Method for selecting new epitopes

Abstract

The present invention relates to a method for selecting new epitopes for an individual by selecting epitopes that bind to MHC I and / or MHC II and ranking their clinical utility. The present invention also provides a cancer vaccine obtained by the method described herein.

Description

Technical Field

[0001] The present invention relates to a method for selecting neoepitopes for an individual by selecting neoepitopes that bind to MHC I and / or MHC II and ranking their clinical utility. The present invention also provides a cancer vaccine obtained by the method described herein. Background Art

[0002] Although in the past few decades, especially due to early detection and diagnosis, the treatment of cancer has improved, which has greatly increased the survival rate, only about 60% of patients diagnosed with cancer are still alive 5 years after diagnosis.

[0003] Most cancer treatments used are surgical treatment, radiation, and cytotoxic chemotherapy. However, they all have serious side effects. Recently, antibodies against known cancer-related antigens or immunomodulatory molecules have also been used for treatment.

[0004] In recent years, cancer immunotherapy (e.g., cancer vaccines) that targets cancer cells with the patient's own immune system has attracted interest because this therapy can reduce or even eliminate some of the side effects seen in traditional cancer treatments.

[0005] The basis of immunology is based on self-nonself discrimination. Most pathogens that induce infectious diseases contain molecular signatures that can be recognized by the host and trigger an immune response. However, tumor cells are derived from normal cells and generally do not express any foreign molecular signatures, making it more difficult to distinguish them from normal cells.

[0006] However, most tumor cells express different types of tumor antigens. One class of tumor antigens is the so-called tumor-associated antigens, such as antigens that are expressed at low levels in normal tissues and at much higher levels in tumor tissues. In the past decade, such tumor-associated antigens have been the targets of cancer vaccines. However, immunotherapy against tumor-associated antigens presents several challenges, as tumor cells can evade the immune system by downregulating the antigens in question, and the treatment may also cause toxicity due to normal cell destruction.

[0007] Recently, another class of tumor antigens, namely the so-called tumor neoantigens, which are tumor-specific antigens, have been identified. Tumor neoantigens arise due to one or more mutations in the tumor genome, which result in changes in the amino acid sequence of the protein in question. Since these mutations do not exist in normal tissues, side effects of treatment with immunotherapy against tumor-specific neoantigens do not occur.

[0008] However, in order to produce an effective vaccine, it is important to select the most immunogenic neoepitopes and use them for the vaccine. SUMMARY OF THE INVENTION

[0010] The inventors of the present invention have developed a new epitope selection method to select new epitopes with properties that have been proven to be important for immunogenicity. By using this method, highly immunogenic new epitopes that are predicted to bind to the major histocompatibility complex (MHC) can be selected for vaccines, thereby generating vaccines capable of inducing strong and powerful immune responses. Thus, vaccines suitable for personalized cancer treatment can be produced. Accordingly, methods for ranking them according to the clinical utility of the new epitopes are provided herein, which are particularly suitable for the case of producing personalized vaccines for cancer treatment.

[0011] Accordingly, the present invention relates to a method for selecting A new epitopes for an individual, the method comprising the steps of:

[0012] a. obtaining one or more new epitopes from the individual, each new epitope comprising at least one minimal epitope, wherein each new epitope comprises at least one mutation, such as an immunogenic mutation, compared to a reference sequence;

[0013] b. determining MHC I and / or MHC II binding affinities for at least one minimal epitope within each of the new epitopes, such as at least two, three or four minimal epitopes;

[0014] c. selecting new epitopes comprising at least one minimal epitope predicted to bind to MHC I and / or MHC II, thereby obtaining new epitopes that bind to MHC;

[0015] d. ranking the new epitopes that bind to MHC according to the likelihood of clinical utility;

[0016] e. selecting A new epitopes from among the highest-ranked new epitopes that bind to MHC,

[0017] thereby selecting A new epitopes with clinical utility.

[0018] Also provided is a method for selecting A new epitopes for an individual, the method comprising the steps of:

[0019] a. obtaining one or more new epitopes from the individual, each new epitope comprising at least one minimal epitope, wherein each new epitope comprises at least one mutation, such as an immunogenic mutation, compared to a reference sequence;

[0020] b. determining MHC I binding affinity for each of the new epitopes, and determining the number of minimal epitopes that bind to MHC I for each of the new epitopes;

[0021] c. ranking the new epitopes as follows:

[0022] i. Prioritize new epitopes containing a large number of minimal epitopes that bind to MHC I and / or MHC II, and select a first group of new epitopes with high scores;

[0023] ii. Optionally, prioritize the new epitopes from the first group of new epitopes as follows:

[0024] 1) If the mutation is in an anchor position and the binding difference of the minimal epitope is equal to or greater than 20, the minimal epitope has the highest score;

[0025] 2) If the mutation is in a non - anchor position and the binding difference of the minimal epitope is equal to or greater than 20, the score of the minimal epitope is lower than that of the minimal epitope in 1);

[0026] 3) If the mutation is in an anchor position and the binding difference of the minimal epitope is less than 20 and equal to or greater than 3, the score of the minimal epitope is lower than that of the minimal epitope in 2);

[0027] 4) If the mutation is in a non - anchor position and the binding difference of the minimal epitope is less than 20 and equal to or greater than 3, the score of the minimal epitope is lower than that of the minimal epitope in 3);

[0028] 5) If the mutation is in an anchor position and the binding difference of the minimal epitope is less than 3 and equal to or greater than 1, the score of the minimal epitope is lower than that of the minimal epitope in 4);

[0029] 6) If the mutation is in a non - anchor position and the binding difference of the minimal epitope is less than 3 and equal to or greater than 1, the score of the minimal epitope is lower than that of the minimal epitope in 5);

[0030] 7) If the minimal epitope has a mutation with a binding difference less than 1, the score of the minimal epitope is the lowest, regardless of the position of the mutation;

[0031] wherein, the MHC I binding difference is given by: (% rank score (MHC I) for reference) / (% rank score (MHC I) for the minimal epitope); and select a second group of new epitopes with high scores;

[0032] iii. Optionally, prioritize the new epitopes from the second group based on the MHC I % rank score, and select a third group of new epitopes with a low MHC I % rank score;

[0033] iv. Optionally, select a fourth group of new epitopes from the second group or the third group, which includes minimal epitopes highly similar to epitopes known to be recognized by T cells,

[0034] v. Optionally, prioritize the new epitopes from the second group, the third group, or the fourth group based on the BLOSUM score, where a BLOSUM score less than a predetermined threshold ranks higher than a BLOSUM score equal to or greater than the threshold, and select the new epitopes of a fifth group with a BLOSUM score less than the threshold, where the threshold is preferably 1;

[0035] vi. Optionally, select new epitopes from the new epitopes of the first group, the second group, the third group, the fourth group, or the fifth group based on the new epitopes found in two or more samples, and select the new epitopes of a sixth group found in two or more samples;

[0036] vii. Optionally, select new epitopes from the new epitopes of the first group, the second group, the third group, the fourth group, the fifth group, or the sixth group based on the identification of mutations by at least two different variant callers, and select the new epitopes of a seventh group containing the mutations identified by at least two different variant callers,

[0037] where the new epitopes of the first group, the second group, the third group, the fourth group, the fifth group, the sixth group, or the seventh group contain the A epitopes.

[0038] Also provided is a method for preparing a cancer vaccine comprising the new epitopes, the method comprising the step of selecting the new epitopes using the method described herein.

[0039] Also described herein are cancer vaccines obtainable by the methods described herein.

[0040] Also provided is a method for selecting a number A of new epitopes for an individual, the method comprising the steps of:

[0041] a. Obtain one or more new epitopes from the individual, each new epitope comprising at least one minimal epitope, where each new epitope comprises at least one mutation, such as an immunogenic mutation, compared to a reference sequence;

[0042] b. Determine the MHC I and / or MHC II binding affinity for each of the new epitopes;

[0043] c. Select the new epitopes comprising at least one minimal epitope (e.g., at least two, three, or four minimal epitopes) predicted to bind to MHC I and / or MHC II, thereby obtaining the new epitopes that bind MHC;

[0044] d. Rank the new epitopes that bind MHC according to the likelihood of clinical utility;

[0045] e. Select A new epitopes from the new epitopes that bind MHC with the highest rank, where A is an integer and A is at least 3, such as at least 4, such as at least 5,

[0046] Thereby, A new epitopes that can induce a CD8+ T cell response when administered to an individual in an immunologically active amount are selected. Brief Description of the Drawings

[0048] Figure 1 . NetMHCpan / NetMHC IIpan predicted binding affinity % rankings of MHC class I (A) and II (B) molecules for immunogenic (left) and non-immunogenic (right) new epitopes (evaluated in mice vaccinated with VB10.NEO).

[0049] Figure 2 . BLOSUM scores of selected new epitopes with low % rankings for MHC class I (A) and II (B) molecules. Immunogenic peptides: left; non-immunogenic peptides: right.

[0050] Figure 3 . Cumulative number of IFN-γ spots of the top 20 new epitopes selected using NeoSELECT on the CT26 (A) and B16 (B) model datasets. The mean cumulative number of IFN-γ spots of 1000 randomly selected new epitopes for each dataset is depicted by the black line.

[0051] Figure 4 . Cumulative number of IFN-γ spots of the top 20 new epitopes (mutations) obtained from the CT26 (A), B16 (B), and LL2 (C) datasets using the NeoSELECT strategy and the wild-type sequences (WT) of these new epitopes. The mean cumulative number of IFN-γ spots of 1000 randomly selected new epitopes from CT26 (A), B16 (B), and LL2 (C) is depicted by the black line.

[0052] Figure 5 . BLOSUM scores of immunogenic and non-immunogenic new epitopes showing reduced MHC class I binding affinity (MHC class I binding difference below 1), increased MHC class I binding affinity (MHC class I binding difference above 3), and no change in MHC class I binding affinity are shown. The figure shows that when the MHC class I binding affinity is changed (left and middle panels), the BLOSUM scores of immunogenic new epitopes (light gray) are lower than those of non-immunogenic new epitopes (dark gray), while no change in BLOSUM scores is observed for new epitopes with unchanged MHC class I binding affinity.

[0053] Figure 6.In 2015, Kreiter et al. and in 2012, Castle et al. studied 10 different neoepitopes (pep1 - 10), all of which were predicted to bind to MHC class I (CD8+ T cell response), to investigate whether they could induce a CD8+ T cell response in a murine B16 - F10 melanoma tumor model. When administered as the vaccibody (“VB10.NEO”) described herein, the responses induced by 6 neoepitopes are shown in the figure above. When administered in the form of peptide plus poly ICLC adjuvant, RNA, and vaccibody, the CD4 and CD8 responses are summarized in the figure below, where white indicates no response, light gray indicates weak response, medium gray indicates medium response, and dark gray indicates strong response.

[0054] Figure 7 .Overview of the neoepitope selection process.

[0055] Figure 8 .Total number of predicted minimal epitopes for MHC class I and II (evaluated in mice vaccinated with VB10.NEO) among 27 - amino - acid - long neoepitopes. Im: Immunogenic; Non - Im: Non - immunogenic.

[0056] Figure 9 .Comparison between immunogenic and non - immunogenic neoepitopes in terms of the binding difference of MHC class I molecules between WT and MT for neoepitopes with mutations at the anchor positions.

[0057] Figure 10 .BLOSUM scores of selected neoepitopes with low % rank for MHC class I (A) and II (B) molecules.

[0058] Detailed description

[0059] Definition

[0060] As used herein, the term “tumor neoantigen” or “neoantigen” refers to any tumor - specific antigen that contains one or more mutations compared to the host's healthy tissue exome. Tumor neoantigens are used synonymously with the term cancer neoantigen. The one or more mutations may also be referred to as “neoepitope mutations”. The mutation can be any mutation that results in at least one amino - acid change. Thus, the mutation may be one of the following:

[0061] - Nonsynonymous mutation that results in an amino - acid change

[0062] - Mutation that results in a frameshift, thus forming a completely different open reading frame in the direction after the mutation

[0063] - Read - through mutation, where a stop codon is modified or deleted, resulting in a longer protein with a tumor - specific neoepitope

[0064] - Splicing mutations that result in unique tumor-specific protein sequences

[0065] - Chromosomal rearrangements that produce chimeric proteins with tumor-specific neoepitopes at the junction of two proteins.

[0066] As understood herein, a mutation does not necessarily refer to a mutation of a single residue, but more commonly refers to a difference between a given sequence (e.g., a potential neoepitope) and a reference sequence. Thus, a mutation can refer to a mutation of more than one amino acid residue. In some embodiments, the mutation is an immunogenic mutation.

[0067] As used herein, the term "tumor neoepitope" or "neoepitope" refers to any immunogenic mutation in a tumor antigen and is used synonymously with the term cancer neoepitope. The presence of a mutation is determined by comparing the neoepitope sequence derived from a tumor sample with a reference sequence present in a reference sample (e.g., healthy tissue from the same individual). It typically refers to a peptide of 27 amino acids in length. As defined herein, a neoepitope can contain one or several minimal epitopes. The mutation is typically present at or near the center of the neoepitope, i.e., position 14 in the 27-mer, but not necessarily at the center of the minimal epitope.

[0068] As used herein, the term "tumor neoepitope sequence" or "neoepitope sequence" refers to a sequence that contains a neoepitope in an antigen subunit and is used synonymously with the term cancer neoepitope sequence.

[0069] As used herein, the terms "tumor neoepitope peptide", "neoepitope peptide" refer to the peptide sequence of a neoepitope, wherein the peptide sequence contains a mutation.

[0070] The term "minimal epitope" refers to a subsequence of a neoepitope predicted to bind to MHC I or MHC II, which subsequence contains a mutation that can be immunogenic. In other words, a minimal epitope can be immunogenic, i.e., capable of eliciting an immune response, e.g., if it contains a mutation that confers immunogenicity to the minimal epitope or increases the immunogenicity of the minimal epitope. Such a mutation is referred to herein as an immunogenic mutation. Thus, the term minimal epitope can refer to a short subsequence of a neoepitope that is predicted to bind to MHC I or MHC II and contains the mutation found in the neoepitope. Thus, a 27-mer neoepitope containing a mutation at position 14 can encompass several minimal binding epitopes, i.e., minimal binding epitopes that are shorter than 27 amino acids but each contain the mutation. For example, a minimal epitope can consist of the first 14 amino acids of the neoepitope, provided that it is predicted to bind to MHC I or MHC II, or can consist of amino acids 9 to 18 or amino acids 7 to 22 of the neoepitope.

[0071] As used herein, the term "MHC molecule" includes class I MHC (MHC I) and class II MHC (MHC II) molecules. MHC I represents several loci, such as HLA-A (human leukocyte antigen-A), HLA-B, HLA-C, HLA-E, HLA-F, HLA-G, HLA-H, HLA-J, HLA-K, HLA-L, HLA-P, and HLA-V, while MHC II represents loci such as HLA-DRA, HLA-DRB1-9, HLA-, HLA-DQA1, HLA-DQB1, HLA-DPA1, HLA-DPB1, HLA-DMA, HLA-DMB, HLA-DOA, and HLA-DOB. The terms "MHC molecule" and "HLA molecule" are used interchangeably herein.

[0072] As used herein, the term "selecting a neoepitope" or "selection of a neoepitope" refers to the selection of a neoepitope for potential clinical or therapeutic use, preferably for a cancer vaccine. Thus, when a neoepitope is selected, it is a potential candidate for clinical use or for a cancer vaccine. The selected neoepitope is ranked higher or prioritized over non-selected neoepitopes.

[0073] A nucleotide is defined herein as a monomer of RNA or DNA. A nucleotide is a ribose or deoxyribose ring attached to a base and a phosphate group. Monophosphates, diphosphates, and triphosphates of nucleosides are all called nucleotides.

[0074] As used herein, the term "genome" refers to the total amount of genetic information in the chromosomes of an organism or cell.

[0075] As used herein, the term "exome" refers to the part of the genome formed by exons, which remain in mature RNA after introns are removed by RNA splicing when transcribed. It consists of all DNA that is transcribed into mature RNA in any type of cell outside the transcriptome, which is RNA transcribed only in a specific cell population.

[0076] As used herein, the term "mutation" includes translocations, inversions, deletions, duplications, and point mutations preferably present in the nucleotides encoding a neoepitope. In a preferred embodiment, the mutation is an amino acid substitution preferably present in the neoepitope.

[0077] As used herein, the term "cancer vaccine" refers to a vaccine for treating existing cancer or preventing the occurrence of cancer. A vaccine for treating existing cancer is called a therapeutic cancer vaccine.

[0078] Identification of new epitopes

[0079] The method according to the present invention may include the step of identifying one or more new epitopes by identifying tumor-specific mutations in nucleic acid sequences from a sample obtained from an individual as described herein. The one or more new epitopes are preferably a plurality of new epitopes; however, in some cases, even a single new epitope may be useful in the context of personalized therapy (e.g., personalized cancer therapy) as long as it can induce the desired immune response.

[0080] Preferably, the individual is a cancer patient. One or more samples can be obtained from the cancer patient to identify new epitopes that may be potential candidates for clinical use such as, for example, a personalized immunogenic cancer vaccine.

[0081] Preferably, tumor-specific mutations are identified by comparing the nucleotide sequence of a tumor sample obtained from the individual with a normal nucleotide sequence. The normal nucleotide sequence can be obtained by sequencing nucleic acids from a body fluid sample obtained from the individual or any non-tumor tissue. Preferably, the normal nucleotide sequence is obtained from healthy tissue of the same individual. The normal nucleotide sequence can also be obtained from a database. The term "normal" when applied to a sequence, whether a peptide sequence or a nucleotide sequence, will be used interchangeably herein with the terms "reference" or "wild-type". Thus, the term can apply to the corresponding sequence found in "normal" circumstances within the same individual or in a normal healthy population. Thus, comparing the tumor-specific sequence with the corresponding sequence found in another non-tumor tissue isolated from the same individual can reduce the number of false positives because new epitopes (e.g., containing SNPs that may be individual-specific but not tumor-specific) generated by genetic variation between individuals will be filtered out from the results. The reference sequence isolated from the individual can be from a healthy cell sample obtained from the same individual prior to diagnosis. Samples may also have been obtained before or after the start of treatment.

[0082] As used herein, the term "tumor sample" refers to a sample containing tumor cells. A tumor sample can be obtained by performing a biopsy on the individual or cancer patient. The biopsy can be a small sample of tumor tissue taken with a needle or a minor surgery. The sample can also be a lymph node biopsy. Similarly, several biopsies or tumor biopsies can be performed.

[0083] The reference sample can be a body fluid sample. The body fluid sample can be, for example, a urine sample, a fecal sample, a serum sample or a saliva sample. In a preferred embodiment, the body fluid sample is a blood sample. Preferably, the reference sample is obtained from healthy tissue of the individual in need of treatment.

[0084] Nucleic acids obtained from a sample can be sequenced using any known sequencing method. For example, next-generation sequencing can be used. In some embodiments, the nucleic acid sequences of tumor cells from an individual are compared with the nucleic acid sequences of normal cells (e.g., healthy cells or reference cells from the same individual) to identify sequence differences.

[0085] Binding affinity of MHC I and MHC II

[0086] To select and prioritize novel epitopes as the strongest candidates for personalized cancer vaccines, the present inventors have developed a method for ranking and selecting novel epitopes that elicit strong immune responses.

[0087] To initiate an immune response, a novel epitope should target major histocompatibility complex (MHC)-restricted epitopes, as only peptides capable of binding to MHC molecules can provide eligible T cell targets. Thus, the first step in selecting the most promising novel epitopes may be to determine their MHC binding affinities.

[0088] MHC I is found on the cell surface of all nucleated cells in the body. One function of MHC I is to display intracellular peptides to cytotoxic T cells. The MHC I complex - peptide complex is inserted into the plasma membrane of the cell presenting the peptide to cytotoxic T cells, thereby triggering the activation of cytotoxic T cells against a specific MHC - peptide complex. The peptide lies within the groove of the MHC I molecule, such that the peptide length is typically about 8 - 10 amino acids. MHC I can be used interchangeably with class I MHC.

[0089] MHC II molecules are a family of molecules typically found only on antigen - presenting cells (e.g., dendritic cells, mononuclear phagocytes, some endothelial cells, thymic epithelial cells, and B cells). MHC II can be used interchangeably with class II MHC.

[0090] In contrast to MHC I, antigens presented by MHC class II peptides are derived from extracellular proteins. The extracellular protein is endocytosed, digested in lysosomes, and then the resulting antigenic peptides are loaded onto MHC class II molecules and then presented at the cell surface. The antigen - binding groove of MHC class II molecules is open at both ends and is capable of presenting longer peptides, typically 15 to 24 amino acid residues in length. Additionally, exogenous antigens typically presented by MHC II on the surface of dendritic cells can be presented via cross - presentation through the MHC I pathway. Cross - presentation is essential for immunity against most tumors and viruses.

