WO2011149909A2

WO2011149909A2 - Class i mhc phosphopeptides for cancer immunotherapy and diagnosis

Info

Publication number: WO2011149909A2
Application number: PCT/US2011/037699
Authority: WO
Inventors: Donald F. Hunt; Jeffrey Shabanowitz; Jennifer Cottine; Ann M. English; Andrew Norris; Victor H. Engelhard; Mark Cobbold; Kara L. Cummings; Angela Zarling; Rebecca C. Obeng
Original assignee: Hunt Donald F; Jeffrey Shabanowitz; Jennifer Cottine; English Ann M; Andrew Norris; Engelhard Victor H; Mark Cobbold; Cummings Kara L; Angela Zarling; Obeng Rebecca C
Priority date: 2010-05-24
Filing date: 2011-05-24
Publication date: 2011-12-01
Also published as: AU2017201229A1; EP2576614A4; AU2011258513A1; WO2011149909A3; US20130259883A1; CA2800535A1; EP2576614A2; AU2019204422A1; US20180066017A1; AU2021250866A1; US20220041655A1

Abstract

A set of phosphorylated peptides are presented by HLA A*0101, A*0201, A*0301, B*4402, B*2705, B*1402, and B*0702 on the surface of melanoma cells. They have the potential to (a) stimulate an immune response to the cancer, (b) to function as immunotherapeutics in adoptive T-cell therapy or as a vaccine, (c) to facilitate antibody recognition of the tumor boundaries in surgical pathology samples, and (d) act as biomarkers for early detection of the disease. Phosphorylated peptides are also presented for other cancers.

Description

CLASS I MHC PHOSPHOPEPTIDES FOR CANCER

IMMUNOTHERAPY AND

[01] This invention was made with government support under ROl AI20963 and AI33993 awarded by the National Institutes of Health. The government has certain rights in the invention,

TECHNICAL FIELD OF THE INVENTION

[02] This invention is related to the area of cancer diagnostics, and therapeutics. In particular, it relates to imniunoiogical reactions mediated through MHC class I molecules.

BACKGROUND OF THE INVENTION

The mammalian immune system has evolved a variety of mechanisms to protect the host from cancerous ceils. An important component of this response is mediated by cells referred to as T cells. Cytotoxic T lymphocytes (CTL) are specialized T cells that primarily function by recognizing and killing cancerous cells or infected cells, but they can also function by secreting soluble molecules referred to as cytokines that can mediate a variety of effects on the immune system. T helper cells primarily function by recognizing antigen on specialized antigen presenting ceils, and in turn secreting cytokines that activate B cells, T cells, and macrophages, A variety of evidence suggests that immunotherapy designed to stimulate a tumor-specific CTL response would be effective in controlling cancer. For example, it has been shown that human CTL recognize sarcomas (Slovin et al., 1986, J Immunol 137, 3042- 3048), renal cell carcinomas (Schendel et al., 1993, J Immunol 151, 4209-4220), colorectal carcinomas (Jacob et al, 1997, Int J Cancer 71 , 325-332), ovarian carcinomas (Peoples et al., 1993, Surgery 1 14, 227-234), pancreatic carcinomas (Peiper et al., 1997, Eur J Immunol 27, 1 115-1 123), squamous tumors of the head and neck (Yasurnura et al, 1993, Cancer Res 53, 1461-1468), and squamous carcinomas of the lung (Slingluff et al., 1994, Cancer Res 54, 2731-2737; Yoshino et al., 1994, Cancer Res 54, 3387-3390), The largest number of reports of human tumor-reactive CTLs, however, has concerned melanomas (Boon et al., 1994, Annu Rev Immunol 12, 337-365). The ability of tumor-specific CTL to mediate tumor regression, in both human (Parmiani et aL 2002, J Natl Cancer Inst 94, 805-818; Weber, 2002, Cancer Invest 20, 208-221 ) and animal models, suggests that methods directed at increasing CTL activity would likely have a beneficial effect with respect to tumor treatment.

Melanoma, or skin cancer, is a disease that is diagnosed in approximately 54,200 persons per year. Conventional therapy for the disease includes surgery, radiation therapy, and chemotherapy . In spite of these approaches to treatment, approximately 7,600 individuals die in the United States every year due to melanoma. Overall, the 5- year survival rate for the disease is 88%. The survival rate drops, however, in more advanced stages of the disease with only about 50% of Stage III patients, and 20-30% of Stage IV patients surviving past five years. In patients where the melanoma has metastasized to distant sites, the 5-year survival dips to only 12%. Clearly, there is a population of melanoma patients that is in need of better treatment options. More recently, in an attempt to decrease the number of deaths attributed to melanoma, immunotherapy has been added to the arsenal of treatments used against the disease.

In order for CTL to kill or secrete cytokines in response to a cancer cell, the CTL must first recognize the cancer cell (Townsend and Bodmer, 1989). This process involves the interaction of the T cell receptor, located on the surface of the CTL,, with what is generically referred to as an MHC-peptide complex which is located on the surface of the cancerous cell. MHC (major histocompatibiiity-compIex)-encoded molecules have been subdivided into two types, and are referred to as class I and class II MHC-encoded molecules. In the human immune system, MHC molecules are referred to as human leukocyte antigens (HLA). Within the MHC complex, located on chromosome six, are three different loci that encode for class I MHC molecules. MHC molecules encoded at these loci are referred to as HLA-A, HLA-B, and HLA-C. The genes that can be encoded at each of these loci are extremely polymorphic, and thus, different individuals within the population express different class I MHC molecules on the surface of their cells. HLA-A 1 , HLA-A2, HLA-A3, HLA-B7, HLA- B14, HLA-B27, and HLA-B44 are examples of different class I MHC molecules that can be expressed from these loci. The peptides which associate with the MHC molecules cars either he derived from proteins made within the ceil, in which case they typically associate with class 1 MHC molecules (Rock and Goldberg, 1999, Annu Rev Immunol 17, 739-779); or they can be derived from proteins which are acquired from outside of the cell, in which case they typically associate with class II MHC molecules (Watts, 1997, Annu Rev Immunol 15, 821 -850). The peptides that evoke a cancer-specific CTL response most typically associate with class I MHC molecules. The peptides themselves are typically nine amino acids in length, but can vary from a minimum length of eight amino acids to a maximum of fourteen amino acids in length. Tumor antigens may also bind to class II MHC molecules on antigen presenting cells and provoke a T helper cell response. The peptides that bind to class 11 MHC molecules are generally twelve to nineteen amino acids in length, but can be as short as ten amino acids and as long as thirty amino acids.

