Peptides and combination of peptides for use in immunotherapy against epithelial ovarian cancer and other cancers

Abstract

The present invention relates to peptides, proteins, nucleic acids and cells for use in immunotherapeutic methods. In particular, the present invention relates to the immunotherapy of cancer. The present invention furthermore relates to tumor-associated T-cell peptide epitopes, alone or in combination with other tumor-associated peptides that can for example serve as active pharmaceutical ingredients of vaccine compositions that stimulate anti-tumor immune responses, or to stimulate T cells ex vivo and transfer into patients. Peptides bound to molecules of the major histocompatibility complex (MHC), or peptides as such, can also be targets of antibodies, soluble T-cell receptors, and other binding molecules.

Claims

1. An antibody that specifically binds to a peptide consisting of the amino acid sequence of LPSPVDAAF (SEQ ID NO: 84) or binds to the peptide consisting of the amino acid sequence of LPSPVDAAF (SEQ ID NO: 84) bound to an MHC molecule.

2. A kit comprising: (a) a container comprising a pharmaceutical composition containing the antibody of claim 1 in solution or in lyophilized form; (b) optionally, a second container containing a diluent or reconstituting solution for the lyophilized formulation; (c) optionally, the peptide having the amino acid sequence of LPSPVDAAF (SEQ ID NO: 84), and (d) optionally, instructions for (i) use of the solution or (ii) reconstitution and/or use of the lyophilized formulation.

3. The kit according to claim 2, further comprising one or more of (iii) a buffer, (iv) a diluent, (v) a filter, (vi) a needle, or (v) a syringe.

4. A pharmaceutical composition comprising the antibody of claim 1 and a pharmaceutically acceptable carrier, and optionally, pharmaceutically acceptable excipients and/or stabilizers.

5. The antibody of claim 1, wherein said antibody is a polyclonal antibody, a monoclonal antibody, a bi-specific antibody or a chimeric antibody.

6. The antibody of claim 1, wherein the peptide bound to an MHC molecule is present in a tumor cell.

7. The antibody of claim 6, wherein the tumor cell is at least one selected from the group consisting of lung cancer, kidney cancer, brain cancer, stomach cancer, colon or rectal cancer, liver cancer, prostate cancer, leukemia, breast cancer, Merkel cell carcinoma (MCC), melanoma, ovarian cancer, esophageal cancer, urinary bladder cancer, endometrial cancer, gall bladder cancer, and bile duct cancer, and other tumors that show an overexpression of a protein comprising a peptide having the amino acid sequence of LPSPVDAAF (SEQ ID NO: 84).

8. The antibody of claim 6, wherein the tumor cell is ovarian cancer.

9. The antibody of claim 1, wherein the antibody is labeled with a toxin.

10. The antibody of claim 1, wherein the antibody is labeled with a radionucleotide.

11. The antibody of claim 10, wherein the radionucleotide is .sup.111In, .sup.99Tc, .sup.14C, .sup.131I, .sup.3H, .sup.32P, or .sup.35S.

12. The antibody of claim 1, wherein the affinity value (Kd) of the antibody is less than 1?10 ?M.

13. The antibody of claim 1, wherein the MHC is a MHC class I molecule.

14. The antibody of claim 1, wherein the MHC is a MHC class II molecule.

15. The antibody of claim 1, further comprising a binding affinity of below 20 nanomolar.

16. The antibody of claim 1, wherein the antibody is humanized.

17. The composition of claim 4, wherein the pharmaceutically acceptable carrier is saline, Ringer's solution, dextrose solution, and/or solid hydrophobic polymer.

18. The composition of claim 17, wherein the solid hydrophobic polymer is film, liposome, or microparticle.

19. A pharmaceutical composition comprising the antibody of claim 6 and a pharmaceutically acceptable carrier.

20. A pharmaceutical composition comprising the antibody of claim 9 and a pharmaceutically acceptable carrier.

21. A pharmaceutical composition comprising the antibody of claim 10 and a pharmaceutically acceptable carrier.

22. A pharmaceutical composition comprising the antibody of claim 13 and a pharmaceutically acceptable carrier.

Description

FIGURES

(1) FIG. 1 shows the HLA-A,B,C (a) and HLA-DR (b) expression of different cell subsets within ovarian cancer and benign ovarian tissue. For FIG. 1 the two-tailed unpaired Student's t-test with Welch's correction was used owing to unequal variance between the two comparison groups. HLA class I (A) and HLA-DR (B) expression on different cell types within EOC and benign ovarian tissue after enzymatic dissociation characterized by distinct cell surface markers (leukocyte compartments: CD45+, tumor cells/epithelial cell compartments: CD45?EpCam+, endothelial cell compartments: CD45?CD31+). Each data point represents the mean of triplicate experiments performed for each sample. Two sided t-tests were used to test for significance (*p<0.05; * p<0.01).

(2) FIGS. 2A to D show the comparative profiling of the immunopeptidome of EOC vs. benign tissues. (A) Comparative profiling of HLA class I ligand source proteins represented in EOC (n=34) and benign tissues. The frequency of HLA restricted presentation of source proteins is indicated on the y-axis separately for EOC (above x-axis) and benign sources (below x-axis). The source proteins were ranked (from left to right) according to their frequency of EOC specific presentation. The box on the left side highlights the TOP100 HLA ligand source proteins exclusively presented by EOC. (B) Word cloud of the TOP 100 EOC specific HLA class I ligand source proteins (uniprot recommended gene name). Font size (5-26) correlates with absolute number of cancer patients presenting HLA ligands of respective source proteins. (C) Comparative profiling of HLA class II ligand source proteins represented in EOC (n=22) and benign tissues. (D) Word cloud of the TOP 100 EOC specific HLA class II ligand source proteins (uniprot recommended gene name). Font size (3-11) correlates with absolute number of cancer patients presenting HLA ligands of respective source proteins.

(3) FIG. 3 shows the cellular origin of the TOP100 EOC associated HLA class I ligands. Volcano plots of the relative abundance of HLA ligands in the class I immunopeptidome of enriched cell populations of OvCa 84 analyzed by label free quantitation. Panels show on the left side (A) tumor infiltrating leukocytes (CD45+) vs. tumor cells (CD45?Epcam+) and on the right side (B) stroma cells (CD45?EpCam?) vs. tumor cells. The horizontal dashed line indicates significance threshold (p<0.05). TOP100 EOC exclusive ligands (MUC16 (red), DDR1, EYA2, SOX9, TLR7, OASL) as well as ligands derived from leukocyte associated antigens (CD132, CD8, LSP1) and stroma (endothelial cell) associated antigens (vWF) are highlighted.

(4) FIG. 4 shows the immunohistochemical staining and serum levels as surrogate markers for ligand presentation. Immunohistochemical staining of high-grade serous ovarian carcinomas for MUC16 (CA-125) with low (IRS4), intermediate (IRS6) and high (IRS12) immunoreactivity score (A). Immunohistochemical staining for Mesothelin (right, IRS8) and IDO1 (left, IRS 12; all at 200? magnification) (B). Correlation of HLA ligand presentation and source protein expression of selected TOP100 EOC associated antigens. Expression of MUC16 (n=23), IDO1 (n=23) and MSLN (n=16) was analyzed by immunohistochemical staining (C) or serum marker analysis of CA-125 (n=30) at the day of surgery (D). For MSLN only the cases for which HLA class II immunopeptidome data were available were included. Non parametric Mann-Whitney test was employed to test for statistical significance (p<0.05 was considered significant).