[0091] MHC class I molecules are recognized by the T cell receptor (TCR) and co - receptor on CD8+ T cells, while MHC class II molecules are recognized by the TCR and co - receptor on CD4+ T cells.

[0092] The inventors of the present invention have established a method for selecting new epitopes, wherein the new epitopes are ranked according to clinical utility, and wherein new epitopes having an affinity for MHC (e.g., MHC I and / or MHC II) are selected.

[0093] Accordingly, one aspect of the present invention relates to a method for selecting A new epitopes for an individual, the method comprising the steps of:

[0094] a. obtaining one or more new epitopes from the individual, each new epitope comprising at least one minimal epitope, wherein each new epitope comprises at least one mutation, such as an immunogenic mutation, compared to a reference sequence;

[0095] b. determining MHC I and / or MHC II binding affinities for at least one minimal epitope within each of the new epitopes, such as at least two, three or four minimal epitopes;

[0096] c. selecting new epitopes comprising at least one minimal epitope predicted to bind to MHC I and / or MHC II, thereby obtaining new epitopes that bind MHC;

[0097] d. ranking the new epitopes that bind MHC according to the likelihood of clinical utility;

[0098] e. selecting A new epitopes from among the highest-ranked new epitopes that bind MHC,

[0099] thereby selecting A new epitopes that are likely to have clinical utility.

[0100] New epitopes with a high likelihood of clinical utility are new epitopes that are immunogenic and are likely to be suitable for cancer vaccines. Accordingly, the method of the present invention is a method for selecting the most immunogenic new epitopes that are highly suitable for cancer vaccines.

[0101] MHC alleles in the population exhibit extreme polymorphism. Each genetic locus contains a large number of haplotypes, which contain different alleles encoding different peptides. Accordingly, it is preferred to determine MHC I and MHC II binding affinities between the new epitope sequences and MHC I and / or MHC II sequences obtained from the same individual.

[0102] In order to determine the MHC binding affinity of a new epitope identified in an individual, it is necessary to HLA type the individual. Accordingly, the method of the present invention may also include the step of determining the MHC binding affinity by determining the HLA genotype of the individual. For example, the HLA genotype is determined from a blood sample of the individual.

[0103] Techniques for determining the HLA type of an individual are well known. HLA typing can be performed by using any suitable sequencing method. Preferably, next-generation sequencing is used.

[0104] For example, DNA isolated from a blood sample of the individual can be sequenced on a next-generation sequencing platform such as, for example, an Illumina platform.

[0105] Computer programs that predict the binding affinity between a peptide and MHC I and MHC II molecules can be used to determine or predict MHC I and MHC II binding affinity.

[0106] Thus, in one embodiment, the MHC binding affinity is determined by computer prediction. Preferably, the computer prediction is performed by using a computer program that predicts the binding of peptides to class I MHC and / or class II MHC molecules. By comparing the predicted binding affinity value with the predicted binding affinities of a set of 100,000 random natural 9-mer peptides, the predicted binding affinity value can be converted into a percentage score. Preferably, the % rank score for predicting the binding affinity of new epitopes to MHC I and MHC II molecules is calculated using NetMHCpan as described in the following literature: Nielsen et al. 2016 (Nielsen, M. et al. (2016) “NetMHCpan-3.0: improved prediction of binding to MHC class I molecules integrating information from multiple receptor and peptide length data sets.” Genome Medicine: 8:33) and Vannessa, J. et al. (2017) “NetMHCpan-4.0: Improved Peptide–MHC Class I Interaction Predictions Integrating Eluted Ligand and Peptide Binding Affinity Data.” Vanessa Jurtz, Sinu Paul, Massimo Andreatta, Paolo Marcatili, Bjoern Peters and Morten Nielsen. The Journal of Immunology (2017)) and the NetMHC IIpan computer program as described in Andreatta et al. 2015 (Andreatta M et al Accurate pan-specific prediction of peptide-MHC class II binding affinity with improved binding core identification. Immunogenetics. 2015;67(11-12):641–650). The above databases of NetMHCpan 30.03.2016 and NetMHCIIpan 29.09.2015 are used to calculate the % rank score mentioned herein.

[0107] A low % rank score indicates strong binding affinity, while a higher % rank score indicates weaker binding affinity. Thus, a new epitope with a low %R rank score is preferably ranked above a new epitope with a higher % rank score.

[0108] The % rank score (MHC I) refers to the % rank score predicting the MHC I binding affinity of a peptide, minimal epitope, or new epitope.

[0109] The % rank score (MHC II) refers to the % rank score predicting the MHC II binding affinity of a peptide, minimal epitope, or new epitope.

[0110] As described above, each new epitope can contain one or more minimal epitopes. The % rank score (MHC I) or % rank score (MHC II) of each minimal epitope within a given new epitope can also be determined. Preferably, the % rank score of a new epitope is equal to the lowest % rank score of the minimal epitopes it contains, i.e., the binding affinity of the new epitope is considered the same as that of the best binder among the minimal epitopes it contains.

[0111] In some embodiments, a new epitope or minimal epitope with a % rank score (MHC I) less than or equal to 2 is considered to bind to MHC I. Generally, a lower % rank score (MHC I) indicates higher binding affinity. For example, in some embodiments, a new epitope or minimal epitope with a % rank score (MHC I) between 0.5 and 2 (0.5 < % rank score (MHC I) ≤ 2) can be considered a weak MHC I binder, while a % rank score (MHC I) equal to or below 0.5 indicates that the new epitope or minimal epitope is a strong MHC I binder. In some embodiments, a new epitope or minimal epitope with a % rank score > 2 is considered not to bind to MHC I. In a preferred embodiment, a new epitope that does not contain any minimal epitopes predicted to bind to MHC I is considered not to have clinical utility and is excluded or de-prioritized. In other words, step e of the method can include or consist of the following steps: excluding new epitopes predicted not to bind to MHC I or de-prioritizing new epitopes predicted not to bind to MHC I. De-prioritizing a new epitope means that the new epitope is considered to have a low relevance to clinical use. Excluding a new epitope means that the new epitope is not selected for clinical use and / or not selected for a vaccine.

[0112] Typically, new epitopes that include minimal epitopes with high MHC I binding affinity (i.e., low % rank score (MHC I)) rank higher than new epitopes that include minimal epitopes with lower MHC I binding affinity (i.e., higher % rank score (MHC I)). New epitopes that include minimal epitopes with high MHC I binding affinity (i.e., low % rank score (MHC I)) rank higher than new epitopes that include minimal epitopes with lower MHC I binding affinity (i.e., higher % rank score (MHC I)).

[0113] In some embodiments, new epitopes or minimal epitopes with a % rank score (MHC II) less than or equal to 10 are considered to bind to MHC II. For example, in some embodiments, new epitopes or minimal epitopes with a % rank score (MHC II) between 2 and 10 (2 < % rank score (MHC II) ≤ 10) can be considered weak MHC II binders, while a % rank score (MHC II) equal to or less than 2 indicates that the new epitope or minimal epitope is a strong MHC II binder. In some embodiments, new epitopes or minimal epitopes with a % rank score > 10 are considered not to bind to MHC II. In a preferred embodiment, new epitopes that do not contain any binding minimal epitopes are considered to have no clinical utility and are excluded or de-prioritized. In other words, the steps of the method of the present invention, such as step c or e of the method of the present invention, may include the following steps or consist of the following steps: excluding new epitopes that only contain minimal epitopes predicted not to bind to MHC II. De-prioritizing a new epitope means that the new epitope is considered to be of far less relevance to clinical use. Excluding a new epitope means that the new epitope is not selected for clinical use and / or not selected for a vaccine.

[0114] Typically, new epitopes or minimal epitopes with high MHC II binding affinity (i.e., low % rank score (MHC II)) rank higher than new epitopes or minimal epitopes with lower MHC I binding affinity (i.e., higher % rank score high (MHC II)). New epitopes that include minimal epitopes with high MHC II binding affinity (i.e., low % rank score (MHC II)) rank higher than new epitopes that include minimal epitopes with lower MHC II binding affinity (i.e., higher % rank score (MHC II)).

[0115] In a preferred embodiment, the minimal epitope that binds MHC II has a %-rank (MHC II) score below 10. Thus, the minimal epitope that binds MHC II and is included in the newly selected epitopes in step b of the method according to the invention preferably has a %-rank score (MHC II) below 10. This means that newly selected epitopes containing the minimal epitope are selected for potential clinical applications and constitute potential vaccine candidates, where the minimal epitope is predicted or determined to bind MHC II with a %-rank score below 10. Preferably, the prioritization of newly selected epitopes that contain a minimal epitope predicted to bind MHC II with a %-rank score greater than 10 is reduced or such newly selected epitopes are excluded.

[0116] Generally, newly selected epitopes that bind MHC II and contain a minimal epitope with a high MHC II binding affinity rank higher than newly selected epitopes that contain a minimal epitope with a lower MHC II binding affinity.

[0117] Number of minimal epitopes

[0118] Compared to newly selected epitopes that contain only one minimal epitope, newly selected epitopes that contain more than one minimal epitope can be expected to have higher clinical utility. Thus, in some embodiments, in step d of the above method, the number of minimal epitopes that contain immunogenic mutations and are capable of binding MHC molecules is the most important parameter. However, it is also possible that newly selected epitopes that are selected by the method of the invention and may have clinical utility contain only a single minimal epitope; such newly selected epitopes may be clinically relevant if no other newly selected epitopes are available and / or if the minimal epitope confers binding or strong binding to MHC molecules.

[0119] In one embodiment, newly selected epitopes that contain a higher number of minimal epitopes rank higher than newly selected epitopes that contain a lower number of minimal epitopes. Newly selected epitopes that contain a lower number of minimal epitopes include newly selected epitopes that contain only one minimal epitope.

[0120] Thus, in one embodiment, the method of the invention further includes the step of determining the number of minimal epitopes included in newly selected epitopes that bind MHC I, where newly selected epitopes that bind MHC I and contain a higher number of binding minimal epitopes rank higher than newly selected epitopes that bind MHC I and contain a lower number of binding minimal epitopes.

[0121] In another embodiment, the method of the invention includes or further includes the step of determining the number of minimal epitopes present in newly selected epitopes that bind MHC II, where newly selected epitopes that bind MHC II and contain a higher number of binding minimal epitopes rank higher than newly selected epitopes that bind MHC II and contain a lower number of binding minimal epitopes.

[0122] In some embodiments, the method includes determining the number of minimal epitopes capable of binding to MHC I molecules and the number of minimal epitopes capable of binding to MHC II molecules.

[0123] To rank the new epitopes as a function of the total number of minimal epitopes they contain, in some embodiments, the number of minimal epitopes binding to MHC I is weighted higher, for example, twice as high, compared to the number of minimal epitopes binding to MHC II. This means that the score for the number of minimal epitopes binding to MHC I is higher, for example, twice as high, than the score for the number of minimal epitopes binding to MHC II. For example, a new epitope predicted to have 14 minimal epitopes binding to MHC I and 13 minimal epitopes binding to MHC II has higher immunogenicity than a new epitope containing 12 minimal epitopes binding to MHC I and 15 minimal MHC II epitopes.

[0124] Thereby, the binding score for each new epitope is determined. In practice, the number of minimal (binding) epitopes contained in the new epitope is determined, thereby identifying x minimal epitopes binding to MHC I and / or y minimal epitopes binding to MHC II.

[0125] A suitable scoring scheme for determining the binding score of the new epitope can be:

[0126] a*2x + b*y

[0127] where a higher binding score indicates a higher likelihood of higher clinical utility as described above.

[0128] For example, if y>0, then a = 1; if y = 0, then a = 0; if x>0, then b = 1; if x = 0, then b = 0.

[0129] Another suitable scoring scheme can be:

[0130] a*x + b*y

[0131] where a higher binding score indicates a higher likelihood of higher clinical utility as described above.

[0132] For example, a = 2 and b = 1, or a = 2 if y>0, a = 0 if y = 0, b = 1 if x>0, b = 0 if x = 0.

[0133] In some embodiments, a new epitope comprising at least one minimal epitope that binds MHC I and at least one minimal epitope that binds MHC II ranks higher than a new epitope comprising only a minimal epitope that binds MHC I or only a minimal epitope that binds MHC II. In other embodiments, a new epitope comprising at least one minimal epitope that binds MHC I ranks higher than a new epitope comprising only a minimal epitope that binds MHC II. In other embodiments, a new epitope comprising at least one minimal epitope that binds MHC II ranks higher than a new epitope comprising only a minimal epitope that binds MHC I.

[0134] In some embodiments, a new epitope comprising at least 2 minimal epitopes that bind MHC ranks higher than a new epitope comprising only 1 minimal epitope that binds MHC. In some embodiments, a new epitope comprising at least 3 minimal epitopes that bind MHC ranks higher than a new epitope comprising fewer than 3 minimal epitopes that bind MHC. In some embodiments, a new epitope comprising at least 4 minimal epitopes that bind MHC ranks higher than a new epitope comprising fewer than 4 minimal epitopes that bind MHC. In some embodiments, a new epitope comprising at least 5 minimal epitopes that bind MHC ranks higher than a new epitope comprising fewer than 5 minimal epitopes that bind MHC. In some embodiments, a new epitope comprising at least 6 minimal epitopes that bind MHC ranks higher than a new epitope comprising fewer than 6 minimal epitopes that bind MHC. In some embodiments, a new epitope comprising at least 7 minimal epitopes that bind MHC ranks higher than a new epitope comprising fewer than 7 minimal epitopes that bind MHC. In some embodiments, a new epitope comprising at least 8 minimal epitopes that bind MHC ranks higher than a new epitope comprising fewer than 8 minimal epitopes that bind MHC. In some embodiments, a new epitope comprising at least 9 minimal epitopes that bind MHC ranks higher than a new epitope comprising fewer than 9 minimal epitopes that bind MHC. In some embodiments, a new epitope comprising at least 10 minimal epitopes that bind MHC ranks higher than a new epitope comprising fewer than 10 minimal epitopes that bind MHC. The minimal epitope that binds MHC can be a minimal epitope that binds MHC I or a minimal epitope that binds MHC II.

[0135] In some embodiments, a neoepitope comprising at least 2 minimal epitopes that bind MHC I ranks higher than a neoepitope comprising only 1 minimal epitope that binds MHC I or higher than a neoepitope comprising only minimal epitopes that bind MHC II. In some embodiments, a neoepitope comprising at least 3 minimal epitopes that bind MHC I ranks higher than a neoepitope comprising less than 3 minimal epitopes that bind MHC I or higher than a neoepitope comprising only minimal epitopes that bind MHC II. In some embodiments, a neoepitope comprising at least 4 minimal epitopes that bind MHC I ranks higher than a neoepitope comprising less than 4 minimal epitopes that bind MHC I or higher than a neoepitope comprising only minimal epitopes that bind MHC II. In some embodiments, a neoepitope comprising at least 5 minimal epitopes that bind MHC I ranks higher than a neoepitope comprising less than 5 minimal epitopes that bind MHC I or higher than a neoepitope comprising only minimal epitopes that bind MHC II. In some embodiments, a neoepitope comprising at least 6 minimal epitopes that bind MHC I ranks higher than a neoepitope comprising less than 6 minimal epitopes that bind MHC I or higher than a neoepitope comprising only minimal epitopes that bind MHC II. In some embodiments, a neoepitope comprising at least 7 minimal epitopes that bind MHC I ranks higher than a neoepitope comprising less than 7 minimal epitopes that bind MHC I or higher than a neoepitope comprising only minimal epitopes that bind MHC II. In some embodiments, a neoepitope comprising at least 8 minimal epitopes that bind MHC I ranks higher than a neoepitope comprising less than 8 minimal epitopes that bind MHC I or higher than a neoepitope comprising only minimal epitopes that bind MHC II. In some embodiments, a neoepitope comprising at least 9 minimal epitopes that bind MHC I ranks higher than a neoepitope comprising less than 9 minimal epitopes that bind MHC I or higher than a neoepitope comprising only minimal epitopes that bind MHC II. In some embodiments, a neoepitope comprising at least 10 minimal epitopes that bind MHC I ranks higher than a neoepitope comprising less than 10 minimal epitopes that bind MHC I or higher than a neoepitope comprising only minimal epitopes that bind MHC II.

[0136] In some embodiments, a new epitope comprising at least 2 minimal epitopes that bind MHC II ranks higher than a new epitope comprising only one minimal epitope that binds MHC II. In some embodiments, a new epitope comprising at least 3 minimal epitopes that bind MHC II ranks higher than a new epitope comprising fewer than 3 minimal epitopes that bind MHC II. In some embodiments, a new epitope comprising at least 4 minimal epitopes that bind MHC II ranks higher than a new epitope comprising fewer than 4 minimal epitopes that bind MHC II. In some embodiments, a new epitope comprising at least 5 minimal epitopes that bind MHC II ranks higher than a new epitope comprising fewer than 5 minimal epitopes that bind MHC II. In some embodiments, a new epitope comprising at least 6 minimal epitopes that bind MHC II ranks higher than a new epitope comprising fewer than 6 minimal epitopes that bind MHC II. In some embodiments, a new epitope comprising at least 7 minimal epitopes that bind MHC II ranks higher than a new epitope comprising fewer than 7 minimal epitopes that bind MHC II. In some embodiments, a new epitope comprising at least 8 minimal epitopes that bind MHC II ranks higher than a new epitope comprising fewer than 8 minimal epitopes that bind MHC II. In some embodiments, a new epitope comprising at least 9 minimal epitopes that bind MHC II ranks higher than a new epitope comprising fewer than 9 minimal epitopes that bind MHC II. In some embodiments, a new epitope comprising at least 10 minimal epitopes that bind MHC II ranks higher than a new epitope comprising fewer than 10 minimal epitopes that bind MHC II.

[0137] Binding difference

[0138] In some embodiments, a new epitope comprising a minimal epitope having a higher or much higher MHC I binding affinity compared to the MHC I binding affinity of a corresponding reference peptide is prioritized over a new epitope comprising a minimal epitope that has only a slightly improved MHC I binding affinity compared to the corresponding reference MHC I binding affinity.

[0139] Thus, the % rank score (MHC I) ratio between a new epitope and a corresponding reference peptide can be used to select new epitopes for potential clinical applications. Herein, the % rank score ratio for MHC I binding is also referred to as the MHC I binding difference.

[0140] Similarly, in some embodiments, a new epitope having a higher or much higher MHC II binding affinity than the MHC II binding affinity of a corresponding reference peptide is prioritized over a new epitope expressing a peptide that has only a slightly improved MHC II binding affinity compared to the corresponding reference MHC II binding affinity.

[0141] Thus, the ratio of the % rank score (MHC II) between the new epitope and the corresponding reference peptide can be used to select new epitopes for potential clinical applications. In this article, the % rank score ratio for MHC II binding is also referred to as the MHC II binding difference.

[0142] The MHC I binding difference between the reference peptide and the new epitope is given by: (reference % rank score (MHC I)) / (new epitope % rank score (MHC I)), where the reference % rank score (MHC I) is the predicted binding affinity between the reference peptide and the MHC I molecule, and the new epitope % rank score (MHC I) is the predicted binding affinity between the new epitope peptide and the MHC I molecule.

[0143] In one embodiment of the present invention, step b or c of the above method further comprises determining the MHC I binding difference given by: (reference % rank score (MHC I)) / (new epitope % rank score (MHC I)), where new epitopes with a high MHC I binding difference are ranked higher than new epitopes with a lower MHC I binding difference.

[0144] In one embodiment, if the MHC I binding difference (reference % rank score (MHC I)) / (new epitope % rank score (MHC I)) is less than 2, the prioritization of the new epitope is reduced.

[0145] In one embodiment, if the MHC I binding difference ((reference % rank score (MHC I)) / (new epitope % rank score (MHC I)) is less than 20 (such as less than 3, such as less than 1), the prioritization of the new epitope is reduced. If the binding difference is equal to or greater than 1 (such as equal to or greater than 3, such as equal to or greater than 20), the new epitope is considered to have potential clinical relevance.

[0146] In a similar manner, the MHC I binding difference between the reference peptide and the minimal epitope is given by: (reference % rank score (MHC I)) / (minimal epitope % rank score (MHC I)), where the reference % rank score (MHC I) is the predicted binding affinity between the reference peptide and the MHC I molecule, and the minimal epitope % rank score (MHC I) is the predicted binding affinity between the minimal epitope peptide and the MHC I molecule.

[0147] In one embodiment of the present invention, step b or c of the above method further comprises determining the MHC I binding difference given by the formula: (reference % rank score (MHC I)) / (minimal epitope % rank score (MHC I)), wherein a new epitope containing a minimal epitope with a high MHC I binding difference is ranked higher than a new epitope containing a minimal epitope with a lower MHC I binding difference.

[0148] In one embodiment, if the MHC I binding difference ((reference % rank score (MHC I) of the minimal epitope sequence) / (minimal epitope % rank score (MHC I))) is less than 20 (such as less than 3, such as less than 1), the prioritization of the new epitope is reduced. If the binding difference of the minimal epitope contained in the new epitope is equal to or greater than 1 (such as equal to or greater than 3, such as equal to or greater than 20), the new epitope is considered to have potential clinical relevance.