The process by which intact proteins are degraded into peptides is referred to as antigen processing. Two major pathways of antigen processing occur within cells (Rock and Goldberg, 1999, Annu Rev Immunol 17, 739-779). One pathway, which is largely restricted to professional antigen presenting cells such as dendritic cells, macrophages, and B cells, degrades proteins that are typically phagocytosed or endocytosed into the cell. Peptides derived from this pathway can be presented on either class I or to class II MHC molecules. A second pathway of antigen processing is present in essentially all cells of the body. This second pathway primarily degrades proteins that are made within the cells, and the peptides derived from this pathway primarily bind to class I MHC molecules. Antigen processing by this latter pathway involves polypeptide synthesis and proteolysis in the cytoplasm, followed by transport of peptides to the plasma membrane for presentation. These peptides, initially being transported into the endoplasmic reticulum of the cell, become associated with newly synthesized class I MHC molecules and the resulting complexes are then transported to the cell surface. Peptides derived from membrane and secreted proteins have also been identified. In some cases these peptides correspond to the signal sequence of the proteins which is cleaved from the protein by the signal peptidase. In other cases, it is thought that some fraction of the membrane and secreted proteins are transported from the endoplasmic reticulum into the cytoplasm where processing subsequently occurs. Once bound to the class I MHC molecule, the peptides are recognized by antigen-specific receptors on CTL. Several methods have been developed to identify the peptides recognized by CTL, each method of which relies on the ability of a CTL to recognize and kill only those cells expressing the appropriate class I MHC molecule with the peptide bound to it. Mere expression of the class I MHC molecule is insufficient to trigger the CTL to kill the target cell if the antigenic peptide is not bound to the class I MHC molecule. Such peptides can be derived from a non-self source, such as a pathogen (for example, following the infection of a cell by a bacterium or a virus) or from a self-derived protein within a cell, such as a cancerous cell. The tumor antigens from which the peptides are derived can broadly be categorized as differentiation antigens, cancer/testis antigens, mutated gene products, widely expressed proteins, viral antigens and most recently, phosphopeptides derived from dysregulated signal transduction pathways. (Zarling et al., PNAS 103, 12889-14894, 2006). Immunization with melanoma-derived, class I or class II MITC-encoded molecule associated peptides, or with a precursor polypeptide or protein that contains the peptide, or with a gene that encodes a polypeptide or protein containing the peptide, are forms of immunotherapy that can be employed in the treatment of melanoma. Identification of the immunogens is a necessary first step in the formulation of the appropriate immunotherapeutic agent or agents. Although a large number of tumor- associated peptide antigens recognized by tumor reactive CTL have been identified, there are few examples of antigens that are derived from proteins that are selectively expressed on a broad array of tumors, as well as associated with cellular proliferation and/or transformation. Attractive candidates for this type of antigen are peptides derived from proteins that are differentially phosphorylated on serine (Ser), threonine (Thr), and tyrosine (Tyr) (Zarling et al., 2000, J Exp Med 192 1755-1762). Due to the increased and dysregulated phosphorylation of cellular proteins in transformed cells as compared to normal cells, tumors are likely to present a unique subset of phosphorylated peptides on the cell surface that are available for recognition by cytotoxic T-lymphocytes (CTL). Presently, there is no way to predict which protein phosphorylation sites in a cell will be unique to tumors, survive the antigen processing pathway, and be presented to the immune system in the context of 8-14 residue phosphopeptides bound to class I MHC molecules. Thirty-six phosphopeptides were disclosed as presented in association with HLA A*0201 on cancer cells. Zarling, et al., Proc. Natl Acad, Sciences, 103, 14889- 14894, 2006, Table 1. Parent proteins for four of these peptides (β-catenin, insulin receptor substrate-2 (IRS-2), tensin-3 and Jun-C/D) are known to be associated with cytoplasmic signaling pathways and cellular transformation. While both normal and cancer cells lines express the parent proteins, only the three cancer lines express phospkorylated class I peptide sequences within IRS-2 and β-catenin, respectively.

Mice expressing a transgenic recombinant human A*0201 MHC molecule were immunized with a synthetic class 1 phosphopeptides from IRS-2 and β-catenin that were pulsed onto activated bone-marrow derived dendritic cells. Cytotoxic T-cells were generated that recognized all three cancer cell lines but not the control JY cells. Class I phosphopeptides from IRS-2 and β-catenin are highly immunogenic and are likely candidates for immunotherapy directed toward melanoma and ovarian cancer.

Adoptive T-cell therapy of melanoma is described in two recent publications. Dudley et al, J. Clin. Oncology 2008, 26: 5233-5239 and Rosenberg, Curr. Opinion in Immun. 2009, 21 : 233-240. For adoptive T-cell therapy, fate stage metastatic melanoma patients are treated as if they were undergoing an organ transplant operation. Tumor is resected and cytotoxic T-cells that have infiltrated the tumor are harvested and exposed to a particular class I peptide antigen (MART -1 ). Those that recognize this antigen are then allowed to expand until the total number of M ART- 1 specific cells reach 100 billion. The patient receives whole body irradiation and chemotherapy to wipe out 98% of his/her immune system. The MART specific T- celis are then given back to the patient and circulate throughout the body looking for tumor. In the most recent clinical trial, tumors in 72% of the patients showed objective responses with this therapy at all sites of metastasis including lymph nodes, bone, Sung, liver, and brain. Twenty-eight percent of the patients had complete remission of the disease.

There is a need in the art for additional class I phosphopeptide antigens to permit adoptive T-cell therapy to be extended to cancer patients that may not express the HLA-A*0201 allele, as well as new phosphopeptides for patients that express the HLA*0201 allele. There is a need in the art to treat a variety of other cancers by the same approach.

SUMMARY OF THE INVENTION

112] One aspect of the invention is an isolated and purified phosphopeptide that consists of between 8 and 50 contiguous amino acid residues derived from a native human protein. The phosphopeptide comprises a sequence selected from SEQ ID NO: 1- 13 1 in which at least one serine, threonine, or tyrosine residue in the selected sequence is phosphor)'! ated with a hydrolyzable or non-hydrolyzable phosphate group. Contiguous amino acids adjacent to the selected sequence in the phosphopeptide are selected from the adjacent residues in the native human protein. When the sequence is selected from SEQ ID NO: 1266-1297, the phosphopeptide is phosphorylated with a non-hydroiyzable phosphate group.

