antigen gplOO modified at HLAA<*>0201-binding residues. J.Immunol., 157: 2539-2548, 1996. 75. Guichard, G., Zerbib, A., Le Gal, F. A., Hoebeke, J., Connan, F., Choppin, J.,

Briand, J. P., and Guillet, J. G. Melanoma peptide MART- 1 (27-35) analogues with enhanced binding capacity to the human dass I histocompatibility molecule HLA-A2 by introduction of a beta-amino acid residue: implications for recognition by tumor-infiltrating lymphocytes. J.Med.Chem., 43: 3803-3808, 2000. 76. Clay, T. M., Custer, M. C, McKee, M. D., Parkhurst, M., Robbins, P. F., Kerstann, K., Wunderlich, J., Rosenberg, S. A., and Nishimura, M. I. Changes in the fine specificity of gpl00(209-217)-reactive T cells in patients following vaccination with a peptide modified at an HLA-A2.1 anchor residue. J.Immunol., 162: 1749-1755, 1999. 77. Melief, C. J., van der Burg, S. H., Toes, R. E., Ossendorp, F., and Offnnga, R. Effective therapeutic anticancer vaccines based on precision guiding of cytolytic T lymphocytes. Immunol.Rev., 188: 177-182, 2002. 78. Jager, E., Ringhoffer, M., Altmannsberger, M., Arand, M., Karbach, J., Jager, D., Oesch, F., and Knuth, A. Immunoselection in vivo: independent loss of MHC dass I and melanocyte differentiation antigen expression in metastatic melanoma. Int.J.Cancer, 71: 142-147, 1997. 79. Thurner, B., Haendle, I., Roder, C, Dieckmann, D., Keikavoussi, P., Jonuleit, H., Bender, A., Maczek, C, Schreiner, D., von den, D. P., Brocker, E. B., Steinman, R. M., Enk, A., Kåmpgen, E., and Schuler, G. Vaccination with mage-3Al peptide-pulsed mature, monocyte-derived dendritic cells expand specific cytotoxic T cells and induce regression of some metastases in advanced stage IV melanoma. J.Exp.Med., 190: 1669-1678, 1999. 80. Yee, C, Thompson, J. A., Roche, P., Byrd, D. R., Lee, P. P., Piepkorn, M., Kenyon, K., Davis, M. M., Riddell, S. R., and Greenberg, P. D. Melanocyte destruction after antigen-specific immunotherapy of melanoma: direct evidence of t cell-mediated vitiligo. J.Exp.Med., 192: 1637-1644, 2000. 81. Simon, R. M., Steinberg, S. M., Hamilton, M., Hildesheim, A., Khleif, S., Kwak, L. W., Mackall, C. L., Schlom, J., Topalian, S. L., and Berzofsky, J. A. Clinical trial designs for the early clinical development of therapeutic cancer vaccines. J. CIin. Oncol., 19: 1848-1854, 2001.