(5) FIG. 5 shows the prognostic relevance of MUC16 and MSLN. Immunohistochemical stainings were performed on TMAs with 71 high-grade serous EOC samples from patients with documented optimal tumor debulking. (A) Kaplan Meier plot depicting the influence of MUC16 expression (left panel, low expression score<7, n=41; high expression score?7, n=30) and MSLN expression (right panel, low expression<6, n=15; high expression?6, n=52) on overall survival. (B) Impact of CD3 T-cell infiltration into the intraepithelial compartment (left panel CD3E, low infiltration<7 cell/HPF, n=13; high infiltration?7, n=57) or the fibrovascular stroma (right panel, CD3S, low infiltration<7 cell/HPF, n=40; high infiltration?7, n=30) on overall survival of patients. (C) Subgroup analysis of combined CD3 and MLN staining (all scoring cutoffs as above) for intraepithelial CD3 T-cells (top panel, low MSLN/high CD3E, n=1; low MSLN/low CD3E, n=40; high MSLN/low CD3E, n=14; high MSLN/high CD3E, n=1) or fibrovascular CD3 T-cells (bottom panel, low MSLN/high CD3S, n=30; high MSLN/low CD3S, n=7; low MSLN/low CD3S, n=21; high MSLN/high CD3S, n=8).

(6) FIG. 6 shows the flow cytometric analysis of EOC and benign ovarian tissue. Exemplary presentation of the gating strategy for OvCa 48 showing the selection of CD45+ leukocytes, CD45?CD31+ endothelial cells and CD45?EpCam+ tumor or epithelial cells.

(7) FIG. 7 shows the saturation analysis of HLA ligand source protein identifications for EOC. Saturation analysis for identifications of source proteins is depicted separately for HLA class I (A) and HLA class II (B) ligand proteins. The mean number of unique source proteins has been calculated for each source count by 1000 random samplings from the 34 EOC sources. Exponential regression was used to determine the calculated maximal attainable coverage of source protein accession (dotted lines) for EOC.

(8) FIG. 8 shows the frequency and number of HLA ligand presentation among EOC samples. HLA presentation of selected EOC associated antigens as well as the number of different HLA presented peptides (color coding) is visualized for each individual EOC (patient number on top of each column) both for class I (top) and class II (bottom) antigens.

EXAMPLES

(9) Materials and Methods

(10) Tissue Samples

(11) All tissue samples were collected at the University Hospital of T?bingen after obtaining patient informed consent in accordance with the principles of the Declaration of Helsinki. All study protocols were approved by the local institutional review board. If not stated otherwise samples were stored at ?80? C. until further usage. Two-digit HLA typing was performed by sequence specific primer (SSP) PCR using the HLA-Ready Gene System (Innotrain, Kronberg, Germany) and evaluated by SCORE Software (Olerup, Stockholm, Sweden) at the Department of Transfusion Medicine of the University Hospital of Tubingen. High resolution four-digit HLA typing was performed by next generation sequencing on a GS Junior Sequencer using the GS GType HLA Primer Sets (both Roche, Basel, Switzerland). Normal tissues were obtained from Bio-Options Inc, CA, USA; BioServe, Beltsville, Md., USA; Capital BioScience Inc, Rockville, Md., USA; Geneticist Inc., Glendale, Calif., USA; University Hospital of Geneva; University Hospital of Heidelberg; University Hospital Munich; ProteoGenex Inc., Culver City, Calif., USA; University Hospital of T?bingen. Written informed consents of all patients had been given before surgery or autopsy. Tissues were shock-frozen immediately after excision and stored until isolation of TUMAPs at ?70? C. or below.

(12) Tissue Dissociation

(13) EOC as well as benign ovary and fallopian tube tissues were freshly collected from patients undergoing tumor resection/debulking or salpingoophorectomy. Tissues were minced into small pieces <2 mm.sup.3 and transferred into an enzymatic dissociation solution containing 400 U/ml Collagenase Type IV, 5 U/ml Dispase (both life technologies, Carlsbad, Calif.) and 0.1 mg/ml DNAse (Roche, Basel, Switzerland) in DMEM (life technologies) with 10% fetal calf serum (Lonza, Basel, Switzerland). Dissociation was performed on a rotating shaker (Infors HT, Basel, Switzerland) for 3 hours at 37? C. Remaining tissue fragments (typically <1% of initial weight) were removed using a 100 ?m cell strainer (BD, Franklin Lakes, N.J.). Single cell suspensions were washed twice with PBS and erythrocytes were lysed using ammonium chloride lysis buffer.

(14) HLA Surface Molecule Quantification

(15) HLA surface expression was determined using QIFIKIT quantification flow cytometric assay (Dako, Glostrup, Denmark) according to manufacturer's instructions. Cells were stained with either pan-HLA class I specific monoclonal antibody W6/32, HLA-DR specific L243 or respective isotype control. Discrimination of cell types was based on surface marker staining with fluorescently labeled antibodies directed against CD45 (AmCyan clone 2D1, BD), CD31 (PeCy7, clone WM59, Biolegend, San Diego, Calif.), EpCam (APC, clone HEA125, Miltenyi, Bergisch-Gladbach, Germany) and CD34 (APCCy7,clone 581, Biolegend). 7-AAD (BioLegend) was added as viability marker immediately before analysis on a LSR SORP Fortessa instrument (BD). Triplicates were recorded for each sample with median fluorescence intensities used for calculation of surface molecule expression.

(16) Cell Separation:

(17) Cell separation was performed using two consecutive magnetic activated cell separation (MACS) protocols according to manufacturer's instructions (Miltenyi). Separations were performed using XS columns and a superMACS separator (both Miltenyi). The first separation aimed at positive selection of CD45.sup.+ leukocytes. The negative fraction was subsequently enriched for EpCa.sup.+ tumor cells. The remaining CD45?EpCam? fraction was assumed to represent the stroma cell fraction.

(18) HLA Ligand Isolation

(19) HLA class I and II molecules were isolated by standard immunoaffinity purification as described previously.sup.42. Pan-HLA class I specific mAb W6/32 was employed for HLA class I isolation and pan-HLA class II mAb T039 as well as HLA-DR specific mAb L243 were used for HLA class II isolation.

(20) Immunopeptidome Analysis by LC-MS/MS

(21) Immunopeptidome analysis was performed on an LTQ OrbitrapXL mass spectrometer (Thermo Fisher, Waltham, Mass.) equipped with a nanoelectron spray ion source and coupled to an Ultimate 3000 RSLC Nano UHPLC System (Dionex, Sunnyvale, Calif.). Peptide samples were loaded with 3% of solvent B (20% H.sub.2O, 80% acetonitrile and 0.04% formic acid) on a 2 cm PepMap 100 C18 Nanotrap column (Dionex) at a flow rate of 4 ?L/min for 10 min. Separation was performed on a 50 cm PepMap C18 column with a particle size of 2 ?m (Dionex) mounted in a column oven running at 50? C. The applied gradient ranged from 3 to 30% solvent B within 140 min at a flow rate of 175 nil/min. (Solvent A: 99% H.sub.2O, 1% ACN and 0.1% formic acid; Solvent B: 20% H2O, 80% ACN and 0.1% formic acid). Mass spectrometry analysis was performed in data dependent acquisition mode employing a top five method (i.e. during each survey scan the five most abundant precursor ions were selected for fragmentation). Survey scans were recorded in the Orbitrap at a resolution of 60,000. MS/MS analysis was performed by collision induced dissociation (CID, normalized collision energy 35%, activation time 30 ms, isolation width 1.3 m/z) with subsequent analysis in the linear trap quadrupole (LTQ). Mass range for HLA class I ligands was limited to 400-650 m/z with possible charge states 2+ and 3+ selected for fragmentation. For HLA class II mass range was set to 300-1500 m/z allowing for fragmentation with positive charge states ?2.