[13] Another aspect of the invention is a method of immunizing a mammal to diminish the risk of, the growth of, or the invasiveness of a melanoma. A composition is administered to the mammal that activates CD8^1' T cells. The composition comprises a phosphopeptide that consists of between 8 and 50 contiguous amino acid residues derived from a native human protein. The phosphopeptide comprises a sequence selected from SEQ ID NO: 1-1391 in which at least one serine, threonine, or tyrosine residue in the selected sequence is phosphorylated with a hydrolyzable or non- hydro lyzable phosphate group. Contiguous amino acids adjacent to the selected sequence in the phosphopeptide are selected from the adjacent residues in the native human protein. When the sequence is selected from SEQ ID NO: 1266-1297, the phosphopeptide is phosphorylated with a non-hydrolyzable phosphate group.

[14] Another aspect of the invention is a method that can be used for monitoring, diagnosis, or prognosis. A sample isolated from a patient is contacted with an antibody that specifically binds to a phosphopeptide. The phosphopeptide consists of between 8 and 50 contiguous amino acid residues derived from a native human protein. The phosphopeptide comprises a sequence selected from SEQ ID NO: 1- 1391 in which at least one serine, threonine, or tyrosine residue in the selected sequence is phosphor lated with a hydrolyzable or non-hydro lyzabie phosphate group. Contiguous amino acids adjacent to the selected sequence in the phosphopeptide are selected from the adjacent residues in the native human protein. The antibody does not bind to a peptide consisting of the same amino acid sequence but devoid of phosphorylation. Antibody bound to the sample is measured or detected.

Still another aspect of the invention is a molecule that comprises an antigen-binding region of an antibody. The molecule specifically binds to a phosphopeptide and does not bind to a peptide consisting of the same amino acid sequence but devoid of phosphorylation. The phosphopeptide consists of between 8 and 50 contiguous amino acid residues derived from a native human protein. The phosphopeptide comprises a sequence selected from SEQ ID NO: 1-1391 in which at least one serine, threonine, or tyrosine residue in the selected sequence is phosphorylated with a hydrolyzable or non-hydrolyzable phosphate group. Contiguous amino acids adjacent to the selected sequence in the phosphopeptide are selected from the adjacent residues in the native human protein.

Still another aspect of the invention is a kit for measuring a phosphoproiein consisting of between 8 and 50 contiguous amino acids. The phosphoproiein comprises a sequence selected from SEQ ID NO: 1 -1391 that includes a phosphorylated serine, threonine, or tyrosine residue. The kit comprises a molecule comprising an antigen- binding region of an antibody, wherein the molecule specifically binds to the phosphoproiein and does not bind to a protein consisting of the same amino acid sequence but devoid of phosphorylation.

Yet another aspect of the invention is a method, useful for producing an immunotherapeutic agent or tool. Dendritic cells are contacted in vitro with an isolated phosphopeptide consisting of between 8 and 50 contiguous amino acids. The phosphopeptide comprises a sequence selected from SEQ ID NO: 1 -1391 which includes at least one serine, threonine, or tyrosine residue that is phosphorylated. The dendritic cells thereby become phosphopeptide-loaded. When the sequence is selected from SEQ ID NO: 1266-1297, the phosphopeptide is phosphorylated with a non-hydrolyzable phosphate group. The dendritic cells made by the method provides an in vitro compositions of dendritic cells, useful as an immunotherapeutic agent,

[18] A further aspect of the invention is a synthetic phosphopeptide comprising from 10- 50 amino acid residues, comprising the sequences, RVAsPTSGVK (SEQ ID NO: 53) or RVAsPTSGVKR (SEQ ID NO: 54), wherein the serine residue at position 4 is phosphorylated with a. hydrolyzable or nonhydrolyzable phosphate group, and wherein adjacent amino acid residues to the sequence are adjacent sequences in the human insulin substrate- 2 (IRS-2) protein. The phosphopeptide is useful for loading dendritic cells so that they present phosphopeptide on HLA-A*0301 molecules.

[19] These and other aspects and embodiments which will be apparent to those of skill in the art upon reading the specification provide the art with immunological tools and agents useful for diagnosing, prognosing, monitoring, and treating human cancers.

BRIEF DESCRIPTION OF THE DRAWINGS

[20] Figs. 1A and IB are a graphic illustration of the recognition of naturally processed and presented phosphorylated peptides on cancer cells by the phosphopeptide-specific CTL. Phosphopeptide-specific CTL were incubated with the following cancer cell lines or EBV -trans formed B iymphoblastoid cell lines (BLCL): COV413.AAD.A4 ovarian carcinoma, DM331.AAD.A4 and SLM2.AAD.A1 melanomas, MCF7.AAD.A2 and MDAMB231.AAD breast carcinomas, and JY EBV-BLCL. Supernatants were harvested and evaluated for the presence of murine IFNy (produced by murine CTL lines). As a positive control, cancer cells were pulsed with the specific phosphopeptide to show that they are capable of presenting exogenous ly added peptide. In Fig. ! A, two phosphopeptide-specific CTL cell lines, 6850 and 6960 that are specific for the phosphopeptide GLLGpSPVRA (SEQ ID NO: 1268), recognize the phosphopeptide on all the cancer cell lines, but not the control cell line. In Fig. IB, two phosphopeptide-specific CTL cell lines, 5183 and 63 that are specific for the phosphopeptide RVApSPTSGV (SEQ ID NO: 1289), recognize the phosphopeptide on all the cancer cell lines, but not the control cell line. The designation "pS" denotes a phosphoserine residue. The ordinate indicates murine IFNy in pg/ml. The abscissa indicates each cell line. [21] Figs. 2A-2E present Tables 2A-2E. Fig. 2 A shows melanoma HLA A*0301 phosphopeptides, A*0101 phosphopeptides, B*4402 phosphopeptides, B*27Q5 phosphopeptides, and B* 1402 phosphopeptides. Fig. 2B shows melanoma and/or leukemia HLA B*0702 phosphopeptides. Fig. 2C shows melanoma HLA A*0301 phosphopeptides, A*0101 phosphopeptides, B*4402 phosphopeptides, B*2705 phosphopeptides, and B* 1402 phosphopeptides and their sequence variants. Fig. 2D shows melanoma and/or leukemia HLA B*0702 phosphopeptides and their sequence variants. Fig. 2E shows melanoma HLA-A*0201 phosphopeptides. Fig. 2F shows melanoma HLA-A*02()1 phosphopeptides and their sequence variants.

DETAILED DESCRIPTION OF THE INVENTION

[22] We have identified MHC class 1 phosphopeptides for use in diagnostics, immunotherapeutics, and adoptive T-cell therapy of melanoma patients. We provide over 200 class I MHC peptides presented on the surface of cancer cells in association with the HLA molecules A*010i (SEQ ID NO: 70-97), A*0301 (SEQ ID NO: 1-69), and B*4402 (SEQ ID NO: 98-110), B*2705 (SEQ ID NO: 11 1-162), B* 1402 (SEQ ID NO: 163-164), and B*0702 (SEQ ID NO: 165-246). Variants and mimeiics of these peptides and of additional class I MHC phosphopeptides are also provided.