Claims (33)

1. MHC class I-restricted epitope peptide derived from survivin, said epitope having the sequence RISTFKNWPK as specified by SEQ ID NO: 58 and having at least one of the following characteristics: (i) is capable of binding to the class I HLA-Al molecule with an affinity as measured by the amount of the peptide capable of effecting half maximal recovery of the class I HLA molecule (C50 value), which is a maximum of 50 µM as determined by an HLA stabilization assay; (ii) is capable of eliciting IFN-γ-producing cells in a peripheral blood lymphocyte (PBL) population from a cancer patient at a frequency of at least 1 per 10 <4> PBLs as determined by an ELISPOT assay , and or (iii) is capable of in situ detection in a tumor tissue of CTVs reactive with the epitope peptide. 2. Peptide according to claim 1, which has a C50 value of at most 30 uM. 3. Peptide according to claim 1 or 2, which has a C50 value of at most 20 uM. 4. Peptide according to any one of the preceding claims, comprising a maximum of 20 amino acid residues. 5. Peptide according to claim 4, which comprises a maximum of 10 amino acid residues. 6. Peptide according to any one of the preceding claims, which is capable of triggering IFN-γ-producing cells in a PBL population from a cancer patient at a frequency of at least 10 per 10 <4> PBLs. 7. Peptide according to any one of the preceding claims, which is capable of triggering IFN-γ-producing cells in a PBL population from a patient having a cancer in which survivin is expressed. 8. Peptide according to claim 7, where the cancer is selected from the group consisting of a hematopoietic malignancy which includes chronic lymphocytic leukemia and chronic myeloid leukemia, melanoma, breast cancer, cervix cancer, ovarian cancer, lung cancer, colon cancer, pancreatic cancer and prostate cancer. 9. Peptide according to any one of the preceding claims, which is capable of triggering IFN-γ-producing cells in a PBL population from a patient having a cancer disease, wherein said IFN-γ-producing cells have a cytotoxic effect against survivin-expressing cells of a cancer cell line, including a cell line selected from the group consisting of the breast cancer cell line MCF-7 and the melanoma cell line FM3. 10. Pharmaceutical composition comprising the peptide according to any one of claims 1 to 9. 11. Pharmaceutical mixture according to claim 10, wherein said mixture comprises a combination of two or more MHC class I-restricted epitope peptides derived from survivin, each interacting specifically with another HLA molecule and having at least one of the following characteristics: (i) is capable of binding to the class I HLA molecule to which it is restricted with an affinity as measured by the amount of the peptide capable of effecting half maximal recovery of the class I HLA molecule (C50 value), which is at most 50 µM as determined by an HLA stabilization assay; (ii) is capable of triggering IFN-γ-producing cells in a PBL population of a cancer patient at a frequency of at least 1 per IO <4> PBLs as determined by an ELISPOT assay, and/or (l ii) is capable of in situ detection in a tumor tissue of CTLs which are reactive with the epitope peptide. 12. Pharmaceutical mixture according to claim 11, wherein each epitope peptide interacts specifically with an MHC class I HLA-A molecule, an MHC class I HLA-B molecule or an MHC class I HLA-C molecule. 13. Pharmaceutical mixture according to claim 12, wherein said MHC class I HLA-A molecule includes HLA-A1, HLA-A2, HLA-A3, HLA-A9, HLA-A10, HLA-All, HLA-AW 19, HLA- A23(9), HLA-A24(9), HLA-A25(10), HLA-A26(10)„ HLA-A28, HLA-A29(wl9), HLA-A30(wl9), HLA-A31(wl9) , HLA-A32(wl9), HLA-Aw33(wl9), HLA-Aw34(10), HLA-Aw36, HLA-Aw43, HLA-Aw66(10), HLA-Aw68(28), HLA-A69(28) . 14. Pharmaceutical composition according to any one of claims 12 to 13, wherein said MHC Class I HLA-B molecule includes any of the following: HLA-B5, HLA-B7, HLA-B8, HLA-B12, HLA-B13, HLA-B14, HLA-B15, HLA-B16, HLA-B17, HLA-B18, HLA-B21, HLA-Bw22, HLA-B27, HLA-B35, HLA-B37, HLA-B38, HLA-B39, HLA- B40, HLA-Bw41, HLA-Bw42, HLA-B44, HLA-B45, HLA-Bw46 and HLA-Bw47. 15. Pharmaceutical composition according to any one of claims 12 to 14, wherein said MHC class I HLA-C molecule includes any of the following: HLA-Cwl, HLA-Cw2, HLA-Cw3, HLA-Cw4, HLA-Cw5, HLA-Cw6, HLA-Cw7 and HLA-Cwl6. 