AU2003236645A1 - Method for identifying immunoreactive peptides - Google Patents

Abstract

The invention relates to a method for identifying immunoreactive peptides. According to said method, a sample consisting of tumorous and corresponding healthy tissue is first prepared, the tumour-specific expression profile is subsequently determined and antigenic peptides are isolated from the tumorous tissue and analysed. The respective data that has been obtained is then matched and peptides are identified on the basis of said matched data.

Description

CONFIRMATION I, Dr. H. Otten Rotebthlstrafe 121 70178 Stuttgart Germany declare that I am conversant with the German and English languages and that to the best of my knowledge and belief the accompanying document is a true translation of the International Patent Application No. PCT/EPO3/05038 in the name of Immatics Biotechnologies GmbH Paul-Ehrlich-Str. 15 72076 TUbingen Germany Signed this 18 th day of November 2004 (Dr. H. Otten) Method for identifying immunoreactive peptides The invention relates to a method for identifying and to a method for preparing immunoreactive peptides and to the immuno reactive peptides identified/prepared thereby. Such peptides are being used - for example - in immunotherapy of tumor-associated diseases. When tumor cells are eliminated by the immune system the identification of tumor-associated antigens (TAA) by components of the immune system plays a piv otal role. This mechanism is based upon the fact that there exist qualitative or quantitative differences between tumor cells and normal cells. To induce an anti-tumor response, the tumor cells have to express antigens which induce an immune response being sufficient for the elimination of the tumor. CD8 expressing cytotoxic T lymphocytes (in the following CTL), in particular, are involved in rejection of tumors. To induce 2 such an immune reaction by cytotoxic T cells foreign pro teins/peptides have to be presented to T cells. Antigens are recognized as peptide fragments by T cells only if they are presented by MHC-molecules on cell surfaces. These MHC ("major histocompatibility complex") molecules are peptide receptors which normally bind peptides intracellularly and transport them to the cell surface. This complex of peptide and MHC-molecule is recognized by T cells. Human MHC-molecules are also desig nated as human leukocyte antigens (HLA). In the past, antigen-specific immunotherapy based on T cells has proven successful in the treatment of cancer. Induction of a specific CTL response directed against a tumor, is dependent on identification of MHC class I-ligands derived from tumor-associated antigens (TAA). Such TAA can be exclu sively present in malignant cells, for example as products of mutated genes. Other important classes of tumor-associated antigens are tissue-specific structures such as the cancer testis antigens, and a third class of tumor-associated antigens are proteins overexpressed in tumors. The methods for identification and characterization of TAA, which represent the starting point for a tumor vaccine, are on the one hand - based on the use of patient derived CTL or antibodies. This immunological approach is combined either with a gene expression approach or with a mass-spectrometry (MS) assisted sequencing of the recognized peptides (see van der Bruggen et al., 1991, A gene encoding an antigen recognized by cytolytic T lymphocytes on a human melanoma, Science 254: 1643 1647, and Cox et al., 1994, Identification of a peptide recog- 3 nized by five melanoma-specific human cytotoxic T cell lines, Science 264: 716-719). Methods for identifying TAA which are based on comparative transcription profiling of tumorous and corresponding normal tissue are - for example - hybridization and the use of DNA microarray technology. Celis et al., 1994, Induction of anti-tumor cytotoxic T lympho cytes in normal humans using primary cultures and synthetic peptide epitopes, Proc. Natl. Acad. Sci. USA 91: 2105-2109, applied a method, which takes advantage of the prediction of MHC class I-ligands derived from a selected tumor-associated antigen, and in which these ligands were verified as T cell epitopes in a next step. Drawbacks of patients-derived T cell-based approaches are the extensive culture techniques and their restriction to the fre quency of pre-existing T cells. Another T cell-independent approach also known in the art com bines epitope prediction and screening for predicted peptides in complex peptide mixtures, the peptides being identified by highly sensitive capillary liquid chromatography mass spec trometry (LC-MS) (see Schirle et al., 2000, Identification of tumor-associated MHC class I-ligands by a novel T cell inde pendent approach, Eur. J. Immunol. 30: 2216-2225). DNA microarray technology provides a new approach using com parative expression profiling of tumorous and corresponding autologous normal tissue. Young et al., 2001, Expression pro filing of renal epithelial neoplasms: a method for tumor clas sification and discovery of diagnostic molecular markers, Am.