[23] Although individuals in the human population display hundreds of different HLA alleles, some are more prevalent than others. For example, 88% of melanoma patients cany at least one of the six HLA alleles: HLA-A*0201 (29%), HLA-A*0101 (15%), HLA-A*0301 (14%), HLA-B*4402 (15%), HLA-B*07Q2 (12%), and HLA-B*-2705 (3%). One of our aims is to provide multiple phosphopeptides presented by each of the six most prevalent alleles and to use them as a cocktail, to optimize coverage of the human population and to minimize the possibility that the tumor will be able to escape immune surveillance by down-regulating expression of any one class I phosphopeptide.

[24] Phosphopeptides of the invention are not the entire proteins from which they are derived. They are from 8 to 50 contiguous amino acid residues of the native human protein. They contain at least one of the MHC class 1 binding peptides listed in SEQ ID NO: 1 -1391. Moreover, at least one of the serine, threonine, or tyrosine residues within the recited sequence is phosphoryiated. The phosphorylation may be with a natural phosphorylation (-CH₂-O-PO₃H) or with an enzyme non-degradable, modified phosphorylation, such as (-CH₂-CF₂-PO₃H or -CH₂- CH₂-P0₃H). In certain specified positions, a native amino acid residue in a native human protein may be altered to enhance the binding to the MHC class I molecule. These occur in "anchor" positions of the phosphopeptides, often in positions 1, 2, 3, 9, or 10. Valine, alanine, lysine, leucine tyrosine, arginine, phenylalanine, proline, glutamic acid, threonine, serine, aspartic acid, tryptophan, and methionine may also be used as improved anchoring residues. Anchor residues for different HLA molecules are shown in Table 1. Some phosphopeptides may contain more than one of the peptides listed in SEQ ID NO: 1- 1391, for example, if they are overlapping, adjacent, or nearby within the native protein from which they are derived. Phosphopeptides can also be mixed together to form a cocktail. The phosphopeptides may be in an admixture, or they may be linked together in a concatamer as a single molecule. Linkers between individual phosphopeptides may be used; these may, for example, be formed by any 10 to 20 amino acid residues. The linkers may be random sequences, or they may be optimized for degradation by dendritic cells.

1. Optimal anchor residues for HLA molecules

HLA A*0201 Residue 2 - i . M

Residue 9 or last residue = V

HLA A*0301 Residue 2 = L, M,

Residue 9 or last residue^■=

HLA A*0i0i Residue 2 = T, S

Residue 3 = D, E

Residue 9 or last residue = Y

HLA B*2705 Residue 1 - R

Residue 2 - R

Residue 9 or last residue L, F, K, R, M

HLA B*0702 Residue 2 - P

Residue 9 or last residue = L, M, V, F

HLA B*4402 Residue 2 - E

Residue 9 or last residue = F, Y, W The chemical sirucmre of a phosphopeptide mimetic appropriate for use in the present invention may closely approximate the natural phosphorylated residue which is mimicked, and also be chemically stable (e.g., resistant to dephosphorylation by phosphatase enzymes). This can be achieved with a synthetic molecule in which the phosphorous atom is finked to the amino acid residue, not through oxygen, but through carbon. In one embodiment, a CF2 group links the amino acid to the phosphorous atom. Mimetics of several amino acids which are phosphorylated in nature can be generated by this approach. Mimetics of phosphoserine, phosphothreonine, and phosphotyrosine can be generated by placing a CF2 linkage from the appropriate carbon to the phosphate moiety. The mimetic molecule L-2- amino-4 (diethyiphosphono)-4,4-difluorobuianoic acid (F2Pab) may substitute for phosphoserine (Otaka et al., Tetrahedron Letters 36: 927-930 ( 1995)). L-2-amino-4- phosphono-4,4difluoro-3-metbylbutanoic acid (F2Pmb) may substitute for phosphothreonine. L-2-amino-4-phosphono (difiuoromethyl) phenylalanine (F2Pmp) may substitute for phosphotyrosine (Akamatsu et al., Bioorg & Med Chem. 5: 157- 163 ( 1997); Smyth et al.. Tetrahedron Lett. Tetrahedron Lett. 33,4137-4140 (1992)). Alternatively, the oxygen bridge of the natural amino acid may be replaced with a methylene group.

Compositions comprising the phosphopeptide are typically substantially free of other human proteins or peptides. They can be made synthetically or by purification from a biological source. They can be made recombinant!}'. Desirably they are at least 90 %, at least 95 %, at least 99 % pure. For administration to a human body, they do not contain other components that might be harmful to a human recipient. The compositions are typically devoid of cells, both human and recombinant producing cells. However, as noted below, in some cases, it may be desirable to load dendritic cells with a phosphopeptide and use those loaded dendritic cells as either an immunotherapy agent themselves, or as a reagent to stimulate a patient's T cells ex vivo. The stimulated T cells can be used as an immunotherapy agent. In some cases, it may be desirable to form a complex between a phosphopeptide and an HLA molecule of the appropriate type. Such complexes may be formed in vitro or in vivo. Such complexes are typically tetrameric with respect to an HLA-phosphopeptide complex. Under certain circumstances it may be desirable to add additional proteins Of peptides, for example, to make a cocktail Slaving the ability to stimulate an immune response in a number of different HLA type hosts. Alternatively, additional proteins or peptide cars provide an interacting function within a single host, such as an adjuvant function or a stabilizing function. As an example, other tumor antigens can be used in admixture with the phosphopeptides, such that multiple different immune responses are induced in a single patient.

Administration of phosphopeptides to a mammalian recipient may be accomplished using long phosphopeptides, e.g., longer than 15 residues, or using phosphopeptide- loaded dendritic cells. See Melief, J. Med. Sciences 2009; 2:43-45. The immediate goal is to induce activation of CDS ' T cells. Additional components which can be administered to the same patient, either at the same time or close in time (e.g., within 21 days of each other) include TLR-ligand oligonucleotide CpG and related phosphopeptides that have overlapping sequences of at least 6 amino acid residues. To ensure efficacy, mammalian recipients should express the appropriate human HLA molecules to bind to the phosphopeptides. Transgenic mammals can be used as recipients, for example, if they express appropriate human HLA molecules. If a mammal's own immune system recognizes a similar phosphopeptide then it can be used as model system directly, without introducing a transgene. Useful models and recipients may be at increased risk of developing metastatic cancer, such as metastatic melanoma. Other useful models and recipients may be predisposed, e.g., genetically or environmentally, to develop melanoma or other cancer.