[0149] If a new epitope contains only one minimal epitope that binds to MHC I, the MHC I binding difference of the new epitope is equal to the MHC I binding difference of the minimal epitope. If a new epitope contains two or more minimal epitopes that bind to MHC I, the MHC I binding difference of the new epitope is considered to be equal to the highest MHC I binding difference among the two or more minimal epitopes that bind to MHC I. Similarly, if a new epitope contains only one minimal epitope that binds to MHC II, the MHC II binding difference of the new epitope is equal to the MHC II binding difference of the minimal epitope. If a new epitope contains two or more minimal epitopes that bind to MHC II, the MHC II binding difference of the new epitope is considered to be equal to the highest MHC II binding difference among the two or more minimal epitopes that bind to MHC II. The same applies to the reference peptide.

[0150] In some embodiments, the position of the mutation is also considered. In other words, the clinical utility of a new epitope or minimal epitope is determined not only according to the above, but also as a function of the mutation position. In a preferred embodiment, the mutation is immunogenic. The peptide-binding groove of the MHC molecule accommodates a peptide or peptide fragment, typically 9 amino acids in length, i.e., the minimal epitope. The contact between the minimal epitope and the MHC molecule is mediated by the side chains of the anchor residues. Taking a minimal epitope of 9 amino acids in length as an example, the anchoring positions of the minimal epitope to the MHC I molecule are positions 2 and 9. For binding to the MHC II molecule, the anchoring positions are positions 1, 4, and 9.

[0151] In some embodiments, the following new epitopes are excluded or the prioritization of the following new epitopes is reduced, regardless of their binding differences: new epitopes that include a minimal epitope in which the mutation is at a non-anchoring position. For a minimal epitope that binds to MHC I, this means that in some embodiments, if the mutation is at position 1, 3, 4, 5, 6, 7, or 8 of the minimal epitope or a partial minimal epitope accommodated within the MHC molecule binding groove, the new epitope is excluded or its prioritization is reduced, regardless of how high its binding difference is. Preferred new epitopes include a minimal epitope having a mutation at an anchoring position and a binding difference equal to or greater than 2 (such as equal to or greater than 4, such as equal to or greater than 6, such as equal to or greater than 8, such as equal to or greater than 10, such as equal to or greater than 12, such as equal to or greater than 14, such as equal to or greater than 16, such as equal to or greater than 18, and most preferably such as equal to or greater than 20).

[0152] The MHC II binding difference can also be used as a selection criterion. The MHC II binding difference between a reference peptide and a new epitope is given by: (% rank score (MHC II) of the reference sequence of the minimal epitope) / (% rank score (MHC II) of the minimal epitope), where the % rank score (MHC II) of the minimal epitope is the predicted binding affinity of the most optimally ranked minimal epitope in the new epitope, and the % rank score (MHC II) of the reference is the predicted binding affinity of the corresponding peptide in the reference sequence.

[0153] In one embodiment of the present invention, step b or c of the above method further includes determining the MHC II binding difference given by: (% rank score (MHC II) of the reference sequence of the minimal epitope) / (% rank score (MHC II) of the minimal epitope), where new epitopes having a high MHC II binding difference are ranked higher than new epitopes having a lower MHC II binding difference.

[0154] In a similar manner, the MHC II binding difference between a reference peptide and a minimal epitope is given by: (% rank score (MHC II) of the reference) / (% rank score (MHC II) of the minimal epitope), where the % rank score (MHC II) of the reference is the predicted binding affinity between the reference peptide and the MHC II molecule, and the % rank score (MHC II) of the minimal epitope is the predicted binding affinity between the minimal epitope peptide and the MHC II molecule.

[0155] The method for determining MHC II binding affinity has been described above.

[0156] In one embodiment of the present invention, the following new epitopes are excluded or their prioritization is reduced: the new epitope comprises a minimal epitope that binds to MHC II, and the minimal epitope is predicted to bind to MHC II with a % ranking score (MHC II) lower than 20 (such as lower than 15, such as lower than 14, such as for example lower than 13, lower than 12 or for example lower than 11).

[0157] In a preferred embodiment, the following new epitopes are excluded or their prioritization is reduced: the new epitope comprises a minimal epitope that binds to MHC II with a % ranking score (MHC II) less than 10.

[0158] In another embodiment, the following new epitopes are excluded or their prioritization is reduced: the new epitope comprises a minimal epitope that binds to MHC II with a % ranking score (MHC II) equal to or less than 9 (such as equal to or less than 8, such as for example equal to or less than 7, equal to or less than 6, such as equal to or less than 5, such as for example equal to or less than 4, equal to or less than 3 or for example equal to or less than 2).

[0159] Examples of scoring schemes that can be used in the steps (such as step d) of the above method are given below:

[0160] · Minimal epitopes with a binding difference < 1 (regardless of the position of the mutation): score equal to 1;

[0161] · Minimal epitopes with a mutation at a non-anchoring position and 1 ≤ binding difference < 3: score equal to 2;

[0162] · Minimal epitopes with a mutation at an anchoring position and 1 ≤ binding difference < 3: score equal to 3;

[0163] · Minimal epitopes with a mutation at a non-anchoring position and 3 ≤ binding difference < 20: score equal to 4;

[0164] · Minimal epitopes with a mutation at an anchoring position and 3 ≤ binding difference < 20: score equal to 5;

[0165] · Minimal epitopes with a mutation at a non-anchoring position and a binding difference > 20: score equal to 6;

[0166] · Minimal epitopes with a mutation at an anchoring position and a binding difference > 20: score equal to 7;

[0167] The binding difference can be an MHC I or MHC II binding difference.

[0168] It will be understood that the scores given above are arbitrary values and can be replaced by any other arbitrary values as long as the scores increase as exemplified above. Generally, it can be seen that regardless of the position of the mutation, the smallest epitope or epitopes with a low binding difference (e.g., less than 1) rank the lowest. For the smallest epitope or epitopes with a binding difference in the same range (e.g., 1 ≤ binding difference < 3; or 3 ≤ binding difference < 20; or binding difference > 20), the smallest epitope or new epitope containing a mutation at the anchoring position is ranked higher than the smallest epitope or new epitope containing a mutation at a non-anchoring position. Generally, the higher the binding difference, the higher the score. For similar scores, mutations at the anchoring position score higher than mutations at non-anchoring positions.

[0169] In the above scoring scheme, the smallest epitope ranked lowest has a mutation at a non-anchoring position or a mutation at an anchoring position with a binding difference less than 1.

[0170] Length of new epitopes

[0171] The new epitope sequence preferably has a length suitable for processing and presenting the smallest epitope contained in the new epitope on MHC molecules. Thus, in one embodiment, the length of the new epitope is from 7 to 40 amino acids, such as from 10 to 35 amino acids, or more preferably from 15 to 30 amino acids, such as from 25 to 30 amino acids. In a preferred embodiment, the new epitope is 27 amino acids. The length of the new epitope includes the length of the new epitope that binds both MHC I and MHC II.

[0172] Preferably, the mutation is substantially located in the middle of the new epitope sequence.

[0173] In some embodiments, the smallest epitope of a given new epitope consists of an amino acid number less than or equal to the amino acid number of the new epitope. In the case where the new epitope is the smallest epitope, the new epitope and the smallest epitope have the same length. In such embodiments, the length of the smallest epitope is from 7 to 40 amino acids, such as from 10 to 35 amino acids, or more preferably from 15 to 30 amino acids, such as from 25 to 30 amino acids. In one embodiment, the length of the smallest epitope is 27 amino acids. The length of the smallest epitope includes the length of the new epitope that binds both MHC I and MHC II.

[0174] In other embodiments, the minimal epitope contained within the new epitope is shorter than the new epitope. For example, the new epitope has a length of 7 to 40 amino acids, and the corresponding minimal epitope has a length of 6 to 39 amino acids, or has a length that is at least one amino acid shorter than the length of the new epitope (e.g., at least 2 amino acids, e.g., at least 3 amino acids, e.g., at least 4 amino acids, e.g., at least 5 amino acids, e.g., at least 6 amino acids, e.g., at least 7 amino acids, e.g., at least 8 amino acids, e.g., at least 9 amino acids, e.g., at least 10 amino acids, e.g., at least 11 amino acids, e.g., at least 12 amino acids, e.g., at least 13 amino acids, e.g., at least 15 amino acids, e.g., at least 16 amino acids, e.g., at least 17 amino acids, e.g., at least 18 amino acids, e.g., at least 19 amino acids, e.g., at least 20 amino acids). In a preferred embodiment, the new epitope is 27 amino acids, and the minimal epitope has a length of 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, or 26 amino acids.

[0175] In some embodiments, the minimal epitope binds or is predicted to bind MHC I, and the minimal epitope has a length of 5 to 20 amino acids, such as 6 to 19 amino acids, such as 7 to 18 amino acids, such as 7 to 17 amino acids, such as 7 to 16 amino acids, such as 8 to 15 amino acids, such as 8 to 14 amino acids. Generally, the length of such minimal epitopes is 8 to 14 amino acids.

[0176] In some embodiments, the minimal epitope binds or is predicted to bind MHC II, and the minimal epitope has a length of 5 to 20 amino acids, such as 6 to 19 amino acids, such as 7 to 18 amino acids, such as 7 to 17 amino acids, such as 8 to 16 amino acids, such as 9 to 15 amino acids. Generally, the length of such minimal epitopes is 9 to 15 amino acids.

[0177] Similar to known T cell recognition epitopes

[0178] In some embodiments, the method further comprises the step of selecting a set of new epitopes that contain minimal epitopes that are highly similar to epitopes known to be recognized by T cells. Those skilled in the art know where to find lists of such epitopes. For example, epitopes known to be recognized by T cells can be obtained from the Immune Epitope Database and Analysis Resource (IEDB database), which lists human infectious epitopes (https: / / www.iedb.org / ). Thus, the sequences of the minimal epitopes contained within new epitopes that are considered likely to have clinical utility can be run through the database to determine whether the minimal epitopes have a high degree of similarity to epitopes known to be recognized by T cells.

[0179] When applied to two sequences, the term "high similarity" means that these sequences have significant similarity, i.e., the sum of identical and similar matches is high. Similarity can be determined by calculating an alignment score, for example, using BLOSUM62. A person skilled in the art knows which alignment score threshold can be used to determine whether two sequences have high similarity to each other. For example, for a 9-mer epitope, an alignment score > 26 can be appropriately used as a threshold for selecting epitopes with high similarity.

[0180] Normalized alignment scores can also be used to evaluate whether two peptide sequences have high similarity to each other. For example, the normalized alignment score can be the BLOSUM62 alignment score obtained for a given minimal epitope and a given epitope in the IEDB database, divided by the alignment score between the minimal epitope and itself (corresponding to the maximum alignment score for that minimal epitope). To obtain a measure of similarity, the result is then subtracted from 1. In other words, the following calculation can be performed:

[0181] Similarity to known epitope = 1 – (minimal epitope alignment score) / (known epitope alignment score)

[0182] The higher the score, the higher the similarity to the known epitope.

[0183] Total number of minimal epitopes

[0184] Useful new epitopes contain at least one minimal epitope, the minimal epitope containing mutations such as immunogenic mutations, and the minimal epitope having a length of 5 to 20 amino acids, preferably 6 to 19 amino acids, such as 7 to 18 amino acids, for example 8 to 17 amino acids, such as 9 to 16 amino acids, for example 10 to 15 amino acids, such as 11 to 14 amino acids, for example 12 or 13 amino acids, such as 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19 or 20 amino acids. Preferably, the minimal epitope has a length of 8 to 14 amino acids. The new epitope may contain several minimal epitopes that can overlap. Thus, the new epitope can contain 1 to 100 minimal epitopes, for example 1 to 90 minimal epitopes, for example 1 to 80 minimal epitopes, for example 1 to 70 minimal epitopes, for example 1 to 60 minimal epitopes, for example 1 to 50 minimal epitopes, for example 1 to 40 minimal epitopes, for example 1 to 30 minimal epitopes, for example 1 to 20 minimal epitopes, for example 2 to 19 minimal epitopes, for example 3 to 18 minimal epitopes, for example 4 to 17 minimal epitopes, for example 5 to 16 minimal epitopes, for example 6 to 15 minimal epitopes, for example 7 to 14 minimal epitopes, for example 8 to 13 minimal epitopes, for example between 9 and 12 minimal epitopes, for example 10 or 11 minimal epitopes, all minimal epitopes containing immunogenic mutations. The minimal epitope has an affinity for MHC molecules such as MHC class I molecules or MHC class II molecules. The mutations contained in the new epitope do not necessarily lie at the center of the minimal epitope, but in some embodiments can lie at the center of the minimal epitope. Thus, the mutation can be at the first or last residue of the minimal epitope, or at any other residue between them. As described above, in preferred embodiments, the mutation is in an anchoring position. In some embodiments, minimal epitopes having mutations in non-anchoring positions are excluded or have their priority reduced.

[0185] Generally, a new epitope can contain n different minimal epitopes, the minimal epitopes having a length of k amino acids and all containing mutations. A given new epitope can contain minimal epitopes of different lengths.

[0186] Likelihood of naturally occurring mutations

[0187] In a preferred embodiment, the method of the present invention further comprises the step of determining the probability of a mutation occurring naturally in the newly identified epitope. Altering an amino acid in a protein may reduce its ability to perform its function, or even change its function. Changes in proteins that have important functions in cells may lead to cell death. Conversely, such changes can enable the cell to continue to function in a different manner and may even result in a beneficial / advantageous modification compared to the original protein, such that the mutation is passed on to the offspring of the organism. If such a change does not cause any significant physiological drawbacks to the offspring, there remains a possibility that this mutation will persist in the population. Since the physical and chemical properties of amino acids vary widely, they are grouped into categories with similar properties. Substituting one amino acid with another from the same group has a lesser impact on the structure and function of the protein compared to substituting with an amino acid from a different category. Thus, in some embodiments, newly identified epitopes containing mutant amino acids from physicochemical groups different from the reference amino acid are preferred over newly identified epitopes containing mutant amino acids from physicochemical groups the same as the reference amino acid.

[0188] Preferably, the mutation causes an amino acid substitution. As used herein, the term "substitution pair" refers to a pair consisting of the amino acid being replaced and the amino acid replacing it.

[0189] Thus, in a preferred embodiment of the present invention, step d of the method described herein and above further comprises determining a score reflecting the similarity between the mutant amino acid sequence and the reference amino acid sequence.

[0190] To calculate the score for a specific amino acid substitution, a substitution matrix can be used. The substitution matrix contains probabilities or log scores based on the frequencies of mutations observed in all available protein sequences. The lower the log odds score, the less likely it is to observe that amino acid substitution when comparing naturally occurring amino acid sequences to each other. It has also been shown that a low log odds score correlates well with a high difference in physicochemical properties between the amino acid substitution pair. For example, compared to substitutions with high log odds scores due to a large difference in physicochemical properties between the newly formed mutant peptide (epitope) and the reference peptide, substitutions with low log odds scores or a lower likelihood / frequency of occurrence of the mutation from an evolutionary perspective have a higher chance of being recognized by the T cell receptor. The T cell receptor has the ability not only to tolerate self-peptides but also to tolerate peptides that are highly similar to self-peptides (i.e., mutant peptides with high log odds). This mechanism is called central tolerance.

[0191] The probabilities used in matrix calculations are calculated by looking at "blocks" of conserved sequences found in multiple protein alignments. It is assumed that these conserved sequences are functionally important in the related proteins, and thus their substitution rates will be lower than less conserved regions. To reduce the bias of closely related sequences on the substitution rate, segments in blocks with sequence identity above a certain threshold have been clustered, thus reducing the weight of each such cluster (Henikoff, S; Henikoff, JG (1992). "Amino acid substitution matrices from protein blocks". Proceedings of the National Academy of Sciences of the United States of America. 89(22):10915–9.). For the BLOSUM62 matrix, this threshold has been set to 62%. Then the pairing frequencies between clusters are counted, so pairings are only counted between segments with identity less than 62%. One would use a higher numbered BLOSUM matrix to align two closely related sequences, and a lower numbered BLOSUM matrix for more divergent sequences.

[0192] Thus, in a preferred embodiment of the present invention, a score reflecting the probability that amino acid substitutions present in a new epitope occur randomly is determined using an evolution-based scoring matrix. In a more preferred embodiment, the scoring matrix is a log-odds matrix. In a particularly preferred embodiment, the matrix is a BLOSUM matrix, preferably the BLOSUM62 matrix.

[0193] To calculate the BLOSUM matrix, the following equation is used:

[0194] S ij =(1 / λ)*log(ρ ij / (q i *q j ))

[0195] ρ ij is the probability that two amino acids i and j substitute for each other in homologous sequences, and q i and q j are the background probabilities of finding amino acids i and j in any protein sequence. The factor λ is a scaling factor, set to make the matrix contain easily computable integer values.

[0196] New epitopes containing mutations with lower BLOSUM62 scores for amino acid substitution pairs rank higher than new epitopes with higher BLOSUM62 scores for amino acid substitution pairs.

[0197] In one embodiment, such novel epitopes are down-prioritized or excluded if they do not contain mutations linked to amino acid substitution pairs with a BLOSUM62 score of less than 3 (e.g., less than 2).

[0198] In a preferred embodiment of the present invention, such novel epitopes are down-prioritized or excluded if they do not contain amino acid substitution pairs with a BLOSUM62 score below 1. Thus, preferably, such novel epitopes are prioritized or selected for clinical applications if they contain at least one amino acid substitution pair with a BLOSUM62 score below 1.

[0199] In another embodiment of the present invention, such novel epitopes are down-prioritized or excluded if they do not contain amino acid substitution pairs with a BLOSUM62 score below 1. Thus, in one embodiment, such novel epitopes are prioritized or selected for clinical applications if they contain at least one amino acid substitution pair with a BLOSUM62 score below 1.

[0200] For novel MHC I-binding epitopes to be selected for potential clinical applications, preferably, the novel epitope satisfies at least one of the following two criteria i and ii:

[0201] i. High MHC I binding difference

[0202] ii. Low BLOSUM62 score.

[0203] In one embodiment, a novel MHC I-binding epitope that satisfies at least one of the following two criteria i and ii:

[0204] i. The MHC I binding difference of the novel epitope is greater than 1;

[0205] ii. The novel epitope contains an amino acid substitution with a BLOSUM62 score below 3 (e.g., below 2, preferably below 1)

[0206] The ranking of the novel epitope is higher than that of a novel MHC I-binding epitope that does not meet any of the criteria i and ii.

[0207] In a preferred embodiment, a novel MHC I-binding epitope that satisfies at least one of the following two criteria i and ii:

[0208] i. The MHC I binding difference of the novel epitope is greater than 3;

[0209] ii. The novel epitope contains an amino acid substitution with a BLOSUM62 score below 3 (e.g., below 2, preferably below 1)

[0210] The ranking of the novel epitope is higher than that of a novel MHC I-binding epitope that does not meet any of the criteria i and ii.

[0211] In an even more preferred embodiment, a novel epitope that binds MHC I and satisfies at least one of the following two criteria i and ii:

[0212] i. The MHC I binding difference of the novel epitope is greater than 2;

[0213] ii. The novel epitope contains amino acid substitution pairs with a BLOSUM62 score of less than 3 (e.g., less than 2, preferably less than 1)

[0214] The novel epitope has a higher rank than a novel epitope that binds MHC I and does not satisfy either of the criteria i and ii.

[0215] In some embodiments, preferably, a novel epitope that binds MHC I and does not satisfy either of the above criteria i and ii is excluded or its priority ranking is reduced. Additionally, in some embodiments, as described above, a novel epitope that binds MHC I and satisfies condition i but in which the mutation is at a non-anchor position is excluded or its priority ranking is reduced.

[0216] For a novel epitope that binds MHC II and is to be selected for potential clinical applications, preferably, it satisfies at least one of the following two criteria i and ii:

[0217] i. High MHC II binding difference

[0218] ii. Low BLOSUM62 score.

[0219] In one embodiment, a novel epitope that binds MHC II and satisfies at least one of the following two criteria i and ii:

[0220] i. The MHC II binding difference of the novel epitope is greater than 2;

[0221] ii. The novel epitope contains amino acid substitution pairs with a BLOSUM62 score of less than 3 (e.g., less than 2 or preferably less than 1)

[0222] The novel epitope has a higher rank than a novel epitope that binds MHC II and does not satisfy either of the criteria i and ii.

[0223] In a preferred embodiment, a novel epitope that binds MHC II and satisfies at least one of the following two criteria i and ii:

[0224] i. The MHC II binding difference of the novel epitope is greater than 3;

[0225] ii. The novel epitope contains amino acid substitution pairs with a BLOSUM62 score of less than 3 (e.g., less than 2 or preferably less than 1)

[0226] The new epitopes are ranked higher than the new MHC II-binding epitopes that do not meet any of the criteria i and ii.