4 J. Pathol. 158: 1639-1651, disclosed that by using this tech nique a large number of tumor-associated antigens from individ ual tumor samples can be identified. MHC-I-ligands derived from overexpressed or selectively expressed proteins provide possi ble targets for specific CTL recognition of tumors. Mathiassen et al., 2001, Tumor-associated antigens identified by mRNA expression profiling induce protective anti-tumor immunity, Eur. J. Immuno. 31: 1239-1246, demonstrated, in a mouse model, that by combining expression analysis with epitope prediction a successful vaccine can be prepared. However, a disadvantage is that epitope prediction even for only a few target genes results in the identification of a vast number of candidate peptides, the majority of which are actu ally not presented by MHC molecules and thus do not induce a CTL-response. Therefore, it is an object of the present invention to provide a new method for selective and simple identification of immuno reactive peptides. According to the invention, this object is achieved by means of a method for identifying immunoreactive peptides comprising: (a) providing a sample of tumorous and corresponding healthy tissue; (b) determining tumor-specific expression profile of the pro vided sample, 5 (c) isolating and analyzing of antigenic peptides in the sam ple of tumorous tissue, (d) matching data obtained in step (b) and (c), and (e) identifying peptides on basis of the matched data. The inventors revealed that by combining an expression analysis with antigenic tumor peptides, which have been isolated and analyzed, specific candidates for an individual vaccine can be identified. By isolating antigenic peptides and matching them with gene expression profiles of tumorous tissue it can be avoided that a vast number of possible immunoreactive peptides is obtained. Rather, specific peptides are identified, which are actually presented by MHC-molecules and which are thus suitable as immu noreactive peptides. With the method according to the invention it is, respectively, possible to identify patient-specific peptides, i.e. it is possible to precisely match peptides, which are to be used as vaccine, to the patient, in order to induce a specific immune response. For example, industrial laboratories - after having received patient samples - can systematically and efficiently perform this method, and can - after having identified suitable immuno reactive peptides - provide clinics in charge with the peptide sequences; the clinics can then synthesize and administer the peptides. Nevertheless, it is also possible that a laboratory 6 is carrying out identification as well as production of the peptides suitable for the respective patient. Therefore, the new method is applicable within the scope of a mere service as well as in combination with the supply of the identified immunoreactive peptide. In a preferred embodiment the peptides isolated in step (c) are MHC-ligands. Only peptides which are bound to MHC-molecules can induce a cellular immune response. Peptides which are derived, for exam ple, from overexpressed genes of a tumor but which are not bound to MHC-molecules, do not induce a CTL immune reaction. Therefore, not all peptides, for example, identified solely by epitope prediction are immunoreactive. In a further preferred embodiment step (b) is performed by means of a microarray-analysis and/or reverse transcription polymerase chain reaction. With microarray analysis the expression profile of tumorous tissue is compared with corresponding normal tissue by using certain DNA- or gene-chips, and selectively expressed or over expressed genes are identified. This method is well known in the art and is, for example, disclosed in Schena et al., 1995, Quantitative monitoring of gene expression patterns with a complementary DNA microarray, Science 270: 467-470. Reverse transcription-polymerase chain reaction (in the follow ing RT-PCT) can be utilized to quantify expression of a gene.