(22) HLA class I samples were analyzed in 5 technical replicates while for HLA class II samples 3 technical replicates were typically acquired. Initial runs were performed without dynamic exclusion, whereas for consecutive runs a dynamic exclusion of 5s was enabled.

(23) Mass Spectrometry Data Processing and Analysis

(24) MS data analysis was carried out using Proteome discoverer 1.3 (ThermoFisher). Peak lists were searched against the human proteome as comprised in the Swiss-Prot database (www(dot)uniprot(dot)org, released Sep. 27, 2013; including 20,279 reviewed protein sequences) using Mascot search engine (Mascot 2.2.04, Matrix Science, Boston, Mass.). Mass tolerance for processing was 5 ppm for precursor ions and 0.5 Da for fragment ions. No cleavage specificity was selected and the only dynamic modification allowed was oxidized methionine. Peptide confidence was determined using percolator algorithm with a target value of q 0.05 (5% FDR). Additional post processing filters were a Mascot Ionscore ?20, search engine rank=1 and peptide length of 8-12 amino acids for HLA class I ligands and 12-25 amino acids for HLA class II ligands. Protein grouping was disabled to ensure multiple annotations of peptides, if sequences map into multiple proteins due to conservation. HLA annotation was performed using HLA prediction algorithms hosted at SYFPEITHI (www(dot)syfpeithi(dot)de) and NETMHC 3.4 (http://www(dot)cbs(dot)dtu(dot)dklservices/NetMHC). In case of ambiguous results multiple alleles are mentioned. For comparative profiling one hit wonders i.e. peptides only presented on one source with a PSM count ?5 were removed from both of the datasets.

(25) Label free quantitation of peptides on tumor vs. CD45+ and tumor vs. stroma cells was performed using Sieve 2.1 (Thermo Fisher). At least 3 replicates of MS raw files for each cell enriched fraction as well as results from whole tissue MHC precipitations were aligned altogether with a maximum retention time (RT) shift of 2.5 mins. Frames were generated based on MS.sup.2 scan events with a maximum RT width of 3.5 mins and 5 ppm mass tolerance. Identifications were imported from Proteome discoverer using Mascot search results (see above). Total ion current chromatogram normalization was used to accommodate for differences in sample intensities.

(26) Immunogenicity Analysis of HLA Class I Ligands

(27) Priming of peptide specific cytotoxic lymphocytes (CTLs) was conducted using an established protocol involving artificial antigen presenting cells (aAPCs) (30). aAPCs consisted of streptavidin-coated polystyrene beads (5.6 ?m in diameter; Bangs Laboratories, Fishers, Ind.). Beads were resuspended at 2?10.sup.6 particles per ml and incubated with 10 nM biotinylated peptide-MHC complexes and 10 nM stimulating anti-CD28 antibody (clone 9.3 derived from ATCC, Manassas, Va.) each for 30 min at ambient temperature. T cells were isolated from whole blood of healthy donors using a CD8 magnetic cell isolation kit (Miltenyi). One million T-cells per well were cultured in 96 well plates (Corning, Corning, N.Y., USA) and stimulated with the same number of loaded aAPCs in the presence of 5 ng/ml IL-12 (PromoCell, Heidelberg, Germany). T cells were stimulated 3 times in total with weekly stimulation interval. 40 U/ml IL-2 was added 2 days subsequent to each stimulation. T-cell priming was assessed by MHC-multimer staining one week after the last stimulation round.

(28) Construction of Tissue Microarrays (TMA)

(29) Consecutive paraffin embedded tumor samples of patients with high-grade serous carcinoma of the ovary or fallopian tube (EOC) with at least FIGO stage II-III and operated at the University Women's Hospital in T?bingen between 1999 and 2008 were retrieved from the archives of the Institute of Pathology. After confirmation of histological subtype and grading according to published criteria (43). 154 cases were initially included in the study. A tissue microarray (TMA) was constructed as described previously (44). We used six cores of 0.6 mm diameter of each patient (maximum three cores each from two different sites of the primary tumorsat least two separate cores). In addition we constructed a TMA using paraffin embedded tissue from the primary tumors of the prospectively collected cases for ligandome analysis. 3 ?m thick sections were cut, rehydrated and subjected to specific pretreatment for immunohistochemistry. In total 23 cases were evaluable for immunoscoring and correlation with immunopeptidome data.

(30) Immunohistochemistry

(31) The following primary antibodies and dilutions were used for immunohistochemistry: CD3 (1:100, rat monoclonal SP7, DCS, Hamburg, Germany), CD8 (1:200, mouse monoclonal C8/144B, DAKO), MUC16 (1:450, mouse monoclonal M11, DAKO, Glostrup, Denmark), IDO1 (1:25, mouse monoclonal, ABCAM, Cambridge, UK) and MSLN (1:100, mouse monoclonal SPM143, GeneTex, Irvine, Calif., USA). The tissue sections were pre-treated with EDTA-buffer solution (pH 8.6) at 95? C. for 36 min. Immunohistochemical staining was performed on an automated immunostainer according to the manufacturer's instructions using the iView DAB detection kit (both Ventana, Tucson, Ariz., USA).

(32) Immunoscoring

(33) Quantification of TILs was carried out by first assessing the average number of immunostained cells per high power field (HPF=400?) by counting at least 2 HPF for each core. In a second step, the average number of lymphocytes per HPF for the left and right triple core set was calculated, and for all cores together. This bilateral average count was used for further calculations. The fibro vascular tumor stroma (CD3S and CD8S), and the intraepithelial compartment of the tumor (CD3E and CD8E) were evaluated separately.

(34) For expression of CA 125, IDO1 and MSLN staining intensity was graded from 0-3, multiplied by a score from 1-4 for the percentage of tumor cells (1: 0-10%; 2: 10-50%; 3: 50-80%; 4: 80-100%). For all parameters the cases were separated in quartiles and the best separation between two quartiles defined as cut-off value between high and low expression. Of the 154 cases on the TMA 71 patients had undergone documented optimal tumor debulking (<1 cm residual tumor mass) and could be successfully evaluated for TILs and expression of proteins. Immunoscoring and clinical data analysis were performed by independent investigators.

(35) Statistical Analysis/Visualization

(36) If not mentioned otherwise all figures and statistical analyses were generated using Graphpad Prism 6.0 (Graphpad software, La Jolla, Calif., USA) or Microsoft Office 2010 (Microsoft). Word clouds were created using an online applet (www(dot)wordle(dot)net). Kaplan-Meier analysis was performed using SPSS statistical software (Version 21, IBM Corp., Armonk, N.Y., USA). Two-tailed unpaired student's t-test was performed unless otherwise specified. P values less than 0.05 were considered statistically significant. D'Agostino-Pearson omnibus test was used to verify normality and the F-Test was used to verify equal variance. For FIG. 1 the two-tailed unpaired Student's t-test with Welch's correction was used owing to unequal variance between the two comparison groups. Non-parametric Mann-Whitney-test was used in FIG. 4 because normal distribution could not be assessed in all cases due to small sample sizes. Spearman correlation was used to correlate IHC scores of MSLN and MUC16 as the datasets were not showing normal distribution. P values comparing two Kaplan-Meier survival curves in FIG. 5 were calculated using the log-rank (Mantel-Cox) test in Graphpad Prism.