[0227] In an even more preferred embodiment, new MHC II-binding epitopes that meet at least one of the following two criteria i and ii:

[0228] i. The MHC II-binding difference of the new epitope is greater than 2;

[0229] ii. The new epitope contains an amino acid substitution pair with a BLOSUM62 score below 1

[0230] The new epitopes are ranked higher than the new MHC II-binding epitopes that do not meet any of the criteria i and ii.

[0231] Preferably, new MHC II-binding epitopes that do not meet any of the above criteria i and ii are excluded or their prioritization is reduced. In addition, in some embodiments, as described above, new MHC II-binding epitopes that meet condition i but in which the mutation is at a non-anchor position are excluded or their prioritization is reduced.

[0232] RNA expression level

[0233] It has been found that the RNA expression level of the source gene of the mutation may be an important factor in the immunogenicity of the new epitope. Higher RNA expression levels generally lead to higher protein expression and the presentation of new epitopes on MHC molecules on tumor cells, thus having a higher potential to induce cancer-specific immunogenicity.

[0234] Therefore, in a preferred embodiment, the method of the present invention further includes the step of determining the RNA expression level of the new epitope.

[0235] Generally, new epitopes with high RNA expression levels are ranked higher than new epitopes with lower RNA expression levels.

[0236] In a specific embodiment, in step d of the method, new MHC I-binding epitopes with high RNA expression levels are ranked higher than new MHC I-binding epitopes with lower RNA expression levels. In another specific embodiment, new MHC II-binding epitopes with high RNA expression levels are ranked higher than new MHC II-binding epitopes with lower RNA expression levels.

[0237] Preferably, new epitopes for which the source gene RNA is not detected are excluded or their prioritization is reduced. In particular, new epitopes with a normalized transcript level expressed as 0 transcripts per million (TPM) are excluded or their prioritization is reduced. Transcripts with a detectable normalized transcript level expression (e.g., a transcript level expression higher than 0 TPM) are not excluded.

[0238] The RNA expression level can be determined by performing RNA sequencing on nucleic acids obtained from a tumor sample as defined above. RNA sequencing can be carried out by techniques known in the art. Methods for determining RNA expression levels are well known to those skilled in the art. For example, next-generation sequencing techniques such as the Illumina platform can be used to determine the RNA expression level.

[0239] Autoimmune risk

[0240] To minimize the risk of cross-reactivity, preferably the following new epitopes are excluded or their preferred ranking is reduced: the new epitopes contain a naturally occurring core peptide that matches a peptide sequence in the human proteome (e.g., a peptide of 9 amino acids in length containing a mutation). Thus, the method of the present invention may further include the step of comparing the new epitope peptide sequence with the peptide sequences of the human proteome.

[0241] Here, the human proteome is defined as all translations derived from known genes of the human genome. The known genes are published on www.ensembl.org. Thus, to determine whether the new epitope core peptide sequence matches a peptide sequence in the human proteome, the sequence can be compared with the peptide sequences of the human genome obtained from a database.

[0242] The new epitope core peptide sequence is included in the new epitopes as described above. Preferably, the following new epitopes are excluded or their priority ranking is reduced: the new epitopes contain a peptide sequence containing a mutation that matches a naturally occurring peptide sequence in the human proteome.

[0243] This is because, without being bound by theory, it is expected that mutations that match naturally occurring peptide sequences have a greater risk of being tolerated by T cell receptors due to central tolerance, and thus it is expected that such mutations have lower immunogenicity compared to mutations that do not match naturally occurring sequences.

[0244] In one embodiment, the new epitope core peptide sequence compared with the peptide sequences of the human genome contains 9 to 27 amino acids (such as, for example, 10 to 25 amino acids, such as 11 to 15 amino acids or such as 12, 13 or 14 amino acids), or consists of them.

[0245] In one embodiment, the new epitope peptide sequence compared with the peptide sequences of the human genome contains 5 to 15 amino acids (such as, for example, 6 to 12 amino acids, preferably 7 to 11 amino acids, or more preferably 8 to 10 amino acids), or consists of them.

[0246] In a preferred embodiment, the new epitope peptide sequence compared with the peptide sequences of the human genome contains 9 amino acids or consists of 9 amino acids.

[0247] In another embodiment, the neoepitope peptide sequence compared to the peptide sequence of the human genome comprises 15 amino acids, 14 amino acids, 13 amino acids, 12 amino acids, 11 amino acids or for example 10 amino acids, or consists of the same.

[0248] The neoepitope peptide sequence comprises a neoepitope mutation. In a preferred embodiment, the mutation is an amino acid substitution. It will be clear to the person skilled in the art that the sequence of the minimal epitope contained in the neoepitope can be compared to the peptide sequence of the human genome, rather than the sequence of a given neoepitope to the peptide sequence of the human genome.

[0249] To further minimize the risk of organ-specific autoimmunity, in some embodiments, neoepitopes found in genes highly expressed in a particular organ / tissue are excluded or their prioritization is reduced.

[0250] In one embodiment, neoepitopes or minimal epitopes present in the following genes are excluded or their prioritization is reduced, wherein the genes show an RNA expression level that is at least 3-fold higher, for example at least 4-fold higher, in any organ compared to other tissues.

[0251] In a preferred embodiment, neoepitopes or minimal epitopes (such as neoepitopes that bind to MHC I and / or MHC II) present in the following genes are excluded or their prioritization is reduced, wherein the genes show an RNA expression level that is at least 5-fold higher in any organ compared to other tissues.

[0252] In another embodiment, neoepitopes or minimal epitopes (such as neoepitopes or minimal epitopes that bind to MHC I and / or MHC II) present in the following genes are excluded or their prioritization is reduced, wherein the genes show an RNA expression level that is at least 6-fold higher, at least 7-fold higher or for example at least 8-fold higher in any organ compared to other tissues.

[0253] Preferably, the organ and tissue are from the same species, such as human. The RNA expression level can be obtained, for example, from a database such as www.gtexportal.org. The RNA expression level is also described in Uhlén, M. et al. (2015) (Uhlén, M. et al. (2015) “Proteomics. Tissue-based map of the human proteome.” Science; 347(6220)). Alternatively, the RNA expression level is obtained from the individual from whom the neoepitope is derived.

[0254] The organ can be selected, for example, from the heart and the brain. In another embodiment, the organ is selected from the liver, lung, stomach, kidney, spleen, colon and intestine.

[0255] Allele frequency

[0256] In one embodiment, the method of the present invention includes the step of determining the allelic frequency of the mutations present in each new epitope. As used herein, the term allelic frequency refers to the relative frequency of an allelic sequence containing a specific genomic mutation at a particular locus compared to the total number of sequences of all alleles in a specific genomic region. The term mutant allelic frequency refers to the allelic frequency containing a mutation or a new epitope mutation, and is used interchangeably herein with the term variant allelic frequency (VAF). Generally, a high mutant allelic frequency or VAF indicates higher clinical utility. A high mutant allelic frequency or VAF indicates a higher proportion of cancer cells containing the same mutation. Thus, without being bound by theory, it is expected that new epitopes with a high allelic frequency or VAF are present in cancer cells at a higher proportion than new epitopes with a low mutant allelic frequency or VAF.

[0257] In a particular embodiment, new epitopes that bind MHC I with a high mutant allelic frequency or VAF are ranked higher than new epitopes that bind MHC I with a lower mutant allelic frequency or VAF. In another particular embodiment, new epitopes that bind MHC II with a high mutant allelic frequency or VAF are ranked higher than new epitopes that bind MHC II with a lower mutant allelic frequency or VAF.

[0258] The mutant allelic frequency can be determined, for example, by analyzing sequencing data from a tumor sample and comparing the frequency of the mutant allele with the frequencies of other alleles present in the same gene.

[0259] In some embodiments, a mutant allelic frequency or VAF greater than 0.05 indicates the clinical utility of the minimal epitope or new epitope containing the mutation. Thus, in some embodiments, new epitopes or minimal epitopes containing a mutation with a VAF greater than 0.05 (e.g., greater than 0.06, e.g., greater than 0.07, e.g., greater than 0.08, e.g., greater than 0.09, e.g., greater than 0.1, e.g., greater than 0.15, e.g., greater than 0.20) are ranked higher, respectively, than new epitopes or minimal epitopes containing a mutation with a VAF equal to or less than 0.05 (e.g., equal to or less than 0.06, e.g., equal to or less than 0.07, e.g., equal to or less than 0.08, e.g., equal to or less than 0.09, e.g., equal to or less than 0.1, e.g., equal to or less than 0.15, e.g., equal to or less than 0.20).

[0260] Variant calling

[0261] Many types of somatic mutation identification software (variant callers) are available in the art. In some embodiments, neoepitopes or minimal epitopes containing mutations identified by at least two different variant callers are considered likely to have clinical utility. Thus, in some embodiments, the ranking of neoepitopes or minimal epitopes depends on how many variant callers identify the mutations they contain. Neoepitopes or minimal epitopes containing mutations identified by at least two different variant callers are ranked higher than neoepitopes or minimal epitopes containing mutations identified by only one variant caller. In some embodiments, the mutations are identified by at least 3 different variant callers (e.g., at least 4 different variant callers, e.g., at least 5 different variant callers, e.g., at least 6 different variant callers, e.g., at least 7 different variant callers, e.g., at least 8 different variant callers, e.g., at least 9 different variant callers, e.g., 10 or more different variant callers).

[0262] Similarity to known cancer-related genes

[0263] In some embodiments, neoepitopes derived from known cancer-related genes are prioritized.

[0264] Cancer-related genes are defined as genes involved in cancer pathogenesis. Cancer-related genes can be found in, for example, the "Catalog of Somatic Mutations in Cancer" (COSMIC) database, or other databases containing curated lists of cancer-related mutations and / or genes, as known to those of skill in the art. Thus, a match with a known cancer-related gene can be found by comparing the source gene, genomic location, or variant location (HGVS-nomenclature) of the neoepitope with the corresponding information in the COSMIC database.

[0265] In some embodiments, a neoepitope is considered likely to have clinical utility if the gene it contains is a known cancer-related gene. Thus, such neoepitopes are prioritized over neoepitopes contained in genes unrelated to cancer.

[0266] Clonality

[0267] As used herein, the term "clonality" refers to the occurrence of neoepitopes in a tumor sample. If a neoepitope or minimal epitope is found, for example, in more than one tumor biopsy, this indicates high clonality and thus a higher likelihood of being present in most cancer cells and thus a higher likelihood of having clinical utility.

[0268] The clonality of a neoepitope can be determined by obtaining at least two tumor samples from the same individual and determining whether the mutation is present in more than one biopsy. Preferably, neoepitopes found in more than one tumor sample are prioritized over neoepitopes found in only one biopsy. "More than one tumor sample" is understood to mean several samples from the same individual, which samples are from the same lesion or different lesions.

[0269] In some embodiments, a minimal epitope or neoepitope having a mutation found in at least two different samples from the same individual may have higher clinical utility compared to a minimal epitope or neoepitope having a mutation found in only one tumor sample. In some embodiments, the at least two different samples are samples from:

[0270] - the same tumor or lesion, such as biopsies from multiple angles of the same tumor or lesion

[0271] - at least two different tumors or lesions

[0272] - at least one tumor or lesion and at least one archival tumor material, such as an archival biopsy or archival resection material from the tumor or lesion,

[0273] Finding an epitope or neoepitope containing a mutation in at least two different samples indicates higher clinical utility. The term "archival tumor material" refers to material sampled from tumor tissue of the same individual at an earlier time point, such as an archival biopsy or archival resection material from the tumor or lesion.

[0274] In some examples, the at least two different samples are at least three different samples, such as at least four different samples, such as at least five different samples, such as at least six different samples, such as at least seven different samples, such as at least eight different samples, such as at least nine different samples, such as at least ten different samples or more. In some embodiments, at least one of the at least two different samples is archival tumor material from the individual, and at least one of the at least two different samples is a sample from a tumor or lesion.

[0275] Liquid biopsy

[0276] When selecting a neoepitope containing a minimal epitope that may have clinical utility, it may be advantageous to determine whether the neoepitope or minimal epitope is also found in the plasma of the individual. Thus, in some embodiments, the method includes sequencing nucleic acid sequences from cell-free DNA (e.g., circulating tumor DNA in plasma from a liquid biopsy obtained from the individual). This may be useful for identifying, validating, and / or tracking neoepitopes and can be done, for example, by sequencing cell-free DNA.

[0277] In some embodiments, neoepitopes found in circulating tumor DNA in plasma samples from an individual rank higher than neoepitopes not found in said samples.

[0278] Individual

[0279] The individual according to the method of the invention is preferably a human. Preferably, the individual or human is a cancer patient. Thus, the individual is preferably a human suffering from cancer. The cancer can be of all types. Preferably, the individual has at least one tumor.

[0280] The cancer can be any cancer in which the cancer cells contain at least one mutation. The cancer can be a primary tumor, a metastasis, or both. The tumor examined for mutations can be a primary tumor or a metastasis. The cancer to be treated can be a cancer known to have a high mutational burden, such as melanoma or lung cancer. The cancer to be treated can also be a cancer characterized by only one cancer-specific mutation.

[0281] Further criteria

[0282] Preferably, the neoepitopes that bind to MHC I and / or bind to MHC II meet at least one of the following criteria i-vii:

[0283] i. High RNA expression level

[0284] ii. High allele frequency

[0285] iii. High clonality

[0286] iv. Matches known cancer-related genes

[0287] v. Contains mutations identified by at least two different variant callers

[0288] vi. Is also found in the plasma of the individual, as determined, for example, by sequencing cell-free DNA

[0289] vii. Is also found in at least two samples from the same individual, such as two samples from the same lesion or from two different lesions, or at least one tumor or lesion and at least one archival tumor material sample from the tumor or lesion.

[0290] The RNA expression level, allele frequency, clonality, match to known cancer-related genes, identification by at least two different variant callers, presence in the plasma of the individual, and presence in at least two samples from the same individual are as defined above.

[0291] In a more preferred embodiment, the novel epitopes that bind MHC I and / or bind MHC II meet at least two of the following criteria i-vii:

[0292] i. High RNA expression level

[0293] ii. High allele frequency

[0294] iii. High clonality

[0295] iv. Matches known cancer-related genes

[0296] v. Contains mutations identified by at least two different variant callers

[0297] vi. Is also found in the plasma of an individual, as determined, for example, by sequencing cell-free DNA

[0298] vii. Is also found in at least two samples from the same individual, such as two samples from the same lesion or from two different lesions, or at least one tumor or lesion and at least one archival tumor material sample from the tumor or lesion.

[0299] In an even more preferred embodiment, the novel epitopes that bind MHC I and / or bind MHC II meet at least three of the following criteria i-vii:

[0300] i. High RNA expression level

[0301] ii. High allele frequency

[0302] iii. High clonality

[0303] iv. Matches known cancer-related genes

[0304] v. Contains mutations identified by at least two different variant callers

[0305] vi. Is also found in the plasma of an individual, as determined, for example, by sequencing cell-free DNA

[0306] vii. Is also found in at least two samples from the same individual, such as two samples from the same lesion or from two different lesions, or at least one tumor or lesion and at least one archival tumor material sample from the tumor or lesion.

[0307] In a preferred embodiment, the novel epitopes that bind MHC I and / or bind MHC II meet at least four of the following criteria i-vii:

[0308] i. High RNA expression level

[0309] ii. High allele frequency

[0310] iii. High clonality

[0311] iv. Matches known cancer-related genes

[0312] v. Contains mutations identified by at least two different variant callers

[0313] vi. Also found in the plasma of an individual, as determined, for example, by sequencing cell-free DNA

[0314] vii. Also found in at least two samples from the same individual, such as two samples from the same lesion or from two different lesions, or at least one tumor or lesion and at least one archival tumor material sample from the tumor or lesion

[0315] In another preferred embodiment, the novel epitopes that bind MHC I and / or bind MHC II meet at least five of the following criteria i-vii:

[0316] i. High RNA expression level

[0317] ii. High allele frequency

[0318] iii. High clonality

[0319] iv. Matches known cancer-related genes

[0320] v. Contains mutations identified by at least two different variant callers

[0321] vi. Also found in the plasma of an individual, as determined, for example, by sequencing cell-free DNA

[0322] vii. Also found in at least two samples from the same individual, such as two samples from the same lesion or from two different lesions, or at least one tumor or lesion and at least one archival tumor material sample from the tumor or lesion

[0323] In another preferred embodiment, the novel epitopes that bind MHC I and / or bind MHC II meet six of the following criteria i-vii:

[0324] i. High RNA expression level

[0325] ii. High allele frequency

[0326] iii. High clonality

[0327] iv. Matches known cancer-related genes

[0328] v. Contains mutations identified by at least two different variant callers

[0329] vi. It is also found in the plasma of an individual, as determined, for example, by sequencing cell-free DNA

[0330] vii. It is also found in at least two samples from the same individual, such as two samples from the same lesion or from two different lesions, or at least one tumor or lesion and at least one archival tumor material sample from the tumor or lesion.

[0331] In another preferred embodiment, the novel epitopes that bind MHC I and / or bind MHC II meet all of the following criteria i-vii:

[0332] i. High RNA expression level

[0333] ii. High allele frequency

[0334] iii. High clonality

[0335] iv. Matches known cancer-related genes

[0336] v. Contains mutations identified by at least two different variant callers

[0337] vi. It is also found in the plasma of an individual, as determined, for example, by sequencing cell-free DNA

[0338] vii. It is also found in at least two samples from the same individual, such as two samples from the same lesion or from two different lesions, or at least one tumor or lesion and at least one archival tumor material sample from the tumor or lesion.

[0339] The above criteria i-vii are not listed in any preferred order and can be prioritized in any way. RNA expression level, allele frequency, clonality, match to known cancer-related genes, identification by at least two different variant callers, presence in an individual's plasma, and presence in at least two samples from the same individual are as defined above.

[0340] Thus, preferably, novel epitopes that bind MHC I and / or bind MHC II with a higher RNA expression level are ranked higher than novel epitopes with a lower RNA expression level.

[0341] Preferably, novel epitopes that bind MHC I and / or bind MHC II with a higher allele frequency are ranked higher than novel epitopes with a lower allele frequency.

[0342] Preferably, novel epitopes that bind MHC I and / or bind MHC II with a higher clonality are ranked higher than novel epitopes with a lower clonality.

[0343] Preferably, the novel MHC class I-binding and / or MHC class II-binding epitopes that match known cancer-related genes are ranked higher than those that do not match known cancer-related genes.

[0344] Preferably, the novel MHC class I-binding and / or MHC class II-binding epitopes that contain mutations identified by at least two different variant callers are ranked higher than those that contain mutations identified by only one variant caller.

[0345] Preferably, the novel MHC class I-binding and / or MHC class II-binding epitopes found in an individual's plasma are ranked higher than those not found in the individual's plasma.

[0346] Preferably, the novel MHC class I-binding and / or MHC class II-binding epitopes found in two or more samples from the same individual are ranked higher than those found in only one sample from the individual. The two or more samples can be samples from the same lesion or different lesions, or samples from at least one tumor or lesion and at least one archival tumor material sample from the tumor or lesion.

[0347] Number of selected new epitopes

[0348] The novel epitopes selected for an individual are preferably used in a cancer vaccine, such as a cancer vaccine construct, preferably a nucleotide vaccine construct. The vaccine is also referred to as a personalized cancer vaccine. The vaccine is immunogenic.

[0349] In the method according to the invention, A novel epitopes with clinical utility are selected. The A novel epitopes are preferably used in a cancer vaccine or a cancer vaccine construct. Thus, the A novel epitopes are preferably included in the same vaccine construct. The vaccine construct is preferably a nucleotide construct. The nucleotide construct can be an RNA construct and / or a DNA construct. Preferably, the vaccine is a DNA construct, also referred to as a vaccinosome DNA vaccine.

[0350] A is an integer. For example, A is at least 1, such as at least 3, such as at least 5, such as at least 7, or such as at least 10, such as at least 20, such as at least 30, such as 40 or greater. A can be a predetermined number.

[0351] In addition, the inventors of the present invention have found that increasing the number of new epitopes in the vaccine construct from 1 to 3 new epitopes or from 3 new epitopes to 10 new epitopes results in a surprising increase in the immune response. Additionally, it has been found that increasing the number of new epitopes in the vaccine construct from 10 new epitopes to 15 or 20 new epitopes results in a further increase in the immune response. In some embodiments, the number of new epitopes in the vaccine construct is between 20 and 40, such as 30. In some embodiments, the number of new epitopes in the vaccine construct is 40 or more.

[0352] In one embodiment, A is an integer from 3 to 100, such as from 3 to 75, such as from 3 to 50, such as from 3 to 30, such as from 3 to 20, such as from 3 to 15 or such as for example from 3 to 10 new epitopes.

[0353] In another embodiment, A is an integer from 5 to 50, such as from 5 to 30, such as for example from 5 to 25, such as from 5 to 20, such as from 5 to 15, such as from 5 to 10.