7 For this purpose cDNA is generated from RNA, which - for exam ple - has been isolated from tumor cells, and the cDNA is con sequently used as template for PCR. In that way, on the basis of the amplified DNA one can compare which genes are being transcribed with which intensity. In a preferred embodiment step (c) is performed by means of a mass spectrometer. Using this technique each peptide can be identified precisely and efficiently with high performance. For example, Schirle et al., 2000, Identification of tumor-associated MHC class I ligands by a novel T cell independent approach, Eur. J. Immu nol. 30: 2216-2225, describe the use of mass spectrometry to identify peptides derived from tumor tissue. In yet a further embodiment, in step (c) candidate antigenic peptides are predicted on the basis of the expression profile using suitable databases, and the mass spectrometer is cali brated with respect to the predicted antigens. Use of databases for prediction of candidate antigens and utilizing the obtained data is - for example - disclosed in Schirle et al., 2001, Combining computer algorithms with experimental approaches permits rapid and accurate identification of T cell epitopes from defined antigens, J. Immunol. Methods, 257: 1-16. Examples for databases can for example be found under http://www.syfpeithi.de and http://www.paproc.de. In yet a further preferred embodiment of the method, step (c) is followed by a further step, in which the reactivity of pe- 8 ripheral leukocytes, preferably of T leukocytes, against the isolated antigenic peptides, is tested. In a preferred embodiment the reactivity of peripheral leuko cytes against the isolated antigenic peptides is tested by means of measuring y-Interferon-mRNA and/or cytokin-mRNA syn thesized by the leukocytes. By detecting y-Interferon- or cytokin-mRNA it is possible to precisely prove the specific reactivity of leukocytes, prefera bly of T lymphocytes against antigenic peptides. Both sub stances are secreted by activated T lymphocytes after their activation by corresponding antigenic peptides. With this additional step candidates of the already identified peptides can be identified even more precisely. In yet another preferred embodiment of the method, following step (c) a further step is performed, in which the presence of the T lymphocytes is detected. Using this method it is possible to specifically detect to what extent T lymphocytes directed against isolated and identified peptides are pre-existing in patients. By performing this step it is possible to apply, as a vaccine, only those peptides against which T lymphocytes are already pre-existing in the patient. The peptides can then be used to activate these spe cific T lymphocytes. In a further preferred method the detection of specific pre existing T lymphocytes is performed by labeling the leukocytes 9 with reconstituted complexes of antigen-presenting molecules and antigenic peptide. With this method the so-called tetramer-technology is utilized. A method for generating such reconstituted complexes ("tetram ers") and for utilizing them is disclosed, for example, in Altman et al., 1996, Phenotypic analysis of antigen-specific T lymphocytes, Science 274: 94-96. The invention further relates to immunoreactive peptides, which are identified and prepared by the method according to the invention. After identification these peptides can be selectively and specifically prepared for each patient. The invention further relates to a pharmaceutical composition comprising one or more peptides which have been identified and/or prepared by the method according to the invention. The composition may be applied, for example, parenterally, for example subcutaneously, intradermally or intramuscularly, or may be administered orally. In doing so, the peptides are dis solved or suspended in a pharmaceutically acceptable carrier, preferably an aqueous carrier; the composition can further comprise additives, for example buffers, binders, etc. The peptides can also be administered together with immunostimulat ing substances, for example cytokins. An extensive description of additives which can be used in compositions of this nature is given, for example, in A. Kibbe, Handbook of Pharmaceutical 10 Excipients, 3. Ed., 2000, American Pharmaceutical Association and Pharmaceutical Press. According to the invention the peptide may be used for treat ment of tumor diseases and for preparing a medicament for treatment of tumor diseases. Tumor diseases to be treated comprise renal, breast, pancreas, gastric, testis and/or skin cancer. Listing of tumor diseases is supposed to be merely illustrative and shall not limit the scope of usage. The peptide can further be used for assessment of the therapy course of a tumor disease. The peptide can also be used for monitoring a therapy in other immunizations or therapies. In that way the peptide according to the invention may not only be used in a therapeutical way but also in a diagnostic way. In a further embodiment the peptides are used for generating an antibody. Polyclonal antibodies can be obtained, in a general manner, by immunization of animals by means of injection of the peptides and subsequent purification of the immunoglobulin. Monoclonal antibodies can be generated according to standard ized protocols, for example as described in Methods Enzymol. (1986), 121, Hybridoma technology and monoclonal antibodies.