Phosphopeptide-loaded dendritic cells can also be used to transfuse a cancer patient or a patient at risk of cancer. The composition of dendritic cells can be provided with a single phosphopeptide loaded in the cells. Thus the dendritic cells are homogenous with respect to the loaded phosphopeptide. The homogeneity may not be perfectly- achievable. The desired phosphopeptide may be form at least 20 %, at least 50 %, at least 70 %, or at least 90 % of the phosphopeptides loaded in the compositions. Additional components may be added to the composition to be administered, such as immune adjuvants, stabilizers, and the like. The particular phosphopeptides were identified on the surfaces of particular cancer cells, but they may be found on other types of cancer cells as well, including but not limited to melanoma, ovarian cancer, breast cancer, colorectal cancer, squamous carcinoma of the lung, sarcoma, renal cell carcinoma, pancreatic carcinomas, squamous tumors of the head and neck, leukemia, brain cancer, liver cancer, prostate cancer, ovarian cancer, and cervical cancer.

Antibodies and antibody-like molecules containing an antigen-binding region are useful, inter alia, for analyzing tissue to determine the pathological nature of tumor margins. Such tissue may be obtained from a biopsy, for example. Other samples which may be tested include blood, serum, plasma, and lymph. Antibodies to peptides may be generated using methods that are well known in the art. For the production of antibodies, various host animals, including rabbits, mice, rats, goats and other mammals, can be immunized by injection with a peptide. They may be conjugated to carrier proteins such as KLH or tetanus toxoid. Various adjuvants may be used to increase the immunological response, depending on the host species, and including but not limited to Freund's (complete and incomplete), mineral gels such as aluminum hydroxide, surface active substances such as lysolecitbin, pluronic polyols, poiyanions, peptides, oil emulsions, keyhole limpet hemocyanins, diniirophenol, and potentially useful human adjuvants such as BCG (bacille Calmette-Guerin) and corynebacterium parvum. Methods of immunization to achieve a polyclonal antibody response are well known in the art, as are methods for generating hybridomas and monoclonal antibodies.

For preparation of monoclonal antibodies, any technique which provides for the production of antibody molecules by continuous cell lines in culture may be used. For example, the hybridoma technique originally developed by Kohler and Milstein (1975, Nature 256:495-497), as well as the trioma technique, the human B-celi hybridoma technique (Kozbor et al., 1983, Immunology Today 4:72), and the EBV- hybridoma technique to produce human monoclonal antibodies (Cole et al., 1985, in Monoclonal Antibodies and Cancer Therapy, Alan R. Liss, Inc., pp. 77-96). Monoclonal antibodies can optionally be produced in germ-free animals (see PCT/US90/02545). Human antibodies may be used and can be obtained by using human hybridomas (Cote et al, 1983, Proc. Natl. Acad. Sci. U.S.A. 80:2026-2030) or by transforming human B cells with EBV virus in vitro (Cole et al., 1985, in Monoclonal Antibodies and Cancer Therapy, Alan R. Liss, pp. 77-96). Techniques developed for the production of "chimeric antibodies" (Morrison et al., 1984, Proc. Natl. Acad. ScL U.S.A. 81 :6851 -6855: Neuberger et al., 1984, Nature 312:604-608; Takeda et al., 1985, Nature 314:452-454) by splicing the genes from a mouse antibody molecule specific for desired epitopes together with genes from a human antibody molecule of appropriate biological activity can be used.

Antibodies may include, but are not limited to, polyclonal, monoclonal, chimeric (i.e., "humanized" antibodies), single chain (recombinant), Fab fragments, and fragments produced by a Fab expression library. Any of these molecules which contain an antigen binding region specific for a phosphopeptide relative to its cognate non- phosphorylated peptide may be used. These molecules can be used as diagnostic agents for the diagnosis of conditions or diseases (such as cancer) characterized by- expression or overexpression of antigen peptides, or in assays to monitor a patient's responsiveness to an anti-cancer therapy. Antibodies specific for one or more of the antigen phosphopeptides can be used as diagnostics for the detection of the antigen phosphopeptides in cancer cells.

The antibodies or antibody fragments of the present invention can be combined with a carrier or diluent to form a composition. In one embodiment, the carrier is a pharmaceutically acceptable carrier. Such carriers and diluents include sterile liquids such as water and oils, with or without the addition of a surfactant and other pharmaceutically and physiologically acceptable carrier, including adjuvants, excipients or stabilizers. Illustrative oils are those of petroleum, animal, vegetable, or synthetic origin, for example, peanut oil, soybean oil, or mineral oil. In general, water, saline, aqueous dextrose, and related sugar solution, and glycols such as, propylene glycol or polyethylene glycol, are preferred liquid carriers, particularly for injectable solutions.

The antigen phosphopeptides are known to be expressed on a variety of cancer cell types. Thus, they can be used where appropriate, in treating, diagnosing, vaccinating, preventing, retarding, and attenuating melanoma, ovarian cancer, breast cancer, colorectal cancer, squamous carcinoma of the lung, sarcoma, renal ceil carcinoma, pancreatic carcinomas, squamous tumors of the head and neck, leukemia, brain cancer, liver cancer, prostate cancer, ovarian cancer, and cervical cancer.

Antibodies generated with specificity for the antigen phosphopeptides can be used to detect the corresponding phosphopeptides in biological samples. The biological sample could come from an individual who is suspected of having cancer and thus detection would serve to diagnose the cancer. Alternatively, the biological sample may come from an individual known to have cancer, and detection of the antigen phosphopeptides would serve as an indicator of disease prognosis, cancer characterization, or treatment efficacy. Appropriate immunoassays are well known in the art and include, but are not limited to, immunohistochemistry, flow cytometry, radioimmunoassay, western blotting, and EL1SA. Biological samples suitable for such testing include, but are not limited to, cells, tissue biopsy specimens, whole blood, plasma., serum, sputum, cerebrospinal fluid, pleural fluid, and urine. Antigens recognized by T cells, whether helper T lymphocytes or CTL, are not recognized as intact proteins, but rather as small peptides that associate with class I or class II MHC proteins on the surface of cells. During the course of a naturally occurring immune response antigens that are recognized in association with class II MHC molecules on antigen presenting cells are acquired from outside the cell, internalized, and processed into small peptides that associate with the class II MHC molecules. Conversely, the antigens that give rise to proteins that are recognized in association with class I MHC molecules are generally proteins made within the cells, and these antigens are processed and associate with class I MHC molecules. It is now well known that the peptides that associate with a given class I or class II MHC molecule are characterized as having a common binding motif, and the binding motifs for a large number of different class I and II MHC molecules have been determined. It is also well known that synthetic peptides can be made which correspond to the sequence of a given antigen and which contain the binding motif for a given class I or II MHC molecule. These peptides can then be added to appropriate antigen presenting cells, and the antigen presenting cells can be used to stimulate a T helper cell or CTL response either in vitro or in vivo. The binding motifs, methods for synthesizing the peptides, and methods for stimulating a T helper cell or CTL response are all well known and readily available. Kits may be composed for help in diagnosis, monitoring, or prognosis. The kits are to facilitate the detecting and/or measuring cancer-specific phosphoproteins. Such kits contain in a single or divided container, a molecule comprising an antigen-binding region. Such molecules are antibodies or antibody-like molecules. Additional components which may be included in the kit include solid supports, detection reagents, secondary antibodies, instructions for practicing, vessels for running assays, gels, control samples, and the like. The antibody or antibody-like molecules may be directly labeled, as an option.