[0354] In a further embodiment, A is an integer from 10 to 50, such as from 10 to 40, such as from 10 to 30, such as from 10 to 20, such as from 10 to 25, such as from 10 to 20 or such as for example from 10 to 15.

[0355] The new epitopes selected by the method of the present invention can be used to design a vaccine body vaccine, which will be described in further detail below. The vaccine body vaccine comprises at least one antigen unit, a targeting unit, and a dimerization unit. The new epitopes are preferably included in the antigen unit.

[0356] The inventors of the present invention have shown that, compared to a vaccine body DNA vaccine comprising only 3 new epitopes, a vaccine body DNA vaccine comprising 10 new epitopes induces a stronger and more extensive total immune response. Similarly, compared to a vaccine comprising only 10 new epitopes, a vaccine body DNA vaccine comprising 20 new epitopes induces a stronger and more extensive total immune response. However, the cancer to be treated may be associated with only one cancer-specific new epitope, and it may not be possible to construct a vaccine body DNA vaccine comprising more than one epitope.

[0357] In a preferred embodiment, A is an integer from 10 to 20.

[0358] In another embodiment, A is an integer from 15 to 50, such as from 15 to 30 or such as for example from 15 to 20.

[0359] In a specific embodiment, A is 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39 or 40.

[0360] In one embodiment, the antigen unit contains one copy of each cancer neoepitope, so that when the vaccine contains 10 neoepitopes, it can elicit a cell-mediated immune response against 10 different neoepitopes.

[0361] In the method of the present invention, the top x neoepitopes that bind to MHC I are selected, and the top y neoepitopes that bind to MHC II are selected, where x + y = A. A is the total number of selected neoepitopes. A is as defined above.

[0362] x and y are integers.

[0363] In a preferred embodiment, x > y. For example, x ≥ 2y. In another embodiment, x ≥ 2.5y. In one example, x ≥ 3y.

[0364] In one embodiment, 0.5A < x < A, where A is as defined above. Preferably, A is an integer from 10 to 20.

[0365] In another embodiment, 0.6A < x < A, such as 0.7A < x < A, such as for example 0.8A < x < A, where A is as defined above. Preferably, A is an integer from 10 to 20.

[0366] Specific method

[0367] In a specific embodiment, the method is a method for selecting a number A of neoepitopes for an individual, the method comprising the following steps:

[0368] a. Obtaining one or more neoepitopes from the individual, each neoepitope containing at least one minimal epitope, wherein each neoepitope contains at least one immunogenic mutation compared to a reference sequence;

[0369] b. Determining the MHC I binding affinity of at least one minimal epitope (e.g., at least two, three or four minimal epitopes) within each of the neoepitopes, and determining the number of minimal epitopes that bind MHC I within each of the neoepitopes; thereby identifying x neoepitopes that bind MHC I;

[0370] c. Ranking the neoepitopes as follows:

[0371] i. Prioritize new epitopes containing a large number of minimal epitopes and select a first group of new epitopes with high scores;

[0372] ii. Optionally, prioritize the new epitopes from the first group of new epitopes as follows:

[0373] 1) If the mutation is at an anchor position and the binding difference of the minimal epitope is equal to or greater than 20, the minimal epitope has the highest score;

[0374] 2) If the mutation is at a non-anchor position and the binding difference of the minimal epitope is equal to or greater than 20, the score of the minimal epitope is lower than that of the minimal epitope in 1);

[0375] 3) If the mutation is at an anchor position and the binding difference of the minimal epitope is less than 20 and equal to or greater than 3, the score of the minimal epitope is lower than that of the minimal epitope in 2);

[0376] 4) If the mutation is at a non-anchor position and the binding difference of the minimal epitope is less than 20 and equal to or greater than 3, the score of the minimal epitope is lower than that of the minimal epitope in 3);

[0377] 5) If the mutation is at an anchor position and the binding difference of the minimal epitope is less than 3, the score of the minimal epitope is lower than that of the minimal epitope in 4);

[0378] 6) If the mutation is at a non-anchor position and the binding difference of the minimal epitope is less than 3, the score of the minimal epitope is lower than that of the minimal epitope in 5);

[0379] 7) If the minimal epitope has a mutation with a binding difference less than 1, the minimal epitope has the lowest score, regardless of the position of the mutation;

[0380] wherein, the MHC I binding difference is given by: (% rank score (MHC I) for reference) / (% rank score (MHC I) for the minimal epitope); and select a second group of new epitopes with high scores;

[0381] iii. Optionally, prioritize the new epitopes from the second group based on the MHC I % rank score and select a third group of new epitopes with a low MHC I % rank score;

[0382] iv. Optionally, select a fourth group of new epitopes that includes minimal epitopes highly similar to epitopes known to be recognized by T cells;

[0383] v. Optionally, based on the BLOSUM score, prioritize the new epitopes from the third group or the fourth group, where a BLOSUM score less than a predetermined threshold ranks higher than a BLOSUM score equal to or greater than the threshold, and select the fifth group of new epitopes with a BLOSUM score less than the threshold, where the threshold is preferably 1;

[0384] where the first group, second group, third group, fourth group, or fifth group of new epitopes comprises the A epitopes.

[0385] In certain embodiments, the method is a method for selecting a number A of new epitopes for an individual, the method comprising the steps of:

[0386] a. Obtain one or more new epitopes from the individual, each new epitope comprising at least one minimal epitope, where each new epitope comprises at least one immunogenic mutation compared to a reference sequence;

[0387] b. Determine the MHC I binding affinity of at least one minimal epitope (e.g., at least two, three, or four minimal epitopes) within each of the new epitopes, and determine the number of minimal epitopes within each of the new epitopes that bind MHC I; thereby identifying x new epitopes that bind MHC I;

[0388] c. Rank the new epitopes as follows:

[0389] i. Prioritize the new epitopes that contain a large number of minimal epitopes and select a first group of new epitopes with high scores;

[0390] ii. Optionally, prioritize the new epitopes from the first group of new epitopes as follows:

[0391] 1) If the mutation is in an anchor position and the binding difference of the minimal epitope is equal to or greater than 20, the minimal epitope has the highest score;

[0392] 2) If the mutation is in a non-anchor position and the binding difference of the minimal epitope is equal to or greater than 20, the score of the minimal epitope is lower than the score of the minimal epitope in 1);

[0393] 3) If the mutation is in an anchor position and the binding difference of the minimal epitope is less than 20 and equal to or greater than 3, the score of the minimal epitope is lower than the score of the minimal epitope in 2);

[0394] 4) If the mutation is in a non-anchor position and the binding difference of the minimal epitope is less than 20 and equal to or greater than 3, the score of the minimal epitope is lower than the score of the minimal epitope in 3);

[0395] 5) If the mutation is in an anchor position and the binding difference of the minimal epitope is less than 3, the score of the minimal epitope is lower than the score of the minimal epitope in 4);

[0396] 6) If the mutation is in a non-anchoring position and the binding difference of the minimal epitope is less than 3, the score of the minimal epitope is lower than that of the minimal epitope in 5);

[0397] 7) If the minimal epitope has a mutation with a binding difference lower than 1, the score of the minimal epitope is the lowest, regardless of the position of the mutation;

[0398] Wherein, the MHC I binding difference is given by: (% ranking score (MHC I) for reference) / (% ranking score (MHC I) for the minimal epitope); and select a second group of newly identified epitopes with high scores;

[0399] iii. Optionally, rank the newly identified epitopes from the second group based on the MHC I % ranking score, and select a third group of newly identified epitopes with a low MHC I % ranking score;

[0400] iv. Optionally, select a fourth group of newly identified epitopes, which includes minimal epitopes highly similar to the epitopes known to be recognized by T cells;

[0401] v. Optionally, rank the newly identified epitopes from the third group or from the fourth group based on the BLOSUM score, wherein a BLOSUM score less than a predetermined threshold ranks higher than a BLOSUM score equal to or greater than the threshold, and select a fifth group of newly identified epitopes with a BLOSUM score less than the threshold, wherein the threshold is preferably 1;

[0402] vi. Optionally, select newly identified epitopes from the first group, the second group, the third group, the fourth group or the fifth group of newly identified epitopes based on the newly identified epitopes found in two or more samples, and select a sixth group of newly identified epitopes found in two or more samples;

[0403] vii. Optionally, select newly identified epitopes from the first group, the second group, the third group, the fourth group, the fifth group or the sixth group of newly identified epitopes based on the identification of mutations by at least two different variant callers, and select a seventh group of newly identified epitopes containing mutations identified by at least two different variant callers,

[0404] Wherein the first group, the second group, the third group, the fourth group, the fifth group, the sixth group or the seventh group of newly identified epitopes contains the A epitopes.

[0405] The method may further include ranking the newly identified epitopes according to any additional parameters described herein (such as clonality, RNA expression level, and allele frequency).

[0406] In some embodiments, the second group is a subgroup of the first group. In some embodiments, the third group is a subgroup of the second group and / or the first group. In some embodiments, the fourth group is a subgroup of the third group and / or the second group and / or the first group. In some embodiments, the fifth group is a subgroup of the fourth group and / or the third group and / or the second group and / or the first group. In some embodiments, the sixth group is a subgroup of the fifth group and / or the fourth group and / or the third group and / or the second group and / or the first group. In some embodiments, the seventh group is a subgroup of the sixth group and / or the fifth group and / or the fourth group and / or the third group and / or the second group and / or the first group. The size of the subgroups may be the same.

[0407] In one embodiment, the method includes step i. above. In another embodiment, the method includes steps i. and ii. above. In another embodiment, the method includes steps i. and iii. above. In another embodiment, the method includes steps i. and iv. above. In another embodiment, the method includes steps i., ii. and iii. above. In another embodiment, the method includes steps i., ii. and iv. above. In another embodiment, the method includes steps i., iii. and iv. above. In another embodiment, the method includes steps i., ii., iii. and iv. above. In some embodiments, the method further includes step v. and / or step vi. and / or step vii. above.

[0408] In one embodiment, a method for selecting a number A of new epitopes for an individual having or suspected of having cancer comprises the steps of:

[0409] a. Obtaining one or more new epitopes from the individual, each new epitope comprising at least one minimal epitope, wherein each new epitope comprises at least one mutation, such as an immunogenic mutation, compared to a reference sequence, wherein the minimal epitope is composed of an amino acid number equal to or less than the amino acid number of the new epitope and comprises the at least one mutation; wherein preferably obtaining the new epitope comprises the step of identifying mutations in a nucleic acid sequence specific to the tumor;

[0410] b. Determining the MHC I and / or MHC II binding affinity for at least one minimal epitope within each of the new epitopes, such as at least two, three or four minimal epitopes, optionally, wherein the binding affinity is determined by computer (in silico) prediction;

[0411] c. Selecting new epitopes comprising at least one minimal epitope predicted to bind to MHC I and / or MHC II, thereby obtaining new epitopes that bind MHC;

[0412] d. Rank the novel MHC-binding epitopes according to the likelihood of clinical utility as described herein;

[0413] e. Select A novel epitopes from among the highest-ranked novel MHC-binding epitopes,

[0414] thereby selecting A novel epitopes that are likely to have clinical utility, wherein step d further comprises ranking the novel epitopes as follows:

[0415] i) For each minimal epitope contained in the novel epitopes, determine whether the mutation is at an anchor position or a non-anchor position of the minimal novel epitope;

[0416] ii) Prioritize the novel epitopes,

[0417] Optionally, wherein prioritizing the novel epitopes in step ii) is performed by assigning to each novel epitope the highest score of the minimal epitopes they contain, and wherein prioritizing the novel epitopes is performed as follows:

[0418] 1) If the mutation is at an anchor position and the binding difference of the minimal epitope is equal to or greater than 20, the minimal epitope has the highest score;

[0419] 2) If the mutation is at a non-anchor position and the binding difference of the minimal epitope is equal to or greater than 20, the minimal epitope has a score lower than that of the minimal epitope in 1);

[0420] 3) If the mutation is at an anchor position and the binding difference of the minimal epitope is less than 20 and equal to or greater than 3, the minimal epitope has a score lower than that of the minimal epitope in 2);

[0421] 4) If the mutation is at a non-anchor position and the binding difference of the minimal epitope is less than 20 and equal to or greater than 3, the minimal epitope has a score lower than that of the minimal epitope in 3);

[0422] 5) If the mutation is at an anchor position and the binding difference of the minimal epitope is less than 3 and equal to or greater than 1, the minimal epitope has a score lower than that of the minimal epitope in 4);

[0423] 6) If the mutation is at a non-anchor position and the binding difference of the minimal epitope is less than 3 and equal to or greater than 1, the minimal epitope has a score lower than that of the minimal epitope in 5);

[0424] 7) If the minimal epitope has a mutation with a binding difference lower than 1, the minimal epitope has the lowest score, regardless of the position of the mutation.

[0425] In one embodiment, a method for selecting A novel epitopes for an individual having or suspected of having cancer comprises the following steps:

[0426] a. Obtain one or more new epitopes from the individual, each new epitope comprising at least one minimal epitope, wherein each new epitope comprises at least one mutation, such as an immunogenic mutation, compared to a reference sequence, wherein the minimal epitope consists of an amino acid number equal to or less than the amino acid number of the new epitope and comprises the at least one mutation; wherein preferably obtaining the new epitope includes the step of identifying mutations in a nucleic acid sequence specific to the tumor;

[0427] b. For at least one minimal epitope within each of the new epitopes, such as at least two, three or four minimal epitopes, determine the MHC I and / or MHC II binding affinity, optionally, wherein the binding affinity is determined by computer prediction;

[0428] c. Select new epitopes comprising at least one minimal epitope predicted to bind to MHC I and / or MHC II, thereby obtaining new epitopes that bind MHC;

[0429] d. Rank the new epitopes that bind MHC according to the likelihood of clinical utility;

[0430] e. Select A new epitopes from the new epitopes that bind MHC with the highest rank,

[0431] thereby selecting A new epitopes that may have clinical utility,

[0432] wherein step d includes ranking the new epitopes according to the number of minimal epitopes comprised by the new epitope, wherein a higher number of minimal epitopes gives a higher rank.

[0433] In another embodiment, a method for selecting A new epitopes for an individual suffering from or suspected of suffering from cancer comprises the following steps:

[0434] a. Obtain one or more new epitopes from the individual, each new epitope comprising at least one minimal epitope, wherein each new epitope comprises at least one mutation, such as an immunogenic mutation, compared to a reference sequence, wherein the minimal epitope consists of an amino acid number equal to or less than the amino acid number of the new epitope and comprises the at least one mutation; wherein preferably obtaining the new epitope includes the step of identifying mutations in a nucleic acid sequence specific to the tumor;

[0435] b. For at least one minimal epitope within each of the new epitopes, such as at least two, three or four minimal epitopes, determine the MHC I and / or MHC II binding affinity, optionally, wherein the binding affinity is determined by computer prediction;

[0436] c. Select new epitopes comprising at least one minimal epitope predicted to bind to MHC I and / or MHC II, thereby obtaining new epitopes that bind MHC;

[0437] d. Rank the MHC-binding neoepitopes according to the likelihood of clinical utility;

[0438] e. Select A neoepitopes from the highest-ranked MHC-binding neoepitopes,

[0439] thereby selecting A neoepitopes that may have clinical utility,

[0440] wherein step d includes ranking the neoepitopes according to the number of samples in which the neoepitopes are found, wherein neoepitopes found in a higher number of samples are ranked higher than neoepitopes found in a lower number of samples, preferably wherein the samples are samples from different lesions.

[0441] Each of the above steps of the specific methods described in this section can be as described elsewhere herein.

[0442] Vaccine

[0443] Another aspect of the invention relates to a method of preparing a cancer vaccine comprising a neoepitope, the method comprising the step of using the method of selecting a neoepitope as defined herein to select the neoepitope.

[0444] In some embodiments, 10 - 40 neoepitopes are selected. That is, A is an integer from 10 to 40. In other embodiments, 10 - 30 or 10 - 20 neoepitopes are selected.

[0445] In one embodiment, the cancer vaccine comprises a nucleotide construct comprising:

[0446] - a targeting unit

[0447] - a dimerization unit

[0448] - a first linker

[0449] an antigen unit, wherein the antigen unit comprises A - 1 antigen subunits, each subunit comprising a sequence encoding at least one of the neoepitopes and a second linker, and the antigen unit further comprises a final sequence encoding one of the neoepitopes, wherein A is an integer from 1 to 100, such as 3 to 50.

[0450] wherein the nucleotide construct is applied to the anti-cancer vaccine in an immunologically effective amount.

[0451] The integer A is as described herein and above. The above DNA vaccine is also referred to as a vaccine body DNA vaccine.

[0452] The invention relates to a vaccine prepared by the above method.

[0453] Thus, the neoantigen vaccine provided by the method described herein may comprise a polynucleotide encoding a polypeptide comprising three units, namely a targeting unit, a dimerization unit, and an antigen unit. Due to the dimerization unit, the polypeptide forms a dimer protein called a vaccine body.

[0454] The genes encoding these three units are genetically engineered to be expressed as one gene. When expressed in vivo, the polypeptide / dimer protein targets antigen-presenting cells (APCs), resulting in enhanced vaccine potency compared to the same non-targeted antigen.

[0455] Antigen unit

[0456] The antigen unit comprises a plurality of tumor neoepitopes, where each neoepitope corresponds to a mutation identified in a tumor neoantigen. As described above, the mutation can be one or more mutations.

[0457] In the antigen unit, all tumor neoepitopes except the last one are arranged in antigen subunits, where each subunit consists of a tumor neoepitope sequence and a second linker, and the last subunit contains only the neoepitope, i.e., without such a second linker. Since the tumor neoepitope sequences are separated by the second linker, each neoepitope is presented to the immune system in an optimal manner, thus ensuring the efficiency of the vaccine as described below.

[0458] The cancer neoepitope sequence preferably has a length suitable for presentation by the MHC molecules described above. The preferred neoepitope length is as described above.

[0459] If the vaccine is directed against the expression product of the gene, in order to avoid tumors escaping the immune system by shutting off the expression of the mutant gene, it is preferred to include a plurality of different neoepitopes in the antigen unit. Generally, the more genes a tumor has to shut off, the less likely it is that the tumor can shut off all these genes and still be able to proliferate or even survive. In addition, tumors may be heterogeneous because not every neoantigen is expressed by all tumor cells. Thus, according to the present invention, the method is to include many neoepitopes in the vaccine to effectively attack the tumor. Preferably, the plurality of neoepitopes target the expression of multiple genes. Similarly, in order to ensure that all neoepitopes are effectively loaded onto the same antigen-presenting cells, they are arranged as one amino acid chain rather than discrete peptides.

[0460] The number of neoepitopes selected and included in the vaccine construct is as described above.

[0461] In one embodiment, the antigen unit contains one copy of each cancer neoepitope, so as to elicit responses against as many different neoepitopes included in the vaccine as possible. For example, when 10 neoepitopes are included in the vaccine, a cell-mediated immune response against 10 different neoepitopes can be elicited, or when 20 neoepitopes are included in the vaccine, a cell-mediated immune response against 20 different neoepitopes can be elicited. In some embodiments, the vaccine may contain more than 20 neoepitopes, so as to elicit responses against as many neoepitopes as possible.

[0462] However, if only some relevant antigen mutations are identified, the antigen unit may contain at least two copies of at least one neoepitope in order to enhance the immune response against these neoepitopes. Also for manufacturing and regulatory reasons, it may be advantageous to keep the plasmid and thus the length of the antigen unit constant or of similar length, and thus it may be advantageous to include more than one copy of the same neoepitope in the antigen unit.

[0463] As described above, it may be advantageous to keep the length of the antigen unit constant, and thus in one embodiment it is preferred that all cancer neoepitope sequences have the same length. However, if one or more of the neoepitopes are generated by mutations causing frameshifts or stop codon mutations, the neoepitopes may have a rather long length, for example consisting of at least the mutated part of the protein, the most antigenic part of the mutant protein or perhaps the whole mutant protein, where the length of at least one neoepitope is substantially longer than that of the neoepitopes generated by non-synonymous point mutations.

[0464] The length of the antigen unit is mainly determined by the length of the neoepitopes and the number of neoepitopes arranged in the antigen unit, and is about 21 to 1500, preferably about 30 amino acids to about 1000 amino acids, more preferably about 50 to about 500 amino acids, such as about 100 to about 400 amino acids, about 100 to about 300 amino acids.

[0465] The cancer neoepitope sequence inserted into the vaccine may contain mutations flanked by amino acid sequences on both sides. Preferably, the mutation is substantially located in the middle of the cancer neoepitope sequence to ensure that the immunogenic mutation is presented by antigen-presenting cells after processing. The amino acid sequences flanking the mutation are preferably the amino acid sequences flanking the mutation in the neoantigen, whereby the cancer neoepitope sequence is a true subsequence of the cancer neoantigen amino acid sequence.