11 In a further aspect the invention further relates to nucleic acid molecules coding for the peptide isolated with the method according to the invention. The nucleic acid molecules can be DNA- or RNA-molecules and can be used for immune therapy of cancer as well. According to the invention the nucleic acid molecules can be provided in a vector. The invention further relates to a cell genetically modified by means of the nucleic acid molecule, such, that the cell is producing a peptide identified according to the invention. The invention further relates to a method for preparing an immunoreactive peptide with which a peptide is identified ac cording to the disclosed method and the identified peptide is synthesized chemically, in vitro or in vivo. Peptides can be prepared by chemical linkage of amino acids, for example by the method of Merrifield, which is known in the art (see Merrifield RB, 1963, J. Am. Chem. Soc. 85: 2149-2154). Peptides can be prepared in vitro, for example, in cell-free systems, and in vivo by using cells. A preferred embodiment of the present invention is a method for preparing a vaccine comprising the steps of a) performing the disclosed method, b) preparing the identified immunoreactive peptides, and 12 c) formulating the prepared immunoreactive peptides. It will be understood that the features which are mentioned above and the features still to be explained below can be used not only in the combinations which are in each case specified but also in other combinations or on their own without depart ing from the scope of the present invention. Embodiments of the invention are displayed and explained in the figures and the example below. Fig. 1 shows the expression analysis of selected genes by means of quantitative RT-PCR; Fig. 2 shows the detection of keratin 18-specific CD8 t T lymphocytes. Example Patient samples Samples of patients having histologically confirmed renal cell carcinoma were obtained from the department of urology, Univer sity of TUbingen. Both patients had not received preoperative therapy. Patient No. 1 (in the following designated RCC01) had the following HLA-typing: HLA-A'02 A'68 B*18 B'44; patient No. 2 (in the following designated RCC13) HLA-A*02 A*24 B'07 B'40.

13 Isolation of MHC class I-bound peptides Shock-frozen tumor samples were processed as described in Schirle, M. et al., Identification of tumor-associated MHC class I ligands by a novel T cell-independent approach, 2000, European Journal of Immunology, 30: 2216-2225. Peptides were isolated according to standard protocols using monoclonal anti body W6/32 being specific for HLA class I or monoclonal anti body BB7.2 being specific for HLA-A2. Production and utiliza tion of these antibodies is described by Barnstable, C.J. et al., Production of monoclonal antibodies to group A erythro cytes, HLA and other human cell surface antigens - New tools for genetic analysis, 1978, Cell, 14:9-20 and Parham, P. & Brodsky, F.M., Partial purification and some properties of BB7.2. A cytotoxic monoclonal antibody with specificity for HLA-A2 and a variant of HLA-A28, 1981, Hum. Immunol., 3: 277 299. Mass spectrometry Peptides from tumor tissue of patient RCC01 were separated by reversed phase HPLC (SMART-system, pRPC C2/C18 SC 2.1/19, Amer sham Pharmacia Biotech) and fractions were analyzed by nanoESI MS. In doing so it was proceeded as described in Schirle, M. et al., Identification of tumor-associated MHC class I ligands by a novel T cell-independent approach, 2000, European Journal of Immunology, 30: 2216-2225. Peptides from tumor tissue of patient RCC13 were identified by online capillary LC-MS as mentioned above with minor modifica tions: Sample volumes of about 100 Ml were loaded, desalted and 14 preconcentrated on a 300 pm * 5 mm C18 p-precolumn (LC pack ings). A syringe pump (PHD 2000, Harvard Apparatus, Inc.) equipped with a gastight 100 pl-syringe (1710 RNR, Hamilton), delivered solvent and sample at 2 pl/min. For peptide separa tion, the preconcentration column was switched in line with a 75 pm * 250 mm C-18-column (LC packings). Subsequently a binary gradient of 25 % - 60 % B within 70 min was performed, applying a 12 pl/min flow rate reduced to approximately 300 nl/min with a precolumn using a TEE-piece (ZT1C, Valco) and a 300 pm * 150 mm C-18-column. A blank run was always included to ensure that the system was free of residual peptides. On-line fragmentation was performed as described and fragment spectra were analyzed manually. Database searches (NCBInr, EST) were made using MASCOT (http://www.matrixscience.com). Preparation of RNA Fragments of normal and malignant renal tissue were dissected, shock-frozen, ground by a mortar and pestle under liquid nitro gen and homogenized with a rotary homogenizer (Heidolph instru ments) in TRIZOL (Life Technologies). Total RNA was prepared according to the manufacturer's protocol followed by a clean-up with RNeasy (QIAGEN). Total RNA from human tissues were ob tained commercially (Human total RNA Master Panel II, Clon tech).