Example 1: HLA Count on Cell Surface and HLA Typing

(37) A major prerequisite for the development of T-cell mediated immunotherapies is the expression of MHC molecules on the surface of tumor cells. Therefore, the inventors analyzed and quantified the number of HLA-A, B, C as well as HLA-DR molecules by flow cytometry on different cell subsets of ovarian tumors (n=11) as well as benign tissues from ovary and fallopian tube (n=8) obtained by enzymatic dissociation. The analysis aimed at the separate quantification of cell type specific HLA expression for leukocytes (CD45.sup.+), tumor/epithelial cells (Epcam.sup.+), and endothelial cells (CD31.sup.+; the latter only in a subset of 7 ovarian tumors). For the complete gating strategy see FIG. 6. The median number of HLA molecules per cell was heterogeneous both among different cell types and individual patients, ranging from ?5,000 to 150,000 HLA class I and ?500 to 330,000 HLA-DR molecules. The number of HLA-A, B, and C molecules was significantly higher (p=0.0205) on leukocytes isolated from tumor vs. benign tissue indicating an ongoing inflammatory reaction within the tumor. Strong differences in HLA class I expression were also seen when comparing tumor cells with epithelial cells derived from benign tissues. HLA class I molecule expression was significantly (p=0.0021) higher on tumor cells (?75,000 molecules/cell) but remained in the range of other stromal cells such as endothelial cells (?95,000 molecules/cell). Surprisingly the inventors evidenced a strong (?105,000 molecules/cell) to some extent extraordinarily high expression of HLA-DR on EOC cells (>300,000 molecules/cell), whereas benign epithelial cells were virtually negative for HLA-DR (p=0.0108). Altogether, the inventors could observe an increased MHC class I and class II expression within the tumors.

(38) HLA ligandome analysis and comparative profiling reveal EOC specific antigen presentation. In order to map the HLA ligand repertoire of EOC the inventors isolated HLA molecules from bulk tumor tissue and performed mass spectrometry to characterize the HLA ligandome for a total of 34 EOCs (for patient characteristics and HLA typing see Table 7).

(39) TABLE-US-00007 TABLE 7 OvCa Tumor TNM HLA typing ID Age Type Staging MHC class I HLA typing MHC class II OvCa 9 65 serous T3cNxM1G2R1 A*02:01, DQB1*03:01, DQA1*03:01, ovarian A*03:01, DQA1*05:01, DRB1*11:01, carcinoma B*07:02, DRB1*04:01, DRB3*02:02, B*40:02, DRB4*01:01, DPB102:01, C*07:02, DPB1*13:01 C*12:01 OvCa 60 serous T3bN1M1G2R1 A*02:01, DQB1*02:02, DQB1*05:01, 10 ovarian A*11:01, DQA1*01:01, DQA1*03:01, carcinoma B*44:05, DRB1*01:01, DRB1*09:01, B*51:01, DRB4*01:01, DPB1*04:01, C*02:02, DPB1*05:01 C*15:02 OvCa 62 serous T3cN0G2R0 A*24:02, DQB1*03:01, DQB1*05:04, 12 ovarian A*31:01, DQA1*01:02, DQA1*03:01, carcinoma B*35:03, DRB1*01:01, DRB1*04:01, B*49:01, DRB4*01:01, DPB1*02:01, C*07:01, DPB1*05:01 C*12:03 OvCa 62 serous T1cN1G3R0 A*02, B*35, DQB1*04, DQB1*06, 13 ovarian B*40, C*03, DRB1*08, DRB1*13 carcinoma C*04 OvCa 75 serous T3cN0G3R0 A*11:01, DQB1*03:01, DQA1*05:01, 15 ovarian A*24:02, DRB1*11:01, DRB1*03:17, carcinoma B*07:02, DRB3*02:02, DPB1*03:01 B*55:01, C*03:03, C*07:02 OvCa 45 serous T3bN1G3R0 A*02, B*40, DQB1*06, DRB1*08, 16 ovarian B*44, C*03, DRB1*13, DRB1*14, DRB3 carcinoma C*05 OvCa 29 serous T3aN1G3R0 A*01, A*03, DQB1*02, DQB1*03, 23 ovarian B*08, B*35, DRB1*03, DRB1*12, DRB3 carcinoma C*04, C*07 OvCa 66 serous T2bN0G3R0 A*01:01, DQB1*05:01, DQB1*06:01, 28 ovarian A*02:01, DQA1*01:01, DQA1*03:01 carcinoma B*27:05, DRB1*01:03, DRB1*15:02, B*52:01, DRB5*01:02, DPB1*04:01 C*01:02, C*02:02 OvCa 45 serous T3cN1G3R1 A*25:01, DQB1*06:02, DQA1*01:02, 39 ovarian A*31:01, DRB1*15:01, DRB1*16:09, carcinoma B*07:02, DRB5*01:01, DRB5*01:11, B*18:01, DPB1*04:01, DPB1*04:02 C*12:03, C*07:02 OvCa 66 serous T3cN0G3R1 A*02, A*24, DQB1*03, DQ7, DRB1*11, 41 and B*18, B*51, DRB3 endo C*02, C*12 metrial ovarian carcinoma OvCa 61 serous T3cN1G3R2 A*02, A*32, DQB1*03, DQB1*05, DQ9, 43 ovarian B*18, B*35, DRB1*01, DRB1*07, DRB4 carcinoma C*04, C*07 OvCa 63 mixed T1cN0G3R0 A*01, A*23, DQB1*02, DRB1*03, 45 differentiated B*08, B*44, DRB1*07, DRB3, DRB4 (mostly C*04, C*07 endo metroid) ovarian carcinoma OvCa 71 serous T3cN1G3R0 A*02:01, DQB1*03:02, DQB1*03:04, 48 ovarian A*25:01, DQA1*03:01, DRB1*04:01, carcinoma B*15:01, DRB1*13:03, DRB3*01:01, B*41:02, DRB4*01:01, DPB1*02:01 C*03:04, C*17:01 OvCa 48 serous T3bN1G3R0 A*02, A*03, DQB1*02, DQB1*03, DQ7, 53 ovarian B*27, B*35, DRB1*03, DRB1*11, DRB3 carcinoma C*02, C*04 OvCa 66 serous T3cN1M1G3R2 A*02:01, DQB1*05:01, DQB1*05:03, 54 ovarian A*11:01, DQA1*01:01, DRB1*01:03, carcinoma B*35:01, DRB1*14:01, DRB3*02:02, B*35:03, DPB1*04:01, DPB1*02:01 C*04:01, C*12:03 OvCa 58 endo T1cN0G1R0 A*25, A*32, DQB1*05, DQB1*06, 57 metrioid B*15, B*18, DRB1*01, DRB1*15, DRB5 ovarian C*03, C*12 carcinoma OvCa 74 serous T3cN1G3R1 A*02, A*03, DQB1*05, DRB1*01 58 ovarian B*35, C*03, carcinoma C*04 OvCa 47 serous T3cN1G3R2 A*03, A*30, DQB1*02, DRB1*07, DRB4 59 ovarian B*13, C*06 carcinoma OvCa 50 serous T3cN1G3R1 A*24:02, DRB1*08:01, DRB1*13:01, 60 ovarian A*25:01, DQB1*04:02, DQB1*06:03, carcinoma B*13:02, DQA1*04:01, DQA1*01:03, B*18:01, DPB1*02:01, DPB1*03:01 C*12:03, C*06:02 OvCa 56 serous T3cN1G3R1 A*01, A*25, DQB1*02, DRB1*03, DRB3 64 ovarian B*08, C*07 carcinoma OvCa 55 serous T3cN1M1G3R1 A*01, A*24, DQB1*03, DQB1*05, 65 ovarian B*15, B*35, DRB1*10, DRB1*11, DRB3 carcinoma C*04, C*14 OvCa 73 serous T2bN0G3R0 A*11:01, DRB1*03, DRB*0701, 66 ovarian A*29:02, DRB3*0202, DRB4*0101, carcinoma B*18:01, DQB1*02:01, DQB1*02:02, B*44:03, DQA1*02:01, DQA1*05:01, C*05:01, DPB1*02:02, DPB1*03:01 C*16:01 OvCa 69 serous T3cN1G3R1 A*02:01, DRB1*10:01, DRB1*04:01, 68 ovarian A*01:01, DRB4*04:01, DQB1*05:01, carcinoma B*44:02, DQB1*03:01, DQA1*01:01, B*37:01, DPB1*04:01 C*06:02, C*05:01 OvCa 68 serous T3cN0G1R1 n/a n/a 69 ovarian carcinoma OvCa 48 serous T3cN1M1G1R1 A*01, A*02, DQB1*03, DQB1*05, 70 ovarian B*07, C*07 DRB1*09, DRB1*14, DRB3, carcinoma DRB4 OvCa 53 serous T3bN1G3R0 A*03:01, DRB1*01:01, DRB1*03:01, 72 ovarian A*01:01, DRB3*01:01, DQB1*05:01, carcinoma B*08:01, DQB1*02:01, DQA1*01:01, B*07:02, DPB1*04:01 C*07:02, C*07:01 OvCa 69 serous T3cN1G3R0 A*01:01, DRB1*03:01, DRB1*03:42, 73 ovarian B*08:01, DRB3*01:01, DRB3*01:14, carcinoma C*07:01 DQB1*02:01, DQA1*05:01, DPB1*04:01 OvCa 79 endo T3bNxG1R1 A*02:01, DRB1*11:04, DRB1*07:01, 74 metrioid B*18:01, DRB3*02:02, DRB4*01:01, ovarian B*51:01, DQB1*03:01, DQB1*02:02, carcinoma C*07:02, DQA1*02:01, DQA1*05:01, C*15:02 DPB1*04:02, DPB1*02:01 OvCa 57 endo T2bN0G2R0 A*01:01, DQB1*03:03, DQA1*02:01, 79 metrioid A*31:01, DRB1*07:01, DRB1*09:01, ovarian B*08:01, DRB4*01:01, DPB1*13:01, carcinoma B*51:01, DPB1*02:01 C*07:01, C*15:02 OvCa 93 serous T3cNxG3R2 A*25:01, DRB1*01:01, DRB1*12:01, 80 ovarian A*32:01, DRB3*02:02, DQB1*03:01, carcinoma B*18:01, DQB1*05:01, DQA1*01:01, B*39:01, DQA1*05:01, DPB1*04:01 C*12:03 OvCa 78 serous T3cNxG3R2 A*02:01, DRB1*04:02, DRBB1*11:01, 81 ovarian B*45:01, DRB4*01:01, DRB3*02:02, carcinoma B*56:01, DQB1*03:01, DQB1*03:02 C*07:02, C*01:02 OvCa 48 serous T3cN1G3R0 A*01:01, DRB1*04:02, DRB1*03:01, 82 ovarian A*03:01, DRB4*01:01, DRB3*01:01, carcinoma B*08:01, DQB1*02:01, DQB1*03:02, B*38:01, DQA1*03:01, DQA1*05:01, C*07:01, DPB1*04:01, DPB1*13:01 C*12:03 OvCa 50 serous T1cN0G2R0 A*02, A*11, DQB1*03, DQB1*05, 83 ovarian B*51, B*55, DRB1*09, DRB1*14, DRB3, carcinoma C*03, C*15 DRB4 OvCa 70 serous T3cN1G3R1 A*02:01, DRB1*15:01, DRB5*01:01, 84 ovarian B*07:02, DQB1*06:02, DQA1*01:02, carcinoma B*44:02, DPB1*04:01, DPB1*04:02 C*07:02, C*05:01