[0466] The second linker is designed to be non-immunogenic and is preferably also a flexible linker, whereby the tumor neoepitopes are optimally presented to T cells despite the presence of a large number of antigen subunits in the antigen unit. Preferably, the length of the second linker is from 4 to 20 amino acids to ensure flexibility. In another preferred embodiment, the length of the second linker is from 8 to 20 amino acids, such as from 8 to 15 amino acids, such as from 8 to 12 amino acids, or such as for example from 10 to 15 amino acids. In a specific embodiment, the length of the second linker is 10 amino acids.

[0467] In a specific embodiment, the vaccine of the present invention comprises 10 neoepitopes, wherein the length of the second linker is from 8 to 20 amino acids, such as from 8 to 15 amino acids, such as from 8 to 12 amino acids or such as for example from 10 to 15 amino acids. In a particular embodiment, the vaccine of the present invention comprises 10 neoepitopes and wherein the second linker has a length of 10 amino acids.

[0468] The second linker is preferably a serine-glycine linker, such as a flexible GGGGS linker, such as GGGSS, GGGSG, GGGGS or multiple variants thereof, such as GGGGSGGGGS or (GGGGS)m, (GGGSS)m, (GGGSG)m, where m is an integer from 1 to 5, from 1 to 4 or from 1 to 3. In a preferred embodiment, m is 2.

[0469] Targeting unit

[0470] Due to the targeting unit, the polypeptide / dimeric protein results in the attraction of dendritic cells (DCs), neutrophils and other immune cells. Thus, the polypeptide / dimeric protein comprising the targeting module not only targets the antigen to specific cells, but also promotes response amplification (adjuvant effect) by recruiting specific immune cells to the site of administration of the vaccine. Since the vaccine itself has an adjuvant effect, this unique mechanism is very important in a clinical setting where patients can receive the vaccine without any additional adjuvant.

[0471] As used herein, the term "targeting unit" refers to a unit that delivers a polypeptide / protein together with its antigen to an antigen-presenting cell for MHC class II-restricted presentation to CD4+ T cells or for cross-presentation via MHC class I restriction to CD8+ T cells.

[0472] The targeting unit is linked to the antigen unit via a dimerization unit, where the latter is at the COOH-terminus or NH 2 -terminus of the polypeptide / dimeric protein. Preferably, the antigen unit is at the COOH-terminus of the polypeptide / dimeric protein.

[0473] The targeting unit is designed to target the polypeptide / dimeric protein of the present invention to surface molecules expressed on relevant antigen-presenting cells, such as molecules specifically expressed on subsets of dendritic cells (DCs).

[0474] Examples of such target surface molecules on APCs are human leukocyte antigen (HLA), cluster of differentiation 14 (CD14), cluster of differentiation 40 (CD40), chemokine receptors, and Toll-like receptors (TLRs). HLA is the human major histocompatibility complex (MHC). Toll-like receptors can include, for example, TLR-2, TLR-4, and / or TLR-5.

[0475] The polypeptide / dimeric protein of the present invention can be targeted to the surface molecule by means of a targeting unit comprising, for example, an antibody-binding region specific for CD14, CD40, or Toll-like receptor; a ligand, such as soluble CD40 ligand; a natural ligand-like chemokine, such as RANTES or MIP-1a; or a bacterial antigen, such as flagellin.

[0476] In one embodiment, the targeting unit has an affinity for MHC class II proteins. Thus, in one embodiment, the nucleotide sequence encoding the targeting unit encodes antibody variable domains (VL and VH) specific for MHC class II proteins selected from the group consisting of anti-HLA-DP, anti-HLA-DR, and anti-HLA-II.

[0477] In another embodiment, the targeting unit has an affinity for surface molecules selected from the group consisting of CD40, TLR-2, TLR-4, and TLR-5. Thus, in one embodiment, the nucleotide sequence encoding the targeting unit encodes antibody variable domains (VL and VH) specific for anti-CD40, anti-TLR-2, anti-TLR-4, and anti-TLR-5. In one embodiment, the nucleotide sequence encoding the targeting unit encodes flagellin. Flagellin has an affinity for TLR-5.

[0478] Preferably, the targeting unit has an affinity for chemokine receptors selected from CCR1, CCR3, and CCR5. More preferably, the nucleotide sequence encoding the targeting unit encodes the chemokine hMIP-1α (LD78β), which binds to its cognate receptors CCR1, CCR3, and CCR5 expressed on the cell surface of APCs. hMIP-1α (human MIP-1α) is also known as chemokine (C-C motif) ligand 3 (CCL3), which is encoded by the CCL3 gene in humans. CCL3, also known as macrophage inflammatory protein-1α (MIP-1α), is a cytokine belonging to the CC chemokine family and is involved in acute inflammatory states in the recruitment and activation of polymorphonuclear leukocytes. Although murine CCL3 is a single-copy gene encoding a 69-amino acid mature chemokine, the human homolog has been duplicated and mutated to produce two non-allelic variants, LD78α (CCL3) and LD78β (CCL3-L1), both of which show 74% homology to murine CCL3.

[0479] The binding of the polypeptide / dimeric protein of the present invention to its cognate receptor results in the internalization of APCs and the degradation of the protein into small peptides including the minimal epitope (and thus the mutation), which are loaded onto MHC molecules and presented to CD4+ and CD8+ T cells to induce a tumor-specific immune response. Once stimulated and with the help of activated CD4+ T cells, CD8+ T cells will target and kill tumor cells expressing the same neoantigen.

[0480] In one embodiment of the present invention, the targeting unit comprises an amino acid sequence having at least 80% sequence identity with the amino acid sequence 5-70 of SEQ ID NO:1. In a preferred embodiment, the targeting unit comprises an amino acid sequence having at least 85% sequence identity (such as at least 86%, such as at least 87%, such as at least 88%, such as at least 89%, such as at least 90%, such as at least 91%, such as at least 92%, such as at least 93%, such as at least 94%, such as at least 95%, such as at least 96%, such as at least 97%, such as at least 98%, such as at least 99% sequence identity) with the amino acid sequence 5-70 of SEQ ID NO:1.

[0481] In a more preferred embodiment, the targeting unit consists of an amino acid sequence having at least 80% sequence identity with amino acid residues 5 - 70 of SEQ ID NO:1 (e.g., having at least 85%, such as at least 86%, such as at least 87%, such as at least 88%, such as at least 89%, such as at least 90%, such as at least 91%, such as at least 92%, such as at least 93%, such as at least 94%, such as at least 95%, such as at least 96%, such as at least 97%, such as at least 98%, such as at least 99%, such as at least 100% sequence identity with amino acid residues 5 - 70 of SEQ ID NO:1).

[0482] Dimerization unit

[0483] As used herein, the term "dimerization unit" refers to the amino acid sequence between the antigen unit and the targeting unit. Thus, the dimerization unit serves to link the antigen unit and the targeting unit and to promote the dimerization of two monomer polypeptides into a dimeric protein. In addition, the dimerization unit also provides flexibility in the polypeptide / dimeric protein to allow optimal binding of the targeting unit to surface molecules on antigen-presenting cells (APCs), even if they are located at variable distances. The dimerization unit can be any unit that meets these requirements.

[0484] Thus, in one embodiment, the dimerization unit may comprise a hinge region and optionally another domain that promotes dimerization, and the hinge region and the other domain may be linked by a third linker.

[0485] The term "hinge region" refers to the peptide sequence of the dimeric protein that promotes dimerization. The hinge region serves as a flexible spacer between the units, thus allowing two targeting units to simultaneously bind to two target molecules on the APC, even if they are expressed at variable distances. The hinge region can be of Ig origin, e.g., derived from IgG3. The hinge region can promote dimerization through the formation of covalent bonds (e.g., disulfide bridges). Thus, in one embodiment, the hinge region has the ability to form one or more covalent bonds. The covalent bond can be, for example, a disulfide bridge.

[0486] In one embodiment, the another domain that promotes dimerization is an immunoglobulin domain, such as the carboxyl-terminal C domain, or a sequence substantially identical to the C domain or its variant. Preferably, the another domain that promotes dimerization is the carboxyl-terminal C domain derived from IgG.

[0487] The immunoglobulin domain promotes dimerization through non-covalent interactions (e.g., hydrophobic interactions). For example, the immunoglobulin domain has the ability to form a dimer via non-covalent interactions. Preferably, the non-covalent interaction is a hydrophobic interaction.

[0488] Preferably, the dimerization unit does not contain a CH2 domain capable of binding to the Fc receptor. In some embodiments, the dimerization unit is completely devoid of a CH2 domain. In other embodiments, the CH2 domain is mutated such that it loses its ability to bind to the Fc receptor.

[0489] In a preferred embodiment, the dimerization unit consists of hinge exons h1 and h4 connected to the CH3 domain of human IgG3 by a third linker.

[0490] CD8+ T cell response

[0491] The vaccines described herein obtained by the ranking method of the present invention can induce a shift in the immune response against a given neoepitope from a CD4+ T cell response to a CD8+ T cell response. Thus, in one aspect, a method for selecting a number A of neoepitopes for an individual is provided, the method comprising the steps of:

[0492] a. obtaining one or more neoepitopes from the individual, each neoepitope comprising at least one minimal epitope, wherein each neoepitope comprises at least one mutation, such as an immunogenic mutation, compared to a reference sequence;

[0493] b. determining the MHC I and / or MHC II binding affinity for at least one minimal epitope within each of the neoepitopes, such as at least two, three or four minimal epitopes;

[0494] c. selecting neoepitopes predicted to bind to MHC I and / or MHC II, thereby obtaining MHC-binding neoepitopes;

[0495] d. ranking the MHC-binding neoepitopes according to the likelihood of clinical utility;

[0496] e. selecting A neoepitopes from the highest-ranked MHC-binding neoepitopes, where A is an integer and A is at least 3, such as at least 4, such as at least 5,

[0497] thereby selecting A neoepitopes that are capable of inducing a CD8+ T cell response when administered as a vaccine as described herein.

[0498] In some embodiments, vaccines comprising the novel epitopes selected by the methods described herein are thus capable of inducing a CD8+ T cell response. In certain embodiments, such vaccines are capable of inducing a dominant CD8+ T cell response as compared to other vaccine formats. For example, novel epitopes that induce a CD4+ T cell response when administered as a peptide vaccine and / or as an RNA vaccine may induce a CD8+ T cell response when administered as a vaccine body DNA vaccine as described herein. In particular, novel epitopes selected by the methods of the present invention may not have been previously found to be capable of inducing a CD8+ T cell response. As shown in Example 13 and Figure 6 as shown, the inventors found that this was the case for several novel epitopes.

[0499] In some embodiments, A is an integer as further described herein. Example

[0500] Example 1 - Source, history, and generation of the VB10.NEO patient-specific host cell line

[0501] The design of the novel epitope antigen module is based on the unique tumor-specific sequences identified for each patient. To predict the most immunogenic novel epitopes, an optimized workflow was established, as shown below, as Figure 7 shown.

[0502] 1) The first step in novel epitope selection is to identify all tumor-specific mutations. Novel epitopes that are not expressed in the tumor or that may, for example, contain a peptide of 9 amino acids (including the mutation site) with an identical sequence match elsewhere in the proteome are excluded or downranked. In addition, novel epitopes in genes that show at least a 5-fold higher RNA expression level in a particular organ / tissue compared to all other tissues are also excluded or downranked.

[0503] 2) The next step is to construct the novel epitopes (27 amino acids long with the mutation in the middle) and rank these novel epitopes based on the best combination of peptide binding affinity for HLA molecules and the nature of their residues.

[0504] 3) In some embodiments, the completion of the novel epitope set involves evaluating the novel epitopes based on carefully selected evaluation criteria consisting of the properties of the novel epitopes and / or their source genes that have been reported to be potentially important for defining immunogenic novel epitopes. This evaluation is performed by a target selection board consisting of a clinician (CMO), an immunologist (CSO), and a bioinformatician.

[0505] Example 2 - Collection and processing of patient tissues

[0506] Solid tumor tissue as well as blood samples are collected from eligible patients during screening.

[0507] Collect at least one core biopsy for whole-exome sequencing and RNA sequencing. For whole-exome and transcriptome sequencing, cryopreservation is the ideal choice for maintaining specimen integrity, so this material is preferably used for sequencing purposes. If fresh tumor material is not available at the time of screening, tumor material stored in FFPE prior to screening is acceptable for sequencing.

[0508] Collect blood samples. For cryopreserved tumor biopsy samples, immerse the samples in liquid nitrogen for rapid freezing and store in liquid nitrogen. For FFPE tumor material, fix the tissue samples in 4% to 10% neutral buffered formalin. Then dehydrate the samples, then embed and store at room temperature.

[0509] Example 3 - Patient exome and RNA sequencing and HLA typing

[0510] Whole-exome and RNA sequencing data can be obtained from blood, tumor, and the patient's HLA type. Sequence the tumor and blood samples using a read length of at least 2x100bp. The raw data output is at least 24Gb.

[0511] RNA sequencing is performed on tumor samples with a read length of at least 2x100 bp. The total data output is at least 100 million reads.

[0512] For HLA typing, sequence the DNA isolated from blood samples on an Illumina platform with a read length of at least 2x150 bp and a coverage of more than 100x.

[0513] Example 4 - Bioinformatics processing and identification of mutations in tumors

[0514] The whole-exome and RNA expression data are provided as FASTQ files and can be downloaded from the internal secure server of the sequencing provider. The analysis of the raw whole-exome sequences (FASTQ files) passing QC follows best practice workflows defined by state-of-the-art techniques and optimized algorithms for fastq processing and variant calling for tumor and normal sample pairs (e.g., Miller et al., 2015; Van der Auwera et al., 2013).

[0515] The raw FASTQ files for RNA sequencing are mapped using state-of-the-art methods in RNA sequencing analysis (e.g., Dobin et al., 2013; Dutton G. 2016).

[0516] Identify the patient's HLA alleles using an Illumina sequencing platform from EDTA blood samples.

[0517] All mutations found in protein-coding genes with detected RNA expression in tumor tissue (variants identified in vcf files) (TPM > 0, as defined in Gubin et al., 2015) were studied for their potential utility as neoepitopes.

[0518] Peptide sequences spanning 13 amino acids on each side of the mutation were extracted to form neoepitope sequences with a total length of 27 amino acids. For nonsynonymous mutations that are closer than 13 amino acids to the C-terminus or N-terminus of the protein, one side of the mutant flanking sequence is shorter than 13 amino acids, and thus the total length of the neoepitope sequence will be shorter than 27 amino acids.

[0519] Example 5 - Exclusion criteria for minimizing autoimmune risk

[0520] To minimize the risk of cross-reactivity, preferably all of the following neoepitopes are excluded or de-prioritized at this stage: the neoepitope has a core peptide of 9 amino acids (including the mutation) that matches any peptide sequence in the normal human proteome. In addition, the following neoepitopes in the gene are also excluded or de-prioritized: the neoepitope shows an RNA expression level at least 5-fold higher in a specific tissue or organ compared to all other tissues or organs (as defined in the Human Proteome Atlas, Uhlén et al., 2015).

[0521] Example 6 - Development of a computational model for prioritizing new epitopes in a murine tumor model

[0522] To identify the most immunogenic neoepitopes, data and results from two in vivo mouse tumor models were used to develop a computational model for predicting neoepitope prioritization.

[0523] Using multiple datasets from preclinical experiments in mice, features related to predicting immunogenic neoepitopes were collected and their predictive potential was evaluated. A strategy for prioritizing neoepitopes based on predicted immunogenicity was developed using neoepitopes identified in the CT26 colon cancer model and their observed immunogenicity in BALB / c mice (H-2d) vaccinated with VB10.NEO. The predictive ability of the strategy was validated using immunogenicity data collected for neoepitopes identified in the B16 melanoma model and the LL2 lung cancer model, and the immunogenicity observed in C57Bl / 6 mice (H-2b) vaccinated with VB10.NEO. A neoepitope was classified as immunogenic if the number of IFN-γ analyzed by an internally conducted IFN-γ ELISpot assay > negative control + 2xSD and the number of spots > 25. The following showed different patterns between immunogenic and non-immunogenic neoepitopes: the binding affinity (% rank) for MHC class I and II molecules using the prediction tools NetMHCpan (Nielsen et al., 2016) and NetMHCIIpan (Andreatta et al., 2015), the total number of binding minimal epitopes, the difference in binding affinity between mutant and wild-type epitopes (binding difference), and the BLOcks Substitution Matrix (BLOSUM) score (Henikoff et al., 1992).

[0524] Example 7 - Binding affinity (% rank), total number of binding epitopes, binding difference, and BLOSUM score

[0525] In mice, only peptides with affinity for MHC class I or II (in human HLA class I or II) can provide eligible T cell targets. One strategy for selecting vaccine targets is to use NetMHCpan and NetMHCIIpan to select candidate epitopes based on the predicted affinity of candidate neoepitopes for MHC class I and II molecules, respectively. For all 27-amino acid neoepitopes, these servers provided binding affinity scores for the selected MHC molecules. The binding affinity prediction servers provided multiple MHC affinity score values, among which the % rank was highly recommended (Nielsen et al., 2016 and Andreatta et al., 2015). A low % rank score indicates strong binding affinity.

[0526] Figure 2 Illustrated is the % rank distribution for MHC class I and MHC class II of immunogenic and non-immunogenic neoepitopes tested in mice vaccinated with VB10.NEO.

[0527] Among the 27 - amino - acid - long neo - epitopes, there may be more than one minimal epitope (8 - 14 amino acids for class I MHC and 9 - 15 amino acids for class II MHC) predicted to bind to the relevant MHC molecules. The neo - epitope containing the highest number of predicted minimal epitopes increases the chance of processing and presenting one or more immunogenic peptides on the MHC molecules of the patient, thus triggering a more effective tumor - specific immune response. Figure 8 Comparison of the total number of minimal epitopes between immunogenic and non - immunogenic neo - epitopes in B16 and CT26 data is shown.

[0528] If the binding affinity of a neo - epitope to an MHC molecule is higher than that of the corresponding wild - type sequence (high binding difference), then the wild - type is likely to be rarely presented or not presented in healthy tissues, and thus, neo - epitope - specific T cells should have a low risk of recognizing healthy cells and a high potential of recognizing cancer cells expressing the neo - epitope. To measure the binding difference between a reference sequence and a mutant neo - epitope sequence, the ratio of their % rank values (reference / mutant) can be used and grouped by residue type: anchor (P2 or P8 / P9) and non - anchor (all other residue positions). Figure 9 Illustrates that if an epitope has a mutation at the anchor position and a high binding difference, it has a higher chance of being immunogenic.

[0529] The BLOSUM score describes the likelihood of a pairwise amino - acid substitution occurring. In our case, the amino acids found in the proteins expressed in healthy tissues have been replaced by mutant amino acids. The lower the score, the less likely such a substitution occurs in the alignment of the relevant proteins. The BLOSUM scores of neo - epitopes in the CT26 melanoma model dataset with high binding affinity are generally lower for immunogenic neo - epitopes ( Figure 10 ).

[0530] To predict the most immunogenic neo - epitopes, a combination of all five factors is included: the % rank values of MHC I and MHC II, the number of minimal epitopes within the neo - epitope, the binding difference at the anchor positions, and the BLOSUM score. The prioritization strategy for neo - epitopes is called NeoSELECT, which weighs these five factors and generates a ranked list of neo - epitopes. The NeoSELECT ranking was tested on neo - epitopes from LL2, CT26, and B16 tumor models, and their observed immunogenicity is as Figure 3 shown.

[0531] For (A) CT26, (B) B16, and (C) LL2 models, Figure 4Shown are the observed cross-reactivities of the top 20 neoepitopes predicted by the NeoSELECT strategy against reference peptide sequences and responses to mutant peptide sequences. For the CT26 model, cross-reactivity measurements were obtained for 12 out of 20 neoepitopes; for the B26 model, cross-reactivity measurements were obtained for 17 out of 20 neoepitopes. For the LL2 model, cross-reactivity measurements were obtained for all 20 neoepitopes. Thus, using the NeoSELECT strategy, the risk of inducing cross-reactive T cells is limited.

[0532] Example 8 - New epitope prediction method used in the VB N-01 clinical trial

[0533] If more than 20 potential neoepitopes are found for a patient, these neoepitopes are ranked according to the NeoSELECT prioritization strategy for neoepitope prediction described above. For patient data, patient-specific HLA alleles were used in the process of obtaining % rank values for reference and WT neoepitopes from NetMHCpan and NetMHCIIpan. Binding differences were calculated and anchor positions were determined. The complete list of predicted minimal epitopes for all HLA alleles was used to calculate the number of minimal epitopes for MHC class I and II. The BLOSUM score for each mutation was extracted from the BLOSUM score matrix. These factors were then used in the NeoSELECT strategy to obtain a ranked list of predicted immunogenic neoepitopes. Binding affinities measured by % rank, binding difference, number of minimal epitopes, and BLOSUM score are standard values and are not affected by the organism in which the neoepitopes are identified. Thus, the prediction model developed based on mouse data can be used for neoepitope prediction in humans without additional customization.