15 High-Density Oligonucleotide Micro-Array Analysis Double-stranded DNA was synthesized from 40 Mg of total RNA using superscript RT II reverse transcriptase (Life Technolo gies). The primer (Eurogentec) were given by the Affymetrix manual. In vitro transcription was performed using the BioAr rayTM High YieldTM RNA Transcript Labeling Kit (ENZO Diagnos tics, Inc.); subsequently, fragmentation and hybridization were carried out on Affymetrix HuGeneFL gene chips, and staining with a streptavidin-phycoerythrin and biotinylated anti streptavidin-antibody followed the manufacturer's protocols (Affymetrix). The Affymetrix GeneArray Scanner was used and data were analyzed with the Microarray Analysis Suite 4.0 Soft ware. Real time RT-PCR cDNA generated for microarray analysis was used for quantita tive PCR analysis. Each gene was run in duplicates (40 cycles, 95 0 C x 15 s, 60 0 C x 1 min) using SYBRGreen chemistry on the ABI PRISM 7700 sequence detection system (Applied Biosystems). Samples were independ ently analyzed two to three times. Primers (MWG-Biotech) were selected to flank an Intron and PCR efficiencies were tested for all primer pairs and found to be close to 1. PCR products were analyzed on 3 % agarose gels for purity and sequence-verified after cloning into pCR4-TOPO vector using the TOPO TA Cloning Kit (Invitrogen). Data analysis involved the delta CT method for relative quantification.