(40) For MHC class I the inventors could identify 22,920 unique peptides (mean 1,263/sample) emanating from 9,136 different source proteins (mean 1,239/sample) reaching >90% of the estimated maximal attainable coverage (see FIG. 7a).

Example 2, Identification of Top Cancer Associated HLA Ligands

(41) Aiming to extract the most specific HLA ligands for EOC from this vast catalogue of data the inventors compared the HLA ligand source proteins with an in-house database of benign sources (HLA benign ligandome database) consisting of samples from PBMCs (n=30), bone marrow (n=10), liver (n=15), colon (n=12), ovary (n=4) and kidney (n=16).

(42) The HLA benign ligandome database contains 31,032 peptides representing 10,012 source proteins and was established using blood or bone marrow from healthy donors as well as histopathologically evaluated normal tissues, all analyzed with exactly the same pipeline as used for EOCs. For comparative profiling one hit wonders (i.e. peptides only presented on one source with low PSM count) were removed from both datasets to accommodate for false positive hits. Comparative analysis of the two respective datasets (see FIG. 2A) revealed 379 MHC class I source proteins to be presented exclusively by EOC in at least three of the tested patients, highlighting an EOC specific HLA peptide repertoire. The TOP100 EOC specific source proteins ranked according to their frequency of presentation are visualized in FIG. 2B. The most important EOC specific HLA ligand source protein yielded by this analysis was mucin 16 (MUC16) also known as cancer antigen 125 (CA-125). Overall more than 80 different MUC16 derived HLA ligands (see Table 8) were presented in nearly 80% of patients (26/34).