The antigens of this invention may take the form of antigen peptides added to autologous dendritic cells and used to stimulate a T helper cell or CTL response in vitro. The in vitro generated T helper cells or CTL can then be infused into a patient with cancer (Yee et al, 2002), and specifically a patient with a form of cancer that expresses one or more of antigen phosphopeptides. The antigen phosphopeptides may also be used to vaccinate an individual. The antigen phosphopeptides may be injected alone, but most often they would be administered in combination with an adjuvant. The phosphopeptides may also be added to dendritic cells in vitro, with the loaded dendritic cells being subsequently transferred into an individual with cancer in order to stimulate an immune response. Alternatively, the loaded dendritic cells may be used to stimulate CD8 ^r T cells ex vivo with subsequent reintroduction of the stimulated T cells to the patient. Although a particular phosphopeptide may be identified on a particular cancer ceil type, it may be found on other cancer cell types. Thus a particular phosphopeptide may have use for treating and vaccinating against multiple cancer types.

Phosphopeptide analogs can readily be synthesized that retain their ability to stimulate a particular immune response, but which also gain one or more beneficial features, such as those described below. a. Substitutions may be made in the phosphopeptide at residues known to interact with the MITC molecule. Such substitutions can have the effect of increasing the binding affinity of the phosphopeptide for the MHC molecule and can also increase the half-life of the phosphopeptide-MHC complex, the consequence of which is that the analog is a more potent stimulator of an immune response than is the original peptide. b. Additionally, the substitutions may have no effect on the immunogenicity of the phosphopeptide per se, but rather than may prolong its biological half-life or prevent it from undergoing spontaneous alterations which might otherwise negatively impact on the immunogenicity of the peptide.

The antigen phosphopeptides of this invention can also be used as a vaccine for cancer, and more specifically for melanoma, leukemia, ovarian, breast, colorectal, or lung squamous cancer, sarcoma, renal cell carcinoma, pancreatic carcinomas, squamous tumors of the head and neck, brain cancer, liver cancer, prostate cancer, ovarian cancer, and cervical cancer. The antigens may take the form of phosphoproteins, or phosphopeptides. The vaccine may include only the antigens of this invention or they may include other cancer antigens that have been identified. Pharmaceutical carriers, diluents and excipients are generally added that are compatible with the active ingredients and acceptable for pharmaceutical use. Examples of such carriers include, but are not limited to, water, saline solutions, dextrose, or glycerol. Combinations of carriers may also be used. The vaccine compositions may further incorporate additional substances to stabilize pH, or to function as adjuvants, wetting agents, or emulsifying agents, which can serve to improve the effectiveness of the vaccine.

The composition may be administered parenteral y, either systemically or topically. Parenteral routes include subcutaneous, intravenous, intradermal, intramuscular, intraperitoneal, intranasal, transdermal, or buccal routes. One or more such routes may be employed. Parenteral administration can be, for example, by bolus injection or by gradual perfusion over time. Alternatively, or concurrently, administration may be by the oral route.

It is understood that a suitable dosage of an immunogen will depend upon the age, sex, health, and weight of the recipient, the kind of concurrent treatment, if any, the frequency of treatment, and the nature of the effect desired, how^rever, the most preferred dosage can be tailored to the individual subject, as determined by the researcher or clinician. The total dose required for any given treatment will commonly be determined with respect to a standard reference dose based on the experience of the researcher or clinician, such dose being administered either in a single treatment or in a series of doses, the success of which will depend on the production of a desired immunological result (i.e., successful production of a T helper cell and/or CTL-mediated response to the antigen, which response gives rise to the prevention and/or treatment desired). Thus, the overall administration schedule must be considered in determining the success of a course of treatment and not whether a single dose, given in isolation, would or would not produce the desired immunologically therapeutic result or effect. Thus, the therapeutically effective amount (i.e., that producing the desired T helper cell and/or CTL-mediated response) will depend on the antigenic composition of the vaccine used, the nature of the disease condition, the severity of the disease condition, the extent of any need to prevent such a condition where it has not already been detected, the manner of administration dictated by the situation requiring such administration, the weight and state of health of the individual receiving such administration, and the sound judgment of the clinician or researcher. Needless to say, the efficacy of administering additional doses, and of increasing or decreasing the interval, may be re-evaluated on a continuing basis, in view of the recipient's immunocompetence (for example, the level of T helper cell and/or CTL activity with respect to tumor-associated or tumor- specific antigens).

The concentration of the T helper or CTL stimulatory peptides of the invention in pharmaceutical formulations are subject to wide variation, including anywhere from less than 0.01% by weight to as much as 50% or more. Factors such as volume and viscosity of the resulting composition should also be considered. The solvents, or diluents, used for such compositions include water, possibly PBS (phosphate buffered saline), or saline itself, or other possible carriers or excipients. The immunogens of the present invention may also be contained in artificially created structures such as liposomes, which structures may or may not contain additional molecules, such as proteins or polysaccharides, inserted in the outer membranes of said structures and having the effect of targeting the liposomes to particular areas of the body, or to particular cells within a given organ or tissue. Such targeting molecules may commonly be some type of immunoglobulin. Antibodies may work particularly well for targeting the liposomes to tumor cells.