16 Laser Capture Microdissection Embedded frozen tissue specimens were cut at 6 pm thickness and transferred in 70 % ethanol for about 15 min. Slides were incu bated 90 seconds in Mayer's hematoxylin (Merck), rinsed in water, incubated for 1 min in 70 % ethanol, 1 min in 95 % etha nol, 30 seconds in 1 % alcoholic eosin Y (Sigma), 2 x 2 min in 95 % ethanol, 2 x 2 min in 100 % ethanol and finally 2 x 2 min in xylene. After air drying for 15 minutes, slides were stored under dry conditions. Normal malignant epithelial tubular cells and carcinoma cells were isolated by a Laser Capture Microdis section (LCM) using the PixCell II LCM system (Arcturus Engi neering). Total RNA was extracted in 400 pl TRIZOL. PBMC, Tetramer production and flow cytometry Peripheral blood mononuclear cells (in the following PBMC) from two healthy donors (HD1 and HD2), which were serologically typed as CMV-positive, were isolated by gradient centrifugation (FicoLite H) and frozen. HLA-A*0201 tetrameric complexes were produced as described by Altman et al., 1996, Phenotypic analysis of antigen-specific T lymphocytes, Science 274: 94-96, as follows: The HLA-A2-binding peptides used for the refolding were ALLNIKVKL from keratin 18 and NLVPMVATV from pp65 HCMVA. Tetramers were assembled by mixing biotinylated monomers with streptavidin-PE or strepta vidin-APC and 2-3 x 106 cells were incubated 30 min at 4 0 C with both tetramers: 10 pg/ml for each monomer in PBS, 0.01 % NaN 3 , 2 mM EDTA, 50 % fetal calf serum). Then, monoclonal antibodies Anti-CD4-FITC (Coulter-Immunotech) and Anti-CD8-PerCP (Becton 17 Dickinson) were added for 20 min. After three washes, samples were fixed in FACS buffer, 1 % formaldehyde. Four-color analy sis was performed on a FACScalibur cytometer (Becton Dickin son). Results The expression of approximately 7000 genes in tumors and corre sponding normal tissue of two renal cell carcinoma was ana lyzed. Between 400 and 500 genes were found to be overexpressed or selectively expressed in the tumor. 70 genes were overex pressed in the tumors of both patients. In patient 1, 268 over expressed and 129 exclusively expressed genes were found. Most of the overexpressed genes are cancer-related, i.e. either oncogenes, tumor suppressor genes or genes already described as overexpressed in cancer, such as CCND1, CA9, cerebrosidesul fotransferase and parathyroid hormone-like hormone. The cancer associated adipose differentiation-related protein (ADFP) or adipophilin, showed the second highest degree of overexpres sion. In addition, this protein was shown to be highly overex pressed in tumorous tissue in comparison to normal tissue of other organs, that is not only in comparison to normal renal tissue. To verify data obtained by microarray analysis the expression of selected genes was analyzed by quantitative PCR. In Fig. 1 the expression-analysis of selected genes by means of RT-PCR is shown. The RT-PCR was performed using the same cDNA as gener ated for microarray analysis. Copy numbers are relative to 18S rRNA and normalized to the normal tissue (=1) of each patient. The black bars correspond to the numbers in tumor tissue of 18 patient RCC01, the white bars represent the numbers of normal tissues and the angular-striped gray bars the numbers in tumor tissue of patient RCC13. It was shown that overexpression of adipophilin (ADFP) and cycline D1 (CCNDl) as proven by microarray could be confirmed by quantitative PCR. Further, it was demonstrated that ets-1 (ETS1) was expressed equally both in normal and in tumor tis sue. Further, relative expression levels detected by both tech niques were roughly comparable. For example adipophilin was overexpressed in tumor tissue of patient RCC01 by a factor of 29.1 as proven by means of mi croarray, compared to 18.1 as measured by means of quantitative PCR (see Fig. 1). With patient RCC13, by means of microarray analysis a factor of 11.4 and by means of quantitative PCR a factor of 6.7 could be demonstrated (see Fig. 1). Galectin 2 (LGALS2) was overexpressed in patient RCC01 and keratin-18 (KRT18) in patient RCC13. An exception to the congruence be tween microarray and quantitative PCR was the overexpression of KIAA0367 and met proto-oncogene (MET) in patient RCC13. Identification of MHC class I-ligands A total of 85 ligands could be obtained from tumor tissue, which were bound to HLA-subtypes HLA-A*02, HLA-A*68, HLA-B*18 or HLA-B*44. Peptides that bind to HLA-A*02 reflected the al lele-specific peptide motive (Leucine, Valine, Isoleucine, Alanine, Methionine on position 2, Leucine, Valine, Isoleucine or Alanine at the C-Terminus). Most ligands were derived from abundantly expressed housekeeping proteins, but ligands from 19 proteins with reported tumor association could be detected also, for example YVDPVITSI derived from met proto-oncogene, ALLNIKVKL derived from keratin 18, and SVASTITGV from adipo philin. HLA-A*68 ligands were identified by their anchor amino acids Threonine, Isoleucine, Valine, Alanine or Leucine on position 2 and arginine or lysine at the C-terminus. Two other ligands from adipophilin were found among HLA-A*68-presented peptides: MTSALPIIQK and MAGDIYSVFR. Ligands carrying glutamic acid on position 2 were assigned to HLA-B*44; since the peptide motive of HLA-B*18 is unknown, a distinction between ligands of these two HLA-B-molecules was not possible. Comparison of microarray data with the isolated ligands indi cated 10 overexpressed genes as sources for MHC-ligands: adipo philin, KIAA0367, SEC14-like 1, B-cell translocation gene 1, aldolase A, cycline Dl, annexin A4, catenin alpha 1, galectin 2 and LMP2. Three of them were also included in the SEREX data base: KIAA0367, aldolase A and catenin alpha 1. A most interesting ligand could be identified from patient RCC13 (ALAAVVTEV) encoded by a "Reading frame" shifted by one nucleotide compared to DEAD/H-box polypeptide 3 (DDX3). ALAAVVTEV is encoded by the nucleotides 317 to 343 of the cod ing strand of DDX3, whereas nucleotides 316 to 342 are coding for GIGSRGDRS of the DDX3 protein.