[0534] Example 9 - Quality control and selection of final new epitopes

[0535] Finally, additional information on all neoepitopes passing these exclusion criteria was collected as selection criteria. These criteria include:

[0536] · Clonality (i.e., prioritizing neoepitopes found in more than one biopsy sample from a patient, such as neoepitopes found in archival tumor material, in multi-angle biopsies of the same lesion, or in cell-free DNA)

[0537] · High RNA expression level (in TPM)

[0538] · Low % rank of neoepitopes for MHC class I and II (high binding affinity)

[0539] · High binding difference (% rank of neoepitopes relative to reference)

[0540] · Low BLOSUM score: residue characteristics of substituted amino acids in neoepitopes compared to wild type

[0541] ·The higher allele frequency (AF) or variant allele frequency (VAF) at the mutation position estimated from the VCF file

[0542] ·In the Catalogue Of Somatic Mutations In Cancer (COSMIC) database, the source gene is a known cancer-related gene

[0543] ·The new epitope has a high similarity to an epitope known to be recognized by T cells (e.g., as obtained from the Immune Epitope Database and Analysis Resource (IEDB database))

[0544] ·A large number of minimal epitopes (8 - 15 amino acids) in the new epitope of 27 amino acids

[0545] ·The position of the mutation in the HLA-TCR complex, as described in Fritsch et al., 2014

[0546] ·A high binding stability score to MHC class I / MHC class II molecules estimated using NetMHCstabpan (Rasmussen et al., 2016) or a similar method

[0547] ·A high proteasome cleavage score predicted by NetChop (Nielsen et al., 2005) or a similar method.

[0548] These criteria, in addition to the prioritization predicted according to the NeoSELECT strategy, also form the basis for selecting the final set of 10 to 20 new epitopes and are evaluated by the target selection committee. The decision criteria are recorded, and the immune response of the patient to each new epitope in the vaccine and the patient's clinical response are analyzed.

[0549] The recommended set of new epitopes is combined and separated by a flexible non-immunogenic glycine / serine-rich linker (see Example 11). The order of the new epitopes depends on the linker sequence between each new epitope and the linker, where all linker sequences of 9 amino acids that have the same match elsewhere in the proteome are avoided.

[0550] Example 10 - Data storage and analysis

[0551] The storage of the original data and the results of exome and RNA sequence analysis are all performed in a secure and controlled computer cluster environment platform that meets the requirements for processing and storing patient-sensitive data. All patients are assigned a unique number at the time of registration to protect the patient's identity.

[0552] Example 11 - Synthesis of new epitope antigen modules

[0553] The VaxiBead company designed a patient-specific neoepitope antigen module based on 10 - 20 27-amino acid long neoepitopes linked to a 10-amino acid long glycine / serine-rich linker. The neoepitope antigen module was de novo synthesized by a DNA synthesis provider and cloned into a plasmid to generate VB10.NEO.

[0554] Example 12 – Compared with other vaccine forms, the vaccine body can induce a strong dominant CD8+ T cell response

[0555] Kreiter et al. in 2015 and Castle et al. in 2012 studied 10 different neoepitopes (pep1 - 10), all of which were predicted to bind to MHC class I (CD8+ T cell response), to investigate whether they could induce a CD8+ T cell response in a murine B16-F10 melanoma tumor model. When administered as peptides, RNA, or the vaccine bodies (“VB10.NEO”) described herein, the responses induced by 6 out of the 10 neoepitopes are shown in Figure 6 in.

[0556] When the neoepitopes were administered as synthetic peptides, one out of six induced a CD8+ T cell response; when the neoepitopes were administered as mRNA, two out of six induced a CD8+ T cell response.

[0557] However, when the same 6 neoepitopes were delivered as vaccine bodies targeting human MIP1α, all neoepitopes induced a CD8+ T cell response, clearly demonstrating the importance of the vaccine format for inducing a strong CD8+ cytotoxic T cell response against neoepitopes.

[0558] Example 13 – Immunogenicity of the selected new epitopes

[0559] The described method was used to select neoepitopes for 4 patients with renal cell carcinoma or squamous cell carcinoma of the head and neck. Personalized vaccines were constructed and administered to the patients. Immunogenicity was measured in peripheral blood mononuclear cells harvested 3 to 9 months after the first dose of the personalized vaccine. T cell responses (subtracting background) were measured in a 10-day pre-stimulated ELISPOT assay. Vaccine-induced response: response with >30 SFU increase after vaccination. Immunogenic neoepitope: >30 SFU at at least one time point.

[0560]

[0561] RCC: Renal cell carcinoma; SCCHN: Squamous cell carcinoma of the head and neck.

[0562] On average, in these 4 patients, 95% of the neoepitopes selected by this method were immunogenic, that is, capable of activating T cell responses in patients bearing the corresponding mutant sequences in their tumors. When the selected neoepitopes were incorporated into a personalized vaccine and administered to the patients, increased T cell responses were induced against an average of 78% of the neoepitopes compared to before vaccination.

[0563] This example shows that the method of selecting neoepitopes results in the selection of neoepitopes that can increase T cell responses when administered to patients.

[0564] Sequence

[0565] SEQ ID NO:1

[0566] C-C motif chemokine 3-like 1 precursor (LD78-β) including signal peptide and mature peptide, aa24-93:

[0567]

[0568] References

[0569] Andreatta, M. et al. (2015) “Accurate pan-specific prediction of peptide-MHC class II binding affinity with improved binding core identification.” Immunogenetics. 2015. 67(11-12):641-50.

[0570] Castle, J.C. et al. (2012) "Exploiting the Mutanome for Tumor Vaccination." Cancer Res; 72(5):1081-91.

[0571] Dobin, A. et al. (2013) “STAR: ultrafast universal RNA-seq aligner.” Bioinformatics. 1; 29(1):15-21.

[0572] Dutton, G. (2016) “From DNA to Diagnosis without Delay. Purpose-Built for Genomics, Dragen Processor Could Form Core of Clinic-Ready Data Systems.” Genet Eng Biotechn N. 36(5):8-9.

[0573] Fritsch, E. F. et al. (2014) “HLA-binding properties of tumor neoepitopes in humans.” Cancer Immunol Res.; 2(6):522-9.

[0574] Gubin, M. M. et al. (2015) “Tumor neoantigens: building a framework for personalized cancer immunotherapy.” J. Clin. Invest. 125(9):3413-21.

[0575] Henikoff, S. et al. (1992) “Amino acid substitution matrices from protein blocks.” Proc Natl Acad Sci U S A.; 89(22):10915–10919.

[0576] Jurtz, V. et al. (2017) “NetMHCpan-4.0: Improved Peptide–MHC Class I Interaction Predictions Integrating Eluted Ligand and Peptide Binding Affinity Data.” Vanessa Jurtz, Sinu Paul, Massimo Andreatta, Paolo Marcatili, Bjoern Peters and Morten Nielsen. The Journal of Immunology(2017))

[0577] Kreiter S et al. Mutant MHC class II epitopes drive therapeutic immune responses to cancer. Nature. 2015 Apr 30;520(7549):692-6

[0578] Kreiter, S. et al. (2015) "Erratum: Mutant MHC Class 2 Epitopes Drive Therapeutic Immune Responses To Cancer." Nature 523.7560:370-370.

[0579] Miller, N.A. et al. (2015) "26-hour system of highly sensitive whole genome sequencing for emergency management of genetic diseases". Genome Med. 7:100.

[0580] Nielsen, M. et al. (2005) “The role of the proteasome in generating cytotoxic T cell epitopes: Insights obtained from improved predictions of proteasomal cleavage.” Immunogenetics., 57(1-2):33-41.

[0581] Nielsen, M. et al. (2016) “NetMHCpan-3.0: improved prediction of binding to MHC class I molecules integrating information from multiple receptor and peptide length data sets.” Genome Medicine: 8:33.

[0582] Rasmussen, M. (2016) “Pan-Specific Prediction of Peptide-MHC Class I Complex Stability, a Correlate of T Cell Immunogenicity.” J Immunol.; 197(4):1517-24.

[0583] Uhlén, M. et al. (2015) “Proteomics. Tissue-based map of the human proteome.” Science; 347(6220)

[0584] Van der Auwera G. A. et al. (2013) ”From FastQ data to high confidence variant calls: the Genome Analysis Toolkit best practices pipeline.” Curr Protoc Bioinformatics. 43.

[0585] Project

[0586] 1. A method for selecting a number A of new epitopes for an individual, the method comprising the steps of:

[0587] a. obtaining one or more new epitopes from the individual, each new epitope comprising at least one minimal epitope, wherein each new epitope comprises at least one mutation, such as an immunogenic mutation, compared to a reference sequence;

[0588] b. determining MHC I and / or MHC II binding affinities for at least one minimal epitope, such as at least two, three or four minimal epitopes, within each of the new epitopes;

[0589] c. selecting new epitopes comprising at least one minimal epitope predicted to bind to MHC I and / or MHC II, thereby obtaining new epitopes that bind to MHC;

[0590] d. ranking the new epitopes that bind to MHC according to the likelihood of clinical utility;

[0591] e. selecting A new epitopes from among the highest-ranked new epitopes that bind to MHC,

[0592] thereby selecting A new epitopes that may have clinical utility.

[0593] 2. The method according to item 1, wherein some or all of the A new epitopes bind to at least MHC I.

[0594] 3. The method according to any one of the preceding items, wherein the at least one minimal epitope is 1 to 50 minimal epitopes, such as 1 to 40 minimal epitopes, such as 1 to 30 minimal epitopes, such as 1 to 20 minimal epitopes.

[0595] 4. The method according to any one of the preceding items, wherein each minimal epitope contains a mutation.

[0596] 5. The method according to any one of the preceding items, wherein step b is carried out by: determining the MHC I and / or MHC II binding affinity of each of the minimal epitopes, thereby identifying x minimal epitopes that bind to MHC I and / or y minimal epitopes that bind to MHC II.

[0597] 6. The method according to any one of the preceding items, wherein the MHC I and MHC II binding affinity of each of the new epitopes is calculated as the highest MHC I and MHC II binding affinity of the minimal epitopes contained within each of the new epitopes.

[0598] 7. The method according to any one of the preceding items, wherein step c comprises the following steps or consists of: selecting new epitopes that contain at least one minimal epitope predicted to bind to MHC I and / or at least one minimal epitope predicted to bind to MHC II, thereby obtaining new epitopes that bind to MHC.

[0599] 8. The method according to any one of the preceding items, wherein step c comprises selecting new epitopes that contain at least one minimal epitope predicted to bind to MHC I and at least one minimal epitope predicted to bind to MHC II.

[0600] 9. The method according to any one of the preceding items, wherein step d comprises the step of determining the number of minimal epitopes that bind to MHC I and / or MHC II within each of the new epitopes, and wherein a higher number of minimal epitopes that bind to MHC I and / or MHC II indicates a higher likelihood of clinical utility of the new epitope.

[0601] 10. The method according to any one of the preceding items, wherein step d comprises the step of determining the binding score of the new epitope, wherein the binding score is a*2x + b*y, wherein preferably a = 0 if y = 0 and b = 0 if x = 0, and wherein a = 1 if y > 0 and b = 1 if x > 0, and wherein a higher binding score indicates a higher likelihood of clinical utility of the new epitope.

[0602] 11. The method according to any one of the preceding items, wherein step d includes the step of determining a binding score of the new epitope, wherein the binding score is a*x + b*y, wherein preferably a = 2 and b = 1, or a = 0 if y = 0 and b = 0 if x = 0, and wherein a = 1 if y > 0 and b = 1 if x > 0, and wherein a higher binding score indicates a higher likelihood of clinical utility of the new epitope.

[0603] 12. The method according to any one of the preceding items, wherein step d further includes the step of determining an MHC I binding difference of the new epitope that binds to MHC I in step c, wherein the MHC I binding difference is given by: (reference % ranking score (MHC I)) / (% ranking score of the new epitope (MHC I)), wherein a new epitope with a high MHC I binding difference ranks higher than a new epitope with a lower MHC I binding difference.

[0604] 13. The method according to any one of the preceding items, wherein step d includes the step of determining an MHC I binding difference of the minimal epitope that binds to MHC I in step c, wherein the MHC I binding difference is given by: (reference % ranking score (MHC I)) / (% ranking score of the minimal epitope (MHC I)), wherein a new epitope containing a minimal epitope with a high MHC I binding difference ranks higher than a new epitope containing a minimal epitope with a lower MHC I binding difference.

[0605] 14. The method according to any one of the preceding items, wherein the MHC I binding difference of each new epitope is equal to the maximum MHC I binding difference of the minimal epitopes it contains.

[0606] 15. The method according to any one of the preceding items, wherein step d further includes the step of determining an MHC II binding difference of the new epitope that binds to MHC II in step c, wherein the MHC II binding difference is given by: (reference % ranking score (MHC II)) / (% ranking score of the new epitope (MHC II)), wherein a new epitope with a high MHC II binding difference ranks higher than a new epitope with a lower MHC II binding difference.

[0607] 16. The method according to any one of the preceding items, wherein step d includes the step of determining an MHC II binding difference of the minimal epitope that binds to MHC II in step c, wherein the MHC II binding difference is given by: (reference % ranking score (MHC II)) / (% ranking score of the minimal epitope (MHC II)), wherein a new epitope containing a minimal epitope with a high MHC II binding difference ranks higher than a new epitope containing a minimal epitope with a lower MHC II binding difference.

[0608] 17. The method according to any one of the preceding items, wherein the MHC II binding difference of each new epitope is equal to the highest MHC II binding difference of the smallest epitope it contains.

[0609] 18. The method according to any one of the preceding items, wherein the new epitope selected in step c binds to MHC II.

[0610] 19. The method according to any one of the preceding items, wherein the new epitope that binds to MHC I is selected in step e only if the plurality of new epitopes do not contain any new epitopes that bind to MHC II.

[0611] 20. The method according to any one of the preceding items, wherein step d further comprises ranking the new epitopes as follows:

[0612] i) For each smallest epitope contained in the new epitope, determine whether the mutation is at an anchor position or a non-anchor position of the smallest new epitope;

[0613] ii) Prioritize the new epitopes, preferably where the new epitopes with higher binding differences are prioritized over the new epitopes with lower binding differences.

[0614] 21. The method according to item 20, wherein the prioritization of the new epitopes in step ii) is performed by assigning them the highest score of the smallest epitope they contain.

[0615] 22. The method according to any one of items 20 or 21, wherein the score of the new epitope with a higher binding difference is higher than the score of the new epitope with a lower binding difference, and wherein the score of the new epitope containing a mutation at an anchor position is higher than the score of the new epitope containing a mutation at a non-anchor position.

[0616] 23. The method according to any one of items 19 to 21, wherein the prioritization of the new epitopes is performed as follows:

[0617] 1) If the mutation is at an anchor position and the binding difference of the smallest epitope is equal to or greater than 20, the score of the smallest epitope is the highest;

[0618] 2) If the mutation is at a non-anchor position and the binding difference of the smallest epitope is equal to or greater than 20, the score of the smallest epitope is lower than the score of the smallest epitope in 1);

[0619] 3) If the mutation is at an anchor position and the binding difference of the smallest epitope is less than 20 and equal to or greater than 3, the score of the smallest epitope is lower than the score of the smallest epitope in 2);

[0620] 4) If the mutation is in a non-anchoring position and the binding difference of the minimal epitope is less than 20 and equal to or greater than 3, the score of the minimal epitope is lower than the score of the minimal epitope in 3);

[0621] 5) If the mutation is in an anchoring position and the binding difference of the minimal epitope is less than 3 and equal to or greater than 1, the score of the minimal epitope is lower than the score of the minimal epitope in 4);

[0622] 6) If the mutation is in a non-anchoring position and the binding difference of the minimal epitope is less than 3 and equal to or greater than 1, the score of the minimal epitope is lower than the score of the minimal epitope in 5);

[0623] 7) If the minimal epitope has a mutation with a binding difference lower than 1, the score of the minimal epitope is the lowest, regardless of the position of the mutation.

[0624] 24. The method according to any one of the preceding items, wherein step d further comprises the step of prioritizing the new epitopes according to the MHC I ranking of the new epitopes.

[0625] 25. The method according to item 24, wherein the % ranking score (MHC I) of a new epitope comprising one or more minimal epitopes predicted to bind MHC I is equal to the lowest % ranking score (MHC I) of said one or more minimal epitopes.

[0626] 26. The method according to any one of the preceding items, wherein step e comprises selecting new epitopes having a % ranking score (MHC I) lower than 2.0, such as lower than 1.5, such as lower than 1, preferably equal to or lower than 0.5.

[0627] 27. The method according to any one of the preceding items, wherein new epitopes that bind MHC I with a % ranking score (MHC I) higher than 2 are excluded or have their prioritization reduced.

[0628] 28. The method according to any one of the preceding items, wherein step d further comprises the step of prioritizing the new epitopes according to the MHC II ranking of the new epitopes.

[0629] 29. The method according to any one of items 24 to 28, wherein the % ranking score (MHC II) of a new epitope comprising one or more minimal epitopes predicted to bind MHC II is equal to the lowest % ranking score (MHC II) of said one or more minimal epitopes.

[0630] 30. The method according to any one of the preceding items, wherein step e comprises selecting new epitopes having a % ranking score (MHC II) lower than 10, such as lower than 2.

[0631] 31. The method according to any one of the preceding items, wherein a novel epitope that binds to MHC I and has a ranking score (MHC I) higher than 10, for example higher than 2, is excluded or its prioritization is reduced.

[0632] 32. The method according to any one of the preceding items, wherein step d further comprises the step of prioritizing the novel epitopes in descending order according to the BLOSUM score of the novel epitopes, wherein the BLOSUM score of a novel epitope is equal to the BLOSUM score of the best-ranked minimal epitope it contains, and wherein a minimal epitope with a BLOSUM score < 1 is ranked higher than a minimal epitope with a BLOSUM score ≥ 1.

[0633] 33. The method according to any one of the preceding items, wherein step d according to item 1 further comprises determining the probability that the amino acid substitutions present in the novel epitope occur randomly, and wherein a novel epitope containing amino acid substitutions that occur randomly with a low probability is ranked higher than a novel epitope containing amino acid substitutions that occur randomly with a higher probability.

[0634] 34. The method according to item 33, wherein the probability that the amino acid substitutions present in the novel epitope occur randomly is determined using an evolution-based scoring matrix.

[0635] 35. The method according to item 34, wherein the scoring matrix is a log-odds matrix.

[0636] 36. The method according to item 35, wherein the log-odds matrix is a BLOSUM matrix.

[0637] 37. The method according to item 36, wherein the BLOSUM matrix is a BLOSUM62 matrix.

[0638] 38. The method according to item 37, wherein a novel epitope that does not have a mutation associated with an amino acid substitution pair with a BLOSUM62 score lower than 1 is excluded or its prioritization is reduced.

[0639] 39. The method according to any one of the preceding items, further comprising determining the RNA expression level of the novel epitope, wherein a novel epitope with undetectable RNA expression is excluded or its prioritization is reduced.

[0640] 40. The method according to item 39, wherein a novel epitope with a high RNA expression level is ranked higher than a novel epitope with a lower RNA expression level.

[0641] 41. The method according to any one of items 39 or 40, wherein the RNA expression level is determined for novel epitopes that bind to MHCI and / or bind to MHC II.

[0642] 42. The method according to any one of the preceding items further comprises the step of comparing the new epitope peptide sequence with the peptide sequences of the human proteome, wherein such new epitopes are excluded or their prioritization is reduced: the new epitopes contain peptide sequences that match the peptide sequences in the human proteome.

[0643] 43. The method according to item 42, wherein the new epitope peptide contains 5 to 15 amino acids or consists of 5 to 15 amino acids.

[0644] 44. The method according to any one of the preceding items further comprises the step of identifying the plurality of new epitopes by identifying one or more tumor-specific mutations in the individual.

[0645] 45. The method according to item 44, wherein the one or more tumor-specific mutations are identified by exome sequencing of the tumor DNA from the individual.

[0646] 46. The method according to any one of items 44 to 45, wherein the one or more tumor-specific mutations result in amino acid substitutions.

[0647] 47. The method according to any one of the preceding items, wherein the MHC binding affinity includes determining the HLA genotype of the individual.

[0648] 48. The method according to any one of the preceding items, wherein the HLA genotype is determined from a blood sample of the individual.

[0649] 49. The method according to any one of the preceding items, wherein the MHC I and / or MHC II binding affinity is determined by computer prediction.

[0650] 50. The method according to item 49, wherein the computer prediction is performed by using a computer program that predicts the binding of peptides to MHC class I and / or MHC class II molecules.

[0651] 51. The method according to any one of the preceding items excludes or reduces the prioritization of new epitopes present in the following genes: wherein the genes show an RNA expression level at least 5 times higher in a given organ or tissue compared to other organs or tissues.

[0652] 52. The method according to item 51, wherein the organ is selected from the heart and the brain.

[0653] 53. The method according to item 52, wherein the organ is selected from the liver, lung, stomach, kidney, spleen, colon, and intestine.

[0654] 54. The method according to any one of the preceding items, wherein step d further comprises the step of determining the allelic frequency of the mutations present in the new epitopes that bind to MHC, wherein the new epitopes that bind to MHC with a high allelic frequency are ranked higher than the new epitopes that bind to MHC with a lower allelic frequency.

[0655] 55. The method according to any one of the preceding items, wherein the new epitopes that bind to MHC and contain mutations found in more than one biopsy are ranked higher than the new epitopes that bind to MHC and contain mutations found in only one biopsy.

[0656] 56. The method according to any one of the preceding items, wherein the length of the new epitope is 7 to 40 amino acids.

[0657] 57. The method according to any one of the preceding items, wherein the length of the new epitope is 15 to 30 amino acids.

[0658] 58. The method according to any one of the preceding items, wherein the length of the new epitope is 25 to 30 amino acids, such as 27 amino acids.

[0659] 59. The method according to any one of the preceding items, wherein the mutation is substantially located in the middle of the new epitope sequence.

[0660] 60. The method according to any one of the preceding items, wherein the individual is a cancer patient.

[0661] 61. The method according to any one of the preceding items, wherein the new epitopes selected for the individual are used in a cancer vaccine.

[0662] 62. The method according to any one of the preceding items, wherein A is an integer from 1 to 100, such as from 5 to 50, such as from 5 to 30.

[0663] 63. The method according to any one of the preceding items, wherein A is an integer from 10 to 20.

[0664] 64. The method according to any one of the preceding items, wherein x > y.

[0665] 65. The method according to any one of the preceding items, wherein x ≥ 2y.

[0666] 66. The method according to any one of the preceding items, wherein x > 2.5y.

[0667] 67. The method according to any one of the preceding items, wherein A ≥ 3.

[0668] 68. The method according to any one of the preceding items, wherein at least some of the A new epitopes, such as A / 2 of the A new epitopes, such as A / 2 to A of the A new epitopes, are capable of inducing a CD8+ T cell response.

[0669] 69. The method according to any one of the preceding items, the method comprising the steps of:

[0670] a. obtaining a plurality of new epitopes from the individual, each new epitope comprising at least 3 minimal epitopes, wherein each new epitope comprises at least one immunogenic mutation compared to a reference sequence;

[0671] b. determining the MHC I and MHC II binding affinities of each of the new epitopes, thereby identifying x new epitopes that bind MHC I and y new epitopes that bind MHC II, wherein x + y ≥ 3;

[0672] c. selecting new epitopes that are predicted to bind at least MHC I and optionally MHC II, thereby obtaining new epitopes that bind MHC;

[0673] d. ranking the new epitopes that bind MHC according to the likelihood of clinical utility; wherein the ranking is performed as follows:

[0674] i. ranking the new epitopes according to the number of minimal epitopes that bind MHC I and optionally bind MHC II comprised in the new epitopes, wherein a higher number gives a higher ranking;

[0675] e. selecting A new epitopes from the new epitopes that bind MHC and are ranked highest,

[0676] thereby selecting A new epitopes that may have clinical utility.

[0677] 70. The method according to item 69, wherein the number of minimal epitopes that bind MHC II comprised in the new epitopes is more than the number of minimal epitopes that bind MHC II comprised in the new epitopes, such as more than twice as many.

[0678] 71. The method according to any one of items 69 to 70, wherein step ii. follows step i. of step d, wherein:

[0679] ii. Determine the MHC I and / or MHC II binding difference scores for the new epitopes, where the MHC I binding difference is given by: (reference % rank score (MHC I)) / (new epitope's % rank score (MHC I)), and the MHC II binding difference is given by: (reference % rank score (MHC II)) / (new epitope's % rank score (MHC II)); where new epitopes with high binding differences are ranked higher than new epitopes with lower binding differences.

[0680] 72. The method according to item 70, wherein the new epitopes are ranked in descending order as follows:

[0681] 1) If the mutation is at an anchor position and the binding difference of the minimal epitope is equal to or greater than 20, the minimal epitope has the highest score;

[0682] 2) If the mutation is at a non - anchor position and the binding difference of the minimal epitope is equal to or greater than 20, the minimal epitope has a score lower than the minimal epitope in 1);

[0683] 3) If the mutation is at an anchor position and the binding difference of the minimal epitope is less than 20 and equal to or greater than 3, the minimal epitope has a score lower than the minimal epitope in 2);

[0684] 4) If the mutation is at a non - anchor position and the binding difference of the minimal epitope is less than 20 and equal to or greater than 3, the minimal epitope has a score lower than the minimal epitope in 3);

[0685] 5) If the mutation is at an anchor position and the binding difference of the minimal epitope is less than 3 and equal to or greater than 1, the minimal epitope has a score lower than the minimal epitope in 4);

[0686] 6) If the mutation is at a non - anchor position and the binding difference of the minimal epitope is less than 3 and equal to or greater than 1, the minimal epitope has a score lower than the minimal epitope in 5);

[0687] 7) If the minimal epitope has a mutation with a binding difference less than 1, the minimal epitope has the lowest score, regardless of the position of the mutation.

[0688] 73. The method according to any one of items 69 to 72, wherein step iii. follows step i. or step ii. of step d, where:

[0689] iii. Rank the new epitopes according to their % rank scores, where new epitopes with low % rank scores are ranked higher than new epitopes with high % rank scores.

[0690] 74. The method according to item 73, wherein the % rank score of a new epitope is equal to the lowest % rank score of the minimal epitope it contains.

[0691] 75. According to the method according to any one of items 73 to 74, wherein the new epitope or minimal epitope is predicted as:

[0692] - If the % rank score (MHC I) ≤ 0.5, it is a strong MHC I binder;

[0693] - If 0.5 < % rank score (MHC I) ≤ 2, it is a weak MHC I binder;

[0694] - If the % rank score (MHC I) > 2, it is a non-MHC I binder.

[0695] 76. According to the method according to any one of items 73 to 75, wherein the new epitope or minimal epitope is predicted as:

[0696] - If the % rank score (MHC II) ≤ 2, it is a strong MHC I binder;

[0697] - If 2 < % rank score (MHC II) ≤ 10, it is a weak MHC I binder;

[0698] - If the % rank score (MHC II) > 10, it is a non-MHC I binder.

[0699] 77. According to the method according to any one of items 69 to 76, wherein after step i., ii. or iii. is step iv, wherein:

[0700] iv. Score the new epitope according to the BLOSUM score of the new epitope, wherein a lower BLOSUM score gives a higher score.

[0701] 78. According to the method according to any one of items 69 to 77, wherein new epitopes with a BLOSUM score < 3, such as a BLOSUM score < 2, such as a BLOSUM score < 1, such as a BLOSUM score < 0 are prioritized.

[0702] 79. A method for selecting a number A of new epitopes for an individual, the method comprising the following steps:

[0703] a. Obtain one or more new epitopes from the individual, each new epitope comprising at least one minimal epitope, wherein each new epitope comprises at least one mutation, such as an immunogenic mutation, compared to a reference sequence;

[0704] b. Determine the MHC I binding affinity of at least one minimal epitope (e.g., at least two, three or four minimal epitopes) within each of the new epitopes, and determine the number of minimal epitopes that bind MHC I for each of the new epitopes;

[0705] c. Rank the new epitopes as follows:

[0706] i. Prioritize the new epitopes that contain the smallest epitopes that bind MHC I in large numbers, and select the first group of new epitopes with high scores;

[0707] ii. Optionally, prioritize the new epitopes from the first group of new epitopes as follows:

[0708] 1) If the mutation is at an anchor position and the binding difference of the smallest epitope is equal to or greater than 20, the smallest epitope has the highest score;

[0709] 2) If the mutation is at a non-anchor position and the binding difference of the smallest epitope is equal to or greater than 20, the score of the smallest epitope is lower than that of the smallest epitope in 1);

[0710] 3) If the mutation is at an anchor position and the binding difference of the smallest epitope is less than 20 and equal to or greater than 3, the score of the smallest epitope is lower than that of the smallest epitope in 2);

[0711] 4) If the mutation is at a non-anchor position and the binding difference of the smallest epitope is less than 20 and equal to or greater than 3, the score of the smallest epitope is lower than that of the smallest epitope in 3);

[0712] 5) If the mutation is at an anchor position and the binding difference of the smallest epitope is less than 3 and equal to or greater than 1, the score of the smallest epitope is lower than that of the smallest epitope in 4);

[0713] 6) If the mutation is at a non-anchor position and the binding difference of the smallest epitope is less than 3 and equal to or greater than 1, the score of the smallest epitope is lower than that of the smallest epitope in 5);

[0714] 7) If the smallest epitope has a mutation with a binding difference less than 1, the score of the smallest epitope is the lowest, regardless of the position of the mutation;

[0715] wherein the MHC I binding difference is given by: (% rank score (MHC I) for reference) / (% rank score (MHC I) for the smallest epitope); and select the second group of new epitopes with high scores;

[0716] iii. Optionally, prioritize the new epitopes from the second group based on the MHC I % rank score, and select the third group of new epitopes with a low MHC I % rank score;

[0717] iv. Optionally, select a fourth group of new epitopes from the second or third group, which includes the smallest epitopes that are highly similar to the epitopes known to be recognized by T cells;

[0718] v. Prioritize the new epitopes from the second group, the third group, or the fourth group based on the BLOSUM score, where a BLOSUM score less than a predetermined threshold ranks higher than a BLOSUM score equal to or greater than the threshold, and select the fifth group of new epitopes with a BLOSUM score less than the threshold, where the threshold is preferably 1;

[0719] where the first group, the second group, the third group, the fourth group, or the fifth group of new epitopes comprises the A epitopes.

[0720] 80. The method according to item 79, wherein step c comprises step i. and ii., or i. and iii., or i. and iv., or i. and v., or i., ii. and iii., or i., ii. and iv., or i., iii. and iv., or i., ii. and v., or i., iii. and v., or i., iv. and v., or i., ii., iii. and iv., or i., ii., iii. and v., or i., ii., iv. and v., or i., iii., iv. and v., or i., ii., iii., iv. and v.

[0721] 81. The method according to any one of items 79 to 80, wherein the method further comprises any feature of the method according to any one of items 1 to 78.

[0722] 82. A method for selecting a number A of new epitopes for an individual, the method comprising the following steps:

[0723] a. Obtain one or more new epitopes from the individual, each new epitope comprising at least one minimal epitope, wherein each new epitope comprises at least one mutation, such as an immunogenic mutation, compared to a reference sequence;

[0724] b. Determine the MHC I binding affinity of at least one minimal epitope (e.g., at least two, three, or four minimal epitopes) within each of the new epitopes, and determine the number of minimal epitopes within each of the new epitopes that bind MHC I;

[0725] c. Rank the new epitopes as follows:

[0726] i. Prioritize the new epitopes that contain a large number of minimal epitopes that bind MHC I, and select the first group of new epitopes with a high score;

[0727] ii. Optionally, prioritize the new epitopes from the first group of new epitopes as follows:

[0728] 1) If the mutation is in an anchor position and the binding difference of the minimal epitope is equal to or greater than 20, the minimal epitope has the highest score;

[0729] 2) If the mutation is at a non-anchoring position and the binding difference of the minimal epitope is equal to or greater than 20, the score of the minimal epitope is lower than the score of the minimal epitope in 1);

[0730] 3) If the mutation is at an anchoring position and the binding difference of the minimal epitope is less than 20 and equal to or greater than 3, the score of the minimal epitope is lower than the score of the minimal epitope in 2);

[0731] 4) If the mutation is at a non-anchoring position and the binding difference of the minimal epitope is less than 20 and equal to or greater than 3, the score of the minimal epitope is lower than the score of the minimal epitope in 3);

[0732] 5) If the mutation is at an anchoring position and the binding difference of the minimal epitope is less than 3 and equal to or greater than 1, the score of the minimal epitope is lower than the score of the minimal epitope in 4);

[0733] 6) If the mutation is at a non-anchoring position and the binding difference of the minimal epitope is less than 3 and equal to or greater than 1, the score of the minimal epitope is lower than the score of the minimal epitope in 5);

[0734] 7) If the minimal epitope has a mutation with a binding difference lower than 1, the score of the minimal epitope is the lowest, regardless of the position of the mutation;

[0735] wherein, the MHC I binding difference is given by: (reference % ranking score (MHC I)) / (% ranking score of the minimal epitope (MHC I)); and select a second group of new epitopes with high scores;

[0736] iii. Optionally, based on the MHC I % ranking score, prioritize the new epitopes from the second group and select a third group of new epitopes with a low MHC I % ranking score;

[0737] iv. Optionally, select a fourth group of new epitopes from the second group or the third group, which includes minimal epitopes highly similar to epitopes known to be recognized by T cells;

[0738] v. Prioritize the new epitopes from the second group, from the third group or from the fourth group based on the BLOSUM score, wherein a BLOSUM score less than a predetermined threshold ranks higher than a BLOSUM score equal to or greater than the threshold, and select a fifth group of new epitopes with a BLOSUM score less than the threshold, wherein the threshold is preferably 1;

[0739] vi. Optionally, based on the new epitopes found in two or more samples, select new epitopes from the first, second, third, fourth or fifth group of new epitopes, and select a sixth group of new epitopes found in two or more samples;

[0740] vii. Optionally, based on the identification of mutations by at least two different variant callers, select new epitopes from the first, second, third, fourth, fifth, or sixth group of new epitopes, and select a seventh group of new epitopes that contain mutations identified by at least two different variant callers,

[0741] wherein the first, second, third, fourth, fifth, sixth, or seventh group of new epitopes contains the A epitopes.

[0742] 83. The method according to item 82, wherein step c includes step i. and ii., or i. and iii., or i. and iv., or i. and v., or i. and vi., or i. and vii., or i., ii. and iii., or i., ii. and iv., or i., ii. and v., or i., ii. and vi., or i., ii. and vii., or i., iii. and iv., or i., iii. and v., or i., iii. and vi., or i., iii. and vii., or i., iv. and v., or i., iv. and vi., or i., iv. and vii., or i., v. and vi., or i., v. and vii., or i., vi. and vii., or ii., iii. and iv., or ii., iii. and v., or ii., iii. and vi., or ii., iii. and vii., or ii., iv. and v., or ii., iv. and vi., or ii., iv. and vii., or ii., v. and vi.. or ii., v. and vii., or ii., vi. and vii., or iii., iv. and v., or iii., iv. and vi., or iii., iv. and vii., or iii., v. and vi., or iii., v. and vii., or iii., vi. and vii., or iv., v. and vi., or iv., v. and vii., or iv., vi. and vii., or v., vi. and vii.

[0743] 84. The method according to any one of items 82 to 83, wherein the method further includes any feature of the method according to any one of items 1 to 78.

[0744] 85. A method for preparing a cancer vaccine comprising new epitopes, the method comprising the step of selecting the new epitopes using the method according to any one of items 1-78 or 79-84.

[0745] 86. The method according to item 85, wherein 10-20 new epitopes are selected.

[0746] 87. The method according to any one of items 85 to 86, wherein the cancer vaccine comprises a nucleotide construct, the nucleotide construct comprising:

[0747] - A targeting unit

[0748] - A dimerization unit

[0749] - A first linker

[0750] - An antigen unit, wherein the antigen unit comprises A - 1 antigen subunits, each subunit comprising a sequence encoding at least one of said neoepitopes and a second linker, and the antigen unit further comprises a final sequence encoding one of said neoepitopes, wherein A is an integer from 1 to 100, preferably A is an integer from 3 to 50,

[0751] wherein the nucleotide construct is applied to the anti - cancer vaccine in an immunologically effective amount.

[0752] 88. The method according to any one of items 85 to 87, wherein the second linker is a serine - glycine linker.

[0753] 89. The method according to any one of items 85 to 88, wherein the targeting unit has an affinity for chemokine receptors selected from CCR1, CCR3, and CCR5.

[0754] 90. The method according to item 89, wherein the targeting unit comprises an amino acid sequence having at least 80% sequence identity with amino acids 5 - 70 of the amino acid sequence SEQ ID NO:1 (MIP1a).

[0755] 91. A cancer vaccine obtained by the method according to any one of items 85 to 90.

[0756] 92. The cancer vaccine according to item 91, wherein the cancer vaccine comprises a nucleotide construct, the nucleotide construct comprising:

[0757] - A targeting unit

[0758] - A dimerization unit

[0759] - A first linker

[0760] - An antigen unit, wherein the antigen unit comprises n - 1 antigen subunits, each subunit comprising a sequence encoding at least one of said neoepitopes and a second linker, and the antigen unit further comprises a final sequence encoding one of said neoepitopes, wherein n is an integer from 1 to 100, such as 3 to 50.

[0761] wherein the nucleotide construct is applied to the anti - cancer vaccine in an immunologically effective amount.

[0762] 93. The cancer vaccine according to any one of items 91 to 92, wherein the second linker is a serine-glycine linker.

[0763] 94. The cancer vaccine according to any one of items 91 to 93, wherein the targeting unit has an affinity for chemokine receptors selected from CCR1, CCR3, and CCR5.

[0764] 95. The cancer vaccine according to any one of items 91 to 94, wherein the targeting unit comprises an amino acid sequence having at least 80% sequence identity with amino acids 5-70 of the amino acid sequence SEQ ID NO:1 (MIP1a).

[0765] 96. A method for selecting a number A of new epitopes for an individual, the method comprising the steps of:

[0766] a. Obtaining one or more new epitopes from the individual, each new epitope comprising at least one minimal epitope, wherein each new epitope comprises at least one mutation, such as an immunogenic mutation, compared to a reference sequence;

[0767] B. Determining the MHC I and / or MHC II binding affinity for each of the new epitopes;

[0768] c. Selecting new epitopes comprising at least one minimal epitope predicted to bind to MHC I and / or MHC II, thereby obtaining new epitopes that bind to MHC;

[0769] d. Ranking the new epitopes that bind to MHC according to the likelihood of clinical utility;

[0770] e. Selecting A new epitopes from the highest-ranked new epitopes that bind to MHC, where A is an integer and A is at least 3, such as at least 4, such as at least 5,

[0771] thereby selecting A new epitopes that are capable of inducing a CD8+ T cell response when administered to the individual in an immunologically active amount.

[0772] 97. The method according to item 96, wherein the method further comprises any feature of the method according to any one of items 1 to 78.

[0773] 98. A method for selecting a number A of new epitopes for an individual suffering from or suspected of suffering from cancer, the method comprising the steps of:

[0774] a. Obtaining one or more new epitopes from the individual, each new epitope comprising at least one minimal epitope, wherein each new epitope comprises at least one mutation, such as an immunogenic mutation, compared to a reference sequence, and wherein the minimal epitope consists of an amino acid number equal to or less than the amino acid number of the new epitope and comprises the at least one mutation; wherein preferably obtaining the new epitope includes the step of identifying mutations in a nucleic acid sequence specific to the tumor;

[0775] b. Determining the MHC I and / or MHC II binding affinity for at least one minimal epitope within each of the new epitopes, such as at least two, three or four minimal epitopes, optionally, wherein the binding affinity is determined by computer prediction;

[0776] c. Selecting a new epitope comprising at least one minimal epitope predicted to bind to MHC I and / or MHC II, thereby obtaining a new epitope that binds to MHC;

[0777] d. Ranking the new epitopes that bind to MHC according to the likelihood of clinical utility;

[0778] e. Selecting A new epitopes from the new epitopes that bind to MHC with the highest ranking,

[0779] thereby selecting A new epitopes that may have clinical utility, wherein step d further includes ranking the new epitopes as follows:

[0780] i) For each minimal epitope contained in the new epitope, determining whether the mutation is at an anchor position or a non-anchor position of the minimal new epitope;

[0781] ii) Prioritizing the new epitopes,

[0782] optionally, wherein in step ii) the new epitopes are prioritized by assigning the highest score of the minimal epitopes they contain to each new epitope, and wherein the new epitopes are prioritized as follows:

[0783] 1) If the mutation is at an anchor position and the binding difference of the minimal epitope is equal to or greater than 20, the minimal epitope has the highest score;

[0784] 2) If the mutation is at a non-anchor position and the binding difference of the minimal epitope is equal to or greater than 20, the score of the minimal epitope is lower than that of the minimal epitope in 1);

[0785] 3) If the mutation is at an anchor position and the binding difference of the minimal epitope is less than 20 and equal to or greater than 3, the score of the minimal epitope is lower than that of the minimal epitope in 2);

[0786] 4) If the mutation is at a non-anchoring position and the binding difference of the minimal epitope is less than 20 and equal to or greater than 3, the score of the minimal epitope is lower than the score of the minimal epitope in 3);

[0787] 5) If the mutation is at an anchoring position and the binding difference of the minimal epitope is less than 3 and equal to or greater than 1, the score of the minimal epitope is lower than the score of the minimal epitope in 4);

[0788] 6) If the mutation is at a non-anchoring position and the binding difference of the minimal epitope is less than 3 and equal to or greater than 1, the score of the minimal epitope is lower than the score of the minimal epitope in 5);

[0789] 7) If the minimal epitope has a mutation with a binding difference lower than 1, the score of the minimal epitope is the lowest, regardless