(43) TABLE-US-00008 TABLE8 Sequence IDNo. Sources HLA AHSKITTAM 3 OvCa80 B*39:01 AVKTETSTSER 4 OvCa12,OvCa79 A*31:01 AVTNVRTSI 5 Ovca59,OvCa60 B*13 DALTPLVTI 6 OvCa74 B*51:01 DALVLKTV 7 OvCa41,OvCa74, B*51 OvCa79,OvCa83 DPYKATSAV 8 OvCa10,OvCa41, B*51 OvCa69OvCa74, OvCa79,OvCa83 EPETTTSFITY 9 OvCa65 B*35 ERSPVIQTL 10 OvCa80 B*39:01 ETILTFHAF 11 OvCa48,OvCa64,OvCa80 A*25 EVISSRGTSM 12 OvCa48,OvCa60,OvCa64,OvCa 80 A*25 EVITSSRTTI 13 OvCa60,Ovca64 A*25 EVTSSGRTSI 14 OvCa60,Ovca64,OvCa80 A*25 FPEKTTHSF 15 OvCa65 B*35 FPHSEETTTM 16 OvCa13,OvCa65 B*35 FPHSEITTL 17 OvCa12,OvCa13,OvCa53 B*35 FQRQGQTAL 18 OvCa48 B*15:01 GDVPRPSSL 19 OvCa72 B*08:01 GHESHSPAL 20 OvCa80 B*39:01 GHTTVSTSM 21 OvCa80 B*39:01 GTHSPVTQR 22 OvCa39,OvCa79 A*31:01 GTSGTPVSK 23 OvCa83 A*11 HPDPQSPGL 24 OvCa65 B*35 IITEVITRL 547 OvCa83 A*02 IPRVFTSSI 25 OvCa41,OvCa74 B*51 ISDEVVTRL 26 OvCa16 C*05 ISIGTIPRI 27 OvCa65 B*15:17 ISKEDVTSI 28 OvCa65 B*15:17 ITETSAVLY 29 OvCa65 A*01 ITRLPTSSI 30 OvCa65 B*15:17 KDTAHTEAM 31 OvCa68 B*44:02 KEDSTALVM 32 OvCa16 B*40/B*44 KEVTSSSSVL 33 OvCa16,OvCa70 B*40/B*44/? KMISAIPTL 548 OvCa81,OvCa83 A*02 LPHSEITTL 34 OvCa12,OvCa13 B*35 LTISTHKTI 35 OvCa65 B*15:17 LTKSEERTI 36 OvCa65 B*15:17 QFITSTNTF 1 OvCa60 A*24:02 RDSLYVNGF 37 OvCa68 B*44:02 RETSTSQKI 38 OvCa60 B*18:01 RSSGVTFSR 39 OvCa79 A*31:01 SAFESHSTV 40 OvCa41,OvCa74,OvCa79,OvCa83 B*51 SATERSASL 41 OvCa13,OvCa16,OvCa70 C*03/?+0 SENSETTAL 42 OvCa16,OvCa70 B*40/B*44/? SEQRTSPSL 43 OvCa70 n.a. SESPSTIKL 44 OvCa13,OvCa70 B*40/?+0 SPAGEAHSL 45 OvCa72,OvCa81,OvCa84 B*07/B*56 SPAGEAHSLLA 46 OvCa81 B*56:01 SPHPVSTTF 47 OvCa84 B*07:02 SPHPVTALL 48 OvCa9,OvCa72,OvCa84 B*07:02 SPLFQRSSL 49 Ovca72 B*0702 SPQNLRNTL 50 OvCa23,OvCa72,OvCa84 B*35/B*07:02 SPRLNTQGNTAL 51 OvCa72,Ovca84 B*07:02 SPSEAITRL 52 Ovca84 B*07:02 SPSKAFASL 53 OvCa9,OvCa23,OvCa39,OvCa B*35/B*07:02 69,OvCa72,OvCa84 SPSSPTPKV 54 OvCa72 B*07:02 SPSSQAPVL 55 OvCa84 B*07:02 SQGFSHSQM 56 OvCa48 B*15:01 SRTEVISSR 57 OvCa53 B*27 SSAVSTTTI 58 OvCa65 B*15:17 SSPLRVTSL 59 OvCa69 n.a. STASSSLSK 60 OvCa83 A*11 STETSTVLY 2 OvCa64,OvCa65,OvCa68 A*01 STQRVTTSM 61 OvCa72 n.a. STSQEIHSATK 62 OvCa83 A*11 SVLADLVTTK 63 OvCa72 A*03:01 SVPDILSTSW 64 OvCa60 A*24:02 TAGPTTHQF 65 OvCa58 C*03 TEISSSRTSI 66 OvCa12 B*49:01 TENTGKEKL 67 OvCa16 B*40/B*44 TETEAIHVF 68 OvCa41,OvCa80 B*18 TEVSRTEVI 69 OvCa12 B*49:01 TExVLQGLL 70 OvCa16,OvCa66,OvCa70 B*40/B*44/? TPGGTRQSL 71 OvCa9,OvCa23,OvCa39,OvCa B*07:02/B*35 72,OvCa84 TPGNRAISL 72 OvCa23,OvCa72,OvCa84 B*07:02/B*35 TPNSRGETSL 73 OvCa72 B*07:02 TSGPVTEKY 74 OvCa58 B*35 TSPAGEAHSL 75 OvCa81 n.a. TTLPESRPS 324 OvCa70 n.a. TYSEKTTLF 549 OvCa12,OvCa41,OvCa60,OvCa65 A*24 VHESHSSVL 76 OvCa80 B*39:01 VPRSAATTL 77 OvCa23,OvCa72,OvCa84 B*07:02/B*35 VTSAPGRSI 78 OvCa65 B*15:17 VTSSSRTSI 79 OvCa65 B*15:17 YPDPSKASSAM 80 OvCa65 B*35
Those data highlight the frequent processing and presentation of MUC16 by a multitude of different HLA allotypes unparalleled by any other EOC specific antigen and mirrored only by frequently (>95%) presented house-keeping proteins such as beta actin (overall 149 different peptides identified). Among the TOP100 EOC specific source proteins other well established tumor associated antigens like MUC1 or KLK10 as well as antigens with well documented immune-evasive functions like Indoleamine-2,3-dioxygenase (IDO1) or Galectin 1 (LGALS1) were identified.

(44) Owing to the power of CD4 T cells in supporting or driving an anti-tumor immune response the inventors used the same approach to further analyze MHC class II presented peptides in EOC (n=22) yielding 9,162 peptides (mean 598/sample) representing 2,330 source protein (mean 319/sample) reaching >80% of attainable coverage (see FIG. 7B). The HLA benign ligand dataset for MHC class II contained 7,267 peptides representing 1,719 source proteins derived from bone marrow (n=5), PBMCs (n=13), colon (n=2), liver (n=7) and kidney (n=17). Analysis of the TOP100 MHC class II presented antigens revealed a more heterogeneous and complex picture (FIG. 2C). Notably, MHC presented peptides of mesothelin (MSLN) an established ligand of MUC16, could be identified in nearly 50% of patients (10/22; FIG. 2D). MUC16 itself was not among the TOP100 class II antigens but respective ligands could nevertheless be detected in four patients.

(45) Besides the TOP100 EOC specific HLA ligand source proteins, the inventors further looked for established cancer-testis and tumor associated antigens that have been previously employed for clinical application to verify their abundance (Her2neu, WT1, NY-ESO-1, hTert and p53). Although the inventors could identify HLA presented peptides for all antigens except for NY-ESO-1, none of them were exclusively presented on EOC (Table 9). The only ligands showing EOC specific presentation, albeit with low frequency (3/34), were HLA class I ligands (but not HLA class II) from Her2neu.

(46) TABLE-US-00009 TABLE9 HLA Sourcesof SEQID Her2neu restriction presentation ERBB2(Receptortyrosine-protein kinaseerbB-2) 554 TYLPTNASLSF A*23/A*24 2xOvCa 153 MPNPEGRYTF B*35 1xOvCa 152 AARPAGATL B*07 1xOvCa 291 AIKVLRENTSPKANKE HLA class111xOvCa 292 DPSPLQRYSEDPTVPLPS HLA classII2xOvCa 293 DPSPLQRYSEDPTVPLPSE HLA classII1xOvCa 294 ELVSEFSRMARD HLA class112xPBMCs 2xPBMCs,1x 295 ELVSEFSRMARDPQ HLA class11Kidney 296 IPVAIKVLRENTSPKANKE HLA class111xOvCa 297 RRLLQETELVEPLTPS HLA class112xLiver 298 SPQPEYVNQPDVRPQPP HLA classII1xOvCa 291 VKPDLSYMPIWKFPDE HLA class111xOvCa WT-1 Wilmstumorprotein 558 RMFPNAPYL A*02 8xPBMCs,1x Liver 557 QRNMTKLQL B*13 2xOvCa,1x Liver,1x PBMCs 555 GVFRGIQDV B*13 2xOvCa 550 ALLPAVPSL A*02 1xOvCa hTert Telomerasereversetranscriptase 556 LMSVYVVEL A*02 2xPBMCs p53 Cellulartumorantigenp53 552 RPILTIITL B*07 4xPBMCs,2x Liver,2x Kidney,3x OvCa 553 TYSPALNKMF A*24 1xPBMCs,1x Liver,2xOvCa 551 GRNSFEVRV B*27 1xPBMC,1x Liver,1x Kidney,1x OvCa

Example 3: Cellular Origin of EOC Associated HLA Presented Peptides

(47) Since EOCs embody not only cancer cells but rather represent a heterogeneous mixture of different cell types the inventors asked, whether the MHC class I TOP100 antigens were indeed originally presented by cancer cells. For this purpose the inventors digested EOCs and separated CD45.sup.+ leukocytes, EpCam.sup.+ tumor cells as well as stroma cells negative for the two markers (for enrichment efficiencies see Table 10) and subsequently the inventors performed HLA ligandomics individually for each of the subsets.