20 Detection of specific T cells in normal CD8* T cell repertoire PBMC from 6 HLA-A2 positive renal cell carcinoma patients were tested for reaction against four of the relevant peptides: HLA A'02-restricted ligands from adipophilin, keratin 18, KIAA0367 and met-proto-oncogene. In doing so, a very sensitive quantita tive PCT assay was carried out to detect y-Interferon-mRNA production by CD8 T cells following a 7 day-in vitro sensitization with peptide. Sporadic responses were seen after stimulation with met-proto-oncogene or keratin 18 or adipo philin peptides. Staining of PBMC of tumor patients and healthy individuals with HLA-A'0201 tetramers was performed with tetramers reconstituted either with adipophilin, keratin 18 or met-proto-oncogene. Fig. 2 shows the detection of keratin 18-specific T lympho cytes. For this purpose, PBMC from four healthy HLA-A*02 t do nors (HDI, 2, 4, 6) were simultaneously stained with HLA A2/keratin 18-PE tetramers, HLA-A2/CMV-APC tetramers, CD8-PerCP and CD4-FITC. Dot plots show samples from one of three inde pendent experiments for 1 x 106 PBMC. In the plots, percentage of tetramer -cells within the CD8 CD4- population are indi cated. Unexpectedly, a significant population of CD8 T lymphocytes specific for keratin 18 (between 0.02 % and 0.2 % of CD8' T cells) was found in 4 out of 22 healthy individuals. This popu lation did not stain with a CMV tetramer showing that the bind ing of keratin 18 tetramer was specific.

21 To summarize it can be concluded that CD8 T lymphocytes spe cific for the keratin 18-peptide are contained in the human T cell repertoire.

Claims (20)

1. Method for identifying immunoreactive peptides comprising the following steps: a) providing a sample of tumorous and corresponding healthy tissue, b) determining the tumor-specific expression profile in the provided sample, c) isolation and analysis of antigenic peptides in the sample of tumorous tissue, d) matching the data obtained in steps b) and c), and e) identification of peptides on the basis of the matched data.

2. Method according to claim 1, characterized in that the antigenic peptides are MHC-ligands.

3. Method according to claim I or claim 2, characterized in that step b) is performed by microarray analysis and/or reversed transcriptase-polymerase chain reaction.

4. Method according to one of claims 1 to 3, characterized in that in step c) analysis is performed by mass spectrome try. 23

5. Method according to claim 4, characterized in that in step c) candidate antigenic peptides are predicted on basis of the expression profile using suitable databases, and that the mass spectrometer is calibrated in view of the pep tides.

6. Method according to one of claims 1 to 5, characterized in that after step c) a further step is performed, in which the reactivity of peripheral leukocytes, preferably T lym phocytes, is tested against the isolated antigenic pep tides.

7. Method according to claim 6, characterized in that the reactivity test is performed by means of measuring cyto kine-mRNA and/or y-Interferon mRNA synthesized by the leu kocytes.

8. Method according to one of claims 1 to 5, characterized in that after step c) a further step is performed, in which the presence of specific T lymphocytes is detected.

9. Method according to claim 8, characterized in that detec tion of specific T lymphocytes is carried out by means of labeling leukocytes with reconstituted complexes of anti gen-presenting molecules and antigenic peptide.

10. Method for preparing an immunoreactive peptide, in which a peptide is identified according to the method according to one of claims 1 to 9 and that the identified peptide is synthesized chemically, in vitro or in vivo. 24

11. Immunoreactive peptide identified by a method according to one of claims 1 to 9 and/or has been prepared by a method according to claim 10.

12. Pharmaceutical composition comprising one or more peptides according to claim 11.

13. Use of a peptide according to claim 11 for treatment of a tumor disease.

14. Use of a peptide according to claim 11 for preparing a medicament for the treatment of tumor diseases.

15. Use of a peptide according to claim 11 to assess the ther apy-course of a tumor disease.

16. Use of a peptide according to claim 11 for preparing an antibody.

17. Nucleic acid molecule coding for a peptide according to claim 11.

18. Vector comprising a nucleic acid molecule according to claim 17.

19. Cell genetically modified by a nucleic acid molecule ac cording to claim 17, which is producing a peptide accord ing to claim 11.

20. Method for preparing a vaccine comprising the steps of: 25 a) performing a method according to one of claims 1 to 9, and b) preparing the identified immunoreactive peptides and c) formulating the prepared immunoreactive peptides.