[42] The vaccine compositions may be used prophylacticaily for the purposes of preventing, reducing the risk of, delaying initiation of a cancer in an individual that does not currently have cancer. Or they may be used to treat an individual that already has cancer, so that recurrence or metastasis is delayed or prevented. Prevention relates to a process of prophylaxis in which the individual is immunized prior to the induction or onset of cancer. For example, individuals with a history of severe sunburn and at risk for developing melanoma, might be immunized prior to the onset of the disease. Alternatively, individuals that already have cancer can be immunized with the antigens of the present invention so as to stimulate an immune response that would be reactive against the cancer. A clinically relevant immune response would be one in which the cancer partially or completely regresses and is eliminated from the patient, and it would also include those responses in which the progression of the cancer is blocked without being eliminated. Similarly, prevention need not be total, but may result in a reduced risk, delayed onset, or delayed progression or metastasis.

[43] The above disclosure generally describes the present invention. All references disclosed herein are expressly incorporated by reference. A more complete understanding can be obtained by reference to the following specific examples which are provided herein for purposes of illustration only, and are not intended to limit the scope of the invention.

EXAMPLE 1

[44] The present example encompasses inter alia a set of phosphoryiated peptides presented by HLA A*0101, A*0301 and B*4402 on the surface of melanoma cells that have the potential to (a) stimulate an immune response to the cancer, (b) to function as immunotherapeutics in adoptive T-cell therapy or as a vaccine, (c) to facilitate antibody recognition of the tumor boundaries in surgical pathology samples, and (d) act as biomarkers for early detection of the disease. The present invention provides at least 246 class 1 MHC peptides presented on the surface of melanoma cells in association with the HLA molecules A*0101, A*0301, and B*4402.

Tables 2A through 2E, are shown in Figure 2A-2E. Sequence identifiers are listed in the first column. UniProt database sequences provide the sequences of the full human proteins from which the peptides are derived. The UniProt sequences are incorporated by reference.

The class I phosphopeptide antigens reported here allow adoptive T-cell therapy to be extended to melanoma patients that do not express the HLA-A*0201 allele and also make it possible to treat a variety of other cancers by the same approach.

We have also shown that we can clone the T-cell receptor on the murine cytotoxic T- celis and then inject the corresponding DNA into normal human T-cells. This process turns them into cytotoxic T-cells that now recognize cancer cells that express the same class 1 phosphopeptides derived from IRS-2 and β-catenin. In short, we have now demonstrated that this process can be used to convert cancer patient T- cells into activated cytotoxic T-cell that recognize class I phosphopeptides and kill their tumor. These experiments also open the door for using class I phosphopeptides in adopted T- cell therapy of cancer. This approach has shown dramatic success in the treatment of advanced stage metastatic melanoma. In conclusion, it should be noted that HLA A*0201 and HLA *A0301 both present peptides from the IRS-2 protein that contain the same phosphorylation site, Serl 100. RVApSPTSGV (SEQ ID NO: 1289) binds to HLA A*0201 and both RVApSPTSGVK (SEQ ID NO: 53) and RVApSPTSGVKR (SEQ ID NO: 54) bind to HLA A*0301. Neither of the A*0301 peptides bind to A*0201 and the A*0201 peptide cannot be presented by the A*0301 molecule. References

The disclosure of each reference cited is expressly incorporated herein.

1, The Cancer Vaccine Roller Coaster, Goldman B and DeFrancesco, L, Nature Biotech, 2009, 27, 129 -

2, Phosphopeptide Antigens Associated with MHC Molecules, US 2005/0277161 Al , Dec 15, 2005

3, P osphoiylated Peptides are Naturally Processed and Presented by Major Hitocompatibility Complex Class 1 Molecules In Vivo, Zarlmg AL, Ficarro SB, White FM, Shabanowitz J, Hunt DF, and Engelhard VH, J. Exp. Med., 2000, 192, 1755-1762.

4, Identification of Class I MHC-Associated Phosphopeptides as Targets for Cancer Immunotherapy, Zarling AL, Polefrone JM, Evans AM, Mikesh LM, Shaban ewis ST, Engelhard VH, and Hunt DF, Proc Natl Acad Sci, USA, 2006, 103, 14889-14894.

5, Phosphorylation-Dependent Interaction between Antigenic Peptides and MHC Class I: A Molecular Basis for the Presentation of Transformed Self, Mohammed F, Cobbold M, Zarling AL, Salim M, Barrett- Wilt GA, Shabanowitz J, Hunt DF, Engelhard VE, Willcox BE, Nat. Immunol. 2008, 11, 1236-43.

6, Adoptive Cell Therapy for Patients with Metastatic Melanoma: Evaluation of Intensive Myeloablative Chemoradiation Preparative Regimens, Dudley ME, Yang JC, Sherry' R, Hughes MS, Royal R, Kammula U, Robbins PF, Huang JP, Citrin DE, Lehman SF, Wunderlich J, Resiifo NP, Tomasian A, Downey SG, Smith FO, Klaper J, Morton K, Laurencot C, White DE and Rosenberg SA, J. Clin Oncology, 2008, 26, 5233-5239.

7, Adoptive Cell Therapy for the Treatment of Patients with Metastatic Melanoma., Rosenberg SA and Dudley ME, Curr Opinion In Immun, 2009, 21 , 233-240

Claims

1. An isolated and purified phosphopeptide consisting of between 8 and 50 contiguous amino acid residues derived from a native human protein, said phosphopeptide comprising a sequence selected from SEQ ID NO: l-139 L wherein at least one serine, threonine, or tyrosine residue in the selected sequence is phosphorylated with a hydrolyzable or non-hydrolyzabie phosphate group, wherein the contiguous amino acids adjacent to the selected sequence in the phosphopeptide are the adjacent contiguous amino acid residues in the native human protein, wherein when the sequence is selected from SEQ ID NO: 1266- 1297, the phosphopeptide is

phosphorylated with a non- hydrolyzable phosphate group.

2. A composition comprising the phosphopeptide of claim L substantially free of other peptides.

3. A composition comprising the phosphopeptide of claim L substantially free of

human cells.

4. A composition comprising the phosphopeptide of claim 1 in a complex with an HLA- A*0101, A*Q3QL B*4402, B*2705, B* 1402, or B*0702 molecule.

5. The composition of claim 4 wherein the complex is a tetramer.

6. The composition of claim 3 which comprises an admixture with one or more distinct peptides.

7. The composition of claim 6 wherein the one or more distinct peptides are melanoma- specific peptides or leukemia-specific peptides.

8. The composition of claim 3 which further comprises an immune adjuvant.

9. The composition of claim 3 wherein at least one phosphopeptide that binds to A*0201 is also in the composition.

10. The composition of claim 3 comprising an admixture of phosphopeptides, wherein a least one peptide that binds to each of HL A- A*0101 , A *0301, B*4402, B*2705, and B*0702 molecule is present in the admixture.

1 1. The composition of claim 10 wherein at least one peptide that binds to HLA-A*0201 is present in the admixture.