(48) TABLE-US-00010 TABLE 10 Cell enrichment efficiencies: Percentage of cells are given in each fraction before (PreSort) and after MACSorting PreSort CD45.sup.+ fraction EpCam.sup.+ fraction EpCam.sup.? fraction OVCa CD45.sup.+ EpCam.sup.+ Viability CD45.sup.+ EpCam.sup.+ Viability CD45.sup.+ EpCam.sup.+ Viability CD45.sup.+ EpCam.sup.+ Viability 84 74.7 18.3 80.2 93.5 6.2 71.6 10.7 85.7 88.2 4.5 22.1 64.0 73 23.1 12.3 81.2 95.7 1.7 77.2 3.4 73.3 87.6 1.7 3.2 87.4 70 76.2 8.83 78.9 96 1.3 82.7 3.4 94 66.4 3.1 4.5 65.4 60 77.4 5.2 92.3 94.8 1.7 90.2 5.2 79.7 88.7 3.8 10.7 89.5 57 31.9 50.5 94.1 93.6 5.0 90.6 1.4 95.3 96.7 0.8 7.2 95.3

(49) The inventors used label free quantification to determine the source of each identified HLA ligand in a total of 5 EOCs (for a representative example see FIG. 3). As expected, MUC16 derived HLA ligands, identified on (4/5) EOC samples, were always found to be overrepresented on enriched cancer cells with a median 5 fold overrepresentation (range 1.8-135 fold) dependent on the efficiency of the enrichment. The same held true for several other frequently presented TOP100 antigens like DDR1, SOX9, CRABP1/2, EYA2, LAMC2, MUC1 or KLK10. However a number of other antigens especially those known to be upregulated by interferon such as toll like receptors (TLR3, TLR7) or 2-5-oligoadenylate synthase-like protein synthase (OASL) could not be unambiguously shown to be presented by tumor cells but rather displayed strong overrepresentation on CD45+ leukocytes and/or stroma cells. Apart from tumor associated antigens the inventors also recognized ligands from source proteins with cell type specific expression. For example ligands derived from CD8, CD132 or lymphocyte specific protein 1 (LSP1) were found highly overrepresented on CD45+ cells and van Willebrand factor (vWF) most likely expressed by endothelial cells in the stroma was found highly overrepresented within the stromal subset emphasizing the strength of this cell type specific approach.

Example 4: Immunogenicity Analysis of MUC16 Derived Ligands

(50) For the applicability of peptide vaccines immunogenicity is a major imperative. In order to evaluate the immunogenic potential of the identified HLA ligands the inventors used a T-cell priming protocol involving artificial antigen presenting cells and T cells isolated from blood of healthy donors. The results of this analysis for the number one EOC associated antigen MUC16 are presented in Table 11. Among 23 different peptides tested so far, 18 were shown to be immunogenic in at least 1/3 donors. This nearly 80% recognition rate verifies the presence of na?ve MUC16 recognizing T cells in the human population. Similar results have been obtained for other TOP100 antigens (e.g. IDO1, LGALS1).

(51) TABLE-US-00011 TABLE11 ImmunogenicityanalysisofEOCpresented HLAligandsfromMUC16/CA-125 SEQ positive/ HLA Sequence ID testeddonors A*01 STETSTVLY 2 0/2 A*02 IITEVITRL 547 3/10 A*02 KMISAIPTL 548 4/6 A*03 SVLADLVTTK 63 0/1 A*11 STSQEIHSATK 62 2/6 A*11 GTSGTPVSK 23 0/5 A*24 TYSEKTLLF 549 2/2 A*24 AVTNVRTSI 5 1/3 A*25 ETILTFHAF 11 2/2 A*25 EVITSSRTTI 13 1/1 A*25 EVTSSGRTSI 14 2/3 A*25 EVISSRGTSM 12 1/3 B*07 SPHPVTALL 48 0/1 B*07 SPQNLRNTL 50 1/1 B*07 LPHSEITTL 34 0/2 B*07 SPSKAFASL 53 2/2 B*07 VPRSAATTL 77 1/2 B*07 TPGNRAISL 72 2/2 B*15 SQGFSHSQM 56 4/5 B*15 FQRQGQTAL 18 1/6 B*27 ERSPVIQTL 10 1/2 B*51 DALVLKTV 7 1/3 B*51 DPYKATSAV 8 3/3 8/10allotypes 18/23HLAligands 34/73

Example 5: Biomarkers for HLA Ligand Presentation

(52) Antigen specific cancer immunotherapy (e.g. peptide vaccination, adoptive T-cell transfer) requires a stringent selection of candidate antigens within a short timeframe. HLA ligandome analysis however, is not always possible due to the lack of appropriate material. A feasible alternative would be the use of biomarkers to predict the presence of HLA ligands on the tumor cells. In order to evaluate whether, protein expression analyzed by immunohistochemistry (immunoreactivity score, IRS) could serve as a surrogate marker for HLA ligand presentation, the inventors analyzed the TOP100 MHC class I antigens MUC16 and IDO1 as well as the TOP100 MHC class II antigen MSLN by immunohistochemistry and correlated the staining intensity (FIG. 4A) to the presence or absence of HLA ligands on the same tumors. For both MUC16 and MSLN, staining scores were significantly higher on tumors, which presented HLA ligands of respective source proteins (FIG. 4C). The same was true for CA-125 serum levels determined at the day of surgery (FIG. 4D), indicating that these parameters could be used for a proper selection of candidate antigens for peptide vaccination. In contrast, IDO1 did not show a significant association with ligand presentation.

Example 6: Prognostic Relevance of the MUC16/MSLN Axis

(53) Because of their importance as targets for immunotherapy the inventors wanted to assess whether MSLN and MUC16 are also of prognostic relevance in a patients similar to our immunopeptidome collective. For this purpose the inventors analyzed the expression of both antigens as well as the extent of T-cell infiltration by immunohistochemistry in a tissue microarray (TMA) of high grade serous ovarian cancers (FIGO stage II-111). In order to avoid prognostically relevant confounders the inventors restricted our analysis to 71 patients with optimally debulked cancers (residual mass below<1 cm).

(54) While the inventors did not observe any prognostic effect for MUC16 staining, strong MSLN staining was associated with a notable borderline significant (p=0.0572) decrease of median overall survival from 50 to 28 months (FIG. 5A). Despite their different prognostic relevance, staining scores for MUC16 and MSLN showed a direct and highly significant correlation (Spearman correlation coefficient r=0.5237; 95% c.i.=0.3159-0.6835, two tailed significance p<0.001).

(55) For the evaluation of T-cell infiltration the inventors assessed the number of CD3 T cells in the intraepithelial compartment of the tumor (CD3E) and the fibrovascular stroma (CD3S) separately. Notably only the number of intraepithelial T cells showed a significant (p<0.0063) prognostic impact, whereas infiltration of the surrounding stroma alone had no prognostic relevance (FIG. 5B). Only in a subgroup analysis combining MSLN and CD3 staining a significant prognostic benefit for tumors with low MSLN and high T-cell infiltration could be observed (FIG. 5C) for both CD3E (p<0.001) and CD3S (p<0.0049). Most strikingly, the combination of high intratumoral T-cell infiltration (CD3E) and low MSLN staining defined a subset of long term cancer survivors (10/11 patients with confirmed survival beyond 3 years).