16. A pharmaceutical composition according to any one of claims 11 to 15 comprising an epitope peptide which is a native sequence of survivin from a mammalian species. 17. Pharmaceutical composition according to any one of claims 10 to 16, comprising an epitope peptide selected from the group consisting of SEQ ID NO: 1, SEQ ID NO: 2, SEQ ID NO: 3, SEQ ID NO: 6, SEQ ID NO: 7, SEQ ID NO: 27, SEQ ID NO: 45 and SEQ ID NO: 66. 18. Pharmaceutical composition according to any one of claims 11 or 12, comprising an epitope peptide selected from the group consisting of: SEQ ID NO: 4, SEQ ID NO: 5 SEQ ID NO: 8, SEQ ID NO: 9, SEQ ID NO 36; SEQ ID NO: 38, SEQ ID NO: 39,. 19. Pharmaceutical composition according to any one of claims 11 to 18, wherein said epitope peptide is capable of triggering IFN-γ-producing cells in a PBL population from a cancer patient at a frequency of at least 10 per 10 <4 > PBLs. 20. A pharmaceutical composition according to any one of claims 11 to 19, wherein said epitope peptide is capable of triggering IFN-γ-producing cells in a PBL population from a patient having a cancer disease in which survivin is expressed. 21. Pharmaceutical mixture according to claim 20, where the cancer is selected from the group consisting of a hematopoietic malignancy which includes chronic lymphocytic leukemia and chronic myeloid leukemia, melanoma, breast cancer, cervical cancer, ovarian cancer, lung cancer, colon cancer, pancreatic cancer and prostate cancer. 22. A pharmaceutical composition according to any one of claims 10 to 21, comprising an epitope peptide which is post-translationally modified. 23. A pharmaceutical composition according to any one of claims 10 to 22, comprising a phosphorylated peptide. 24. Pharmaceutical composition according to any one of claims 10 to 23, comprising an adjuvant. 25. A pharmaceutical composition according to any one of claims 10 to 24, which is a composition for ex vivo or in situ diagnosis of the presence of survivin-reactive T cells among PBLs or in tumor tissue. 26. Multi-epitope vaccine comprising an MHC class I-restricted epitope peptide derived from survivin, wherein said epitope has the sequence RISTFKNWPK as specified by SEQ ID NO: 58 and which has at least one of the following characteristics: (i) is capable of binding the class I HLA-Al molecule to which it is restricted with an affinity as measured by the amount of the peptide capable of effecting half-maximal recovery of the class I HLA molecule (C50 value), as a maximum is 50 µM as determined by an HLA stabilization assay; (ii) is capable of eliciting IFN-γ-producing cells in a PBL population from a cancer patient at a frequency of at least 1 per IO <4> PBLs as determined by an ELISPOT assay, and/or (iii) is capable of in situ detection in a tumor tissue of CTLs reactive with the epitope peptide. 27. Multi-epitope vaccine according to claim 26, which comprises a combination of survivin-derived peptide epitopes depending on the tissue type of the patient in question. 28. Multi-epitope vaccine according to claim 26 or 27, wherein the vaccine is able to trigger an immune response against a cancer in which survivin is expressed. 29. Multi-epitope vaccine according to claim 28, wherein the cancer is selected from the group consisting of a hematopoietic malignancy comprising chronic lymphocytic leukemia and chronic myeloid leukemia, melanoma, breast cancer, cervical cancer, ovarian cancer, lung cancer, colon cancer, pancreatic cancer and prostate cancer. 30. Multi-epitope vaccine according to any one of claims 26 to 29, wherein the vaccine triggers production in the vaccinated individual of effector T cells which have a cytotoxic effect against the cancer cells. 31. Use of a peptide as described in any one of claims 1 to 9 for the production of a medicament for the treatment of cancer. 32. Use according to claim 31, where the cancer is selected from the group consisting of a hematopoietic malignancy which includes chronic lymphocytic leukemia and chronic myeloid leukemia, melanoma, breast cancer, cervix cancer, ovarian cancer, lung cancer, colon cancer, pancreatic cancer and prostate cancer. 33. Use of a peptide as described in any one of claims 1 to 9 for the manufacture of a medicament for the treatment of cancer in combination with conventional cancer treatment such as radiotherapy or chemotherapy.