12. The method of claim 1 wherein the phosphopeptide is phosphorylated with a non- hydrolyzable phosphate group which is a -CF2-PO₃H group.

13. The method of claim 1 wherein the phosphopeptide is phosphorylated with a non- hydrolyzable phosphate group which is a -CH2-PO₃H group.

14. A method of immunizing a mammal to diminish the risk of, the growth of, or the invasiveness of a cancer, comprising:

administering to the mammal a composition according to any of claims 2-10, whereby CDS¹ T cells are activated.

15. The method of claim 14 wherein the peptide comprises at least 15 amino acid

residues.

16. The method of claim 14 further comprising administering TLR-ligand

ofigonueleotide-CpG.

17. The method of claim 14 wherein at least two phosphopeptides are administered that share a sequence of at least 6 amino acid residues.

18. The method of claim 14 wherein the mammal is a transgenic non-human comprising a human HLA.

19. The method of claim 18 wherein the mammal is a dog.

20. The method of claim 14 wherein the mammal has a melanoma.

21. The method of claim 14 wherein the mammal has metastatic melanoma.

22. The method of claim 14 wherein the mammal has an increased risk of developing a melanoma.

23. A method comprising:

contacting a. sample isolated from a patient with an antibody that specifically binds to the phosphopeptide of claim 1 and does not bind to a peptide consisting of the same amino acid sequence but devoid of phosphorylation; measuring or detecting antibody bound to the sample.

24. The method of claim 23 wherein the sample is blood, serum, plasma, or lymph.

25. The method of claim 23 wherein the sample is a biopsy sample from tumor or normal tissue.

26. The method of claim 23 wherein the sample is from a lymph node.

2.7. A molecule comprising an antigen-binding region of an antibody, wherein the

molecule specifically binds to the phosphopeptide of claim 1 and does not bind to a peptide consisting of the same amino acid sequence but devoid of phosphorylation.

28. The molecule of claim 27 which is a monoclonal antibody.

29. The molecule of claim 27 which is a single chain variable region (ScFv).

30. A kit for measuring a p osphoprotein, said phosphoprotein comprising a sequence selected from SEQ ID NO: 1-1391 and including a phosphorylated serine, threonine, or tyrosine residue, comprising;

a molecule comprising an antigen-binding region of an antibody, wherein the molecule specifically binds to the phosphoprotein and does not bind to a protein consisting of the same amino acid sequence but devoid of phosphorylation.

31. The kit of claim 30 further comprising an antibody which specifically binds to a portion of the molecule which is distinct from the antigen-binding region.

32. The kit of claim 30 further comprising a detectable label.

33. The kit of claim 30 further comprising a solid support on which binding complexes of the molecule and the phosphoprotein can be captured.

34. A method, comprising:

contacting dendritic cells in vitro with an isolated phosphopeptide comprising between 8 and 50 contiguous amino acids comprising a sequence selected from SEQ ID NO: 1 -1391 , said phosphopeptide including at least one serine, threonine, or tyrosine residue that is phosphor lated, whereby the dendritic cells become phosphopeptide-loaded, wherein when the sequence is selected from SEQ ID NO: 1266-1297, the phosphopeptide is phosphorylated with a non-hydrolyzable phosphate group.

35. The method of claim 34 further comprising:

transfusing or injecting the phosphopeptide-loaded dendritic cells into a melanoma patient, wherein the sequence is not SEQ ID NO: 165-170, 173- 179, 181, 184-1 85, 187-190, 192-203, 205-232, 234, 237-238, 240-246, 947- 977, 984-1004, 1009-1011 , 1018-1024, 1028-1043, 1047-1086, 1090-1214, 1218-1220, 1227-1232, and 1236-1265.

36. The method of claim 34 further comprising:

transfusing or injecting the phosphopeptide-loaded dendritic cells into a leukemia patient, wherein the sequence is selected from the group consisting of SEQ ID NO: 165-170, 173-179, 181 , 184-185, 187-203, 205-232, 234-235, 237-238, 240-246, 947-977, 984-1004, 1009-1011, 1018-1024, 1028-1086, 1047- 1086, 1090-1214, 1218- 1223, 1227-1232, and 1236-1265.

37. The method of claim 34, further comprising:

contacting the phosphopeptide-loaded dendritic ceils with CD8⁺ T cells in vitro, whereby the CD8⁺ T cells are stimulated.

38. The method of claim 37 further comprising:

transfusing the stimulated CD8⁺ T cells into a melanoma or leukemia patient.

39. The method of claim 38 wherein the CD8 T cells are autologous to the patient.

40. The method of claim 38 wherein the CD8^T T cells are allogeneic to the patient.

41. The method of claim 34 wherein the dendritic cells are contacted with a plurality of said isolated phosphopeptides.

42. The method of claim 34 wherein the dendritic cells are contacted with a plurality of said isolated phosphopeptides which are linked by a spacer of 10-50 amino acid residues

43. An in vitro composition comprising dendritic cells made by the method of claim 33, wherein the dendritic cells are loaded with a phosphopeptide consisting of between 8 and 14 contiguous amino acids comprising a sequence selected from SEQ ID NO: l - 1391, said phosphopeptide including at least one serine, threonine, or tyrosine residue that is phosphorylated, wherein when the sequence is selected from SEQ ID

MO: 1266- 1297, the phosphopeptide is phosphorylated with a non-hydrolyzable phosphate group.

44. The composition of claim 43 wherein the phosphopeptide comprises at least one amino acid residue that is not in its native human protein,

45. The composition of claim 44 wherein the at least one amino acid residue is an optimal anchor residue for its corresponding HLA molecule.

46. The composition of claim 43 wherein the phosphopeptide is phosphorylated with a non-hydrolyzable phosphate group which is a -CF₂-PO₃H group.

47. The composition of claim 43 wherein the phosphopeptide is phosphorylated with a non-hydrolyzable phosphate group which is a -CH₂-P(¾H group.

48. A synthetic phosphopeptide consisting of from 10-50 amino acid residues, comprising the sequence RVAsPTSGVK (SEQ ID NO: 53) or RVAsPTSGVKR (SEQ ID NO: 54), wherein the serine residue at position 4 is phosphorylated with a hydroiyzable or non-hydrolyzable phosphate group, and wherein adjacent amino acid residues to the sequence are adjacent sequences in human insulin substrate-2 (IRS -2) protein.

49. The synthetic phosphopeptide of claim 48 which is in a complex with A*0301 ,

50. A concatamer of at least two phosphopeptides according to claim 1, wherein the phosphopepiides are linked by a spacer of 10-50 amino acid residues.