REFERENCES AS CITED

(56) Allison, J. P. et al., Science 270 (1995) Andersen, R. S. et al., Nat. Protoc. 7 (2012) Appay, V. et al., Eur. J Immunol. 36 (2006) Banchereau, J. et al., Cell 106 (2001) Beatty, G. et al., J Immunol 166 (2001) Beggs, J. D., Nature 275 (1978) Benjamini, Y. et al., Journal of the Royal Statistical Society. Series B (Methodological), Vol. 57 (1995) Boulter, J. M. et al., Protein Eng 16 (2003) Braumuller, H. et al., Nature (2013) Brossart, P. et al., Blood 90 (1997) Bruckdorfer, T. et al., Curr. Pharm. Biotechnol. 5 (2004) Card, K. F. et al., Cancer Immunol. Immunother. 53 (2004) Chanock, S. J. et al., Hum. Immunol. 65 (2004) Cohen, C. J. et al., J Mol. Recognit. 16 (2003a) Cohen, C. J. et al., J Immunol. 170 (2003b) Cohen, S. N. et al., Proc. Natl. Acad. Sci. U.S.A 69 (1972) Coligan J E et al., (1995) Colombetti, S. et al., J Immunol. 176 (2006) Dengjel, J. et al., Clin Cancer Res 12 (2006) Denkberg, G. et al., J Immunol. 171 (2003) Falk, K. et al., Nature 351 (1991) Fong, L. et al., Proc. Natl. Acad. Sci. U.S.A 98 (2001) Gabrilovich, D. I. et al., Nat. Med 2 (1996) Gattinoni, L. et al., Nat. Rev. Immunol. 6 (2006) Gnjatic, S. et al., Proc Natl. Acad. Sci. U.S.A 100 (2003) Godkin, A. et al., Int. Immunol 9 (1997) Green M R et al., 4th, (2012) Greenfield E A, 2nd, (2014) Hwang, M. L. et al., J Immunol. 179 (2007) Jung, G. et al., Proc Natl Acad Sci USA 84 (1987) Kibbe A H, rd, (2000) Krieg, A. M., Nat. Rev. Drug Discov. 5 (2006) Liddy, N. et al., Nat. Med. 18 (2012) Ljunggren, H. G. et al., J Exp. Med 162 (1985) Longenecker, B. M. et al., Ann N.Y. Acad. Sci. 690 (1993) Lukas, T. J. et al., Proc. Natl. Acad. Sci. U.S.A 78 (1981) Lundblad R L, 3rd, (2004) Meziere, C. et al., J Immunol 159 (1997) Morgan, R. A. et al., Science 314 (2006) Mori, M. et al., Transplantation 64 (1997) Mortara, L. et al., Clin Cancer Res. 12 (2006) Mueller, L. N. et al., J Proteome. Res. 7 (2008) Mueller, L. N. et al., Proteomics. 7 (2007) Mumberg, D. et al., Proc. Natl. Acad. Sci. U.S.A 96 (1999) Pinheiro J et al., (2015) Plebanski, M. et al., Eur. J Immunol 25 (1995) Porta, C. et al., Virology 202 (1994) Rammensee, H. G. et al., Immunogenetics 50 (1999) Rini, B. I. et al., Cancer 107 (2006) Rock, K. L. et al., Science 249 (1990) Rodenko, B. et al., Nat. Protoc. 1 (2006) Saiki, R. K. et al., Science 239 (1988) Seeger, F. H. et al., Immunogenetics 49 (1999) Sherman F et al., (1986) Singh-Jasuja, H. et al., Cancer Immunol. Immunother. 53 (2004) Small, E. J. et al., J Clin Oncol. 24 (2006) Sturm, M. et al., BMC. Bioinformatics. 9 (2008) Teufel, R. et al., Cell Mol. Life Sci. 62 (2005) Tran, E. et al., Science 344 (2014) Walter, S. et al., J. Immunol. 171 (2003) Walter, S. et al., Nat Med. 18 (2012) Willcox, B. E. et al., Protein Sci. 8 (1999) Zaremba, S. et al., Cancer Res. 57 (1997) Siegel, R., Ma, J., Zou, Z. & Jemal, CA Cancer J. Clin. 64, 9-29 (2014). Coleman, R. L. et al Nat. Rev. Clin. Oncol. 10, 211-224 (2013). Herzog, T. J. & Pothuri, B. Nat. Clin. Pract. Oncol. 3, 604-611 (2006). Kandalaft, L. E., Powell, D. J., Jr., Singh, N. & Coukos, G. J. Clin. Oncol. 29, 925-933 (2011). Zhang, L., et al. N. Engl. J. Med. 348, 203-213 (2003). Schlienger, K., et al. Clin. Cancer Res. 9, 1517-1527 (2003). Matsuzaki, J., et al. Proc. Natl. Acad. Sci. U.S.A 107, 7875-7880 (2010). Fisk, B., Blevins, T. L., Wharton, J. T. & Ioannides, C. G. J. Exp. Med. 181, 2109-2117 (1995). Curiel, T. J., et al. Nat. Med. 10, 942-949 (2004). Vlad, A. M., et al. Cancer Immunol. Immunother. 59, 293-301 (2010). Hodi, F. S., et al. Proc. Natl. Acad. Sci. U.S.A 105, 3005-3010 (2008). Robert, C., et al. Lancet 384, 1109-1117 (2014). Wolchok, J. D., et al. N. Engl. J. Med. 369, 122-133 (2013). Rosenberg, S. A., et al. Clin. Cancer Res. 17, 4550-4557 (2011). Walter, S., et al. Nat. Med. 18, 1254-1261 (2012). Rosenberg, S. A. Sci. Transl. Med. 4, 127ps128 (2012). Tran, E., et al. Science 344, 641-645 (2014). Mantia-Smaldone, G. M., Corr, B. & Chu, C. S. Hum. Vaccin. Immunother. 8, 1179-1191 (2012). Haridas, D., et a. FASEB J. 28, 4183-4199 (2014). Deng, J., et al. Cancer Metastasis Rev. 32, 535-551 (2013). Luo, L. Y., et al. Cancer Res. 63, 807-811 (2003). Uyttenhove, C., et al. Nat. Med. 9, 1269-1274 (2003). Sorensen, R. B., et al. PLoS One 4, e6910 (2009). van den Brule, F., et al. Lab. Invest. 83, 377-386 (2003). Rubinstein, N., et al Cancer Cell 5, 241-251 (2004). Perez-Diez, A., et al Blood 109, 5346-5354 (2007). Braumuller, H., et al. Nature 494, 361-365 (2013). Hassan, R. & Ho, M. Eur. J. Cancer 44, 46-53 (2008). Schoggins, J. W., et al Nature 472, 481-485 (2011). Walter, S., et al. J. Immunol. 171, 4974-4978 (2003). Couzin-Frankel, J. Cancer immunotherapy. Science 342, 1432-1433 (2013). Mellman, I., Coukos, G. & Dranoff, G. Nature 480, 480-489 (2011). Perez, S. A., et al. Cancer 116, 2071-2080 (2010). Matsushita, H., et al. Nature 482, 400-404 (2012). Robbins, P. F., et al. Nat. Med. 19, 747-752 (2013). Gubin, M. M., et al Nature 515, 577-581 (2014). Andersen, R. S., et al. Cancer Res. 72, 1642-1650 (2012). Lu, Y. C., et al. Clin. Cancer Res. 20, 3401-3410 (2014). Rolland, P., Deen, S., Scott, I., Durrant, L. & Spendlove, I. Clin. Cancer Res. 13, 3591-3596 (2007). Cheng, W. F., et al. Br. J. Cancer 100, 1144-1153 (2009). Berlin, C., et al. Leukemia (2014). Blaustein, A. & Kurman, R. J. Blaustein's pathology of the female genital tract, (Springer, New York, N.Y., 2011). Pham, D. L., et al Int. J. Gynecol. Pathol. 32, 358-367 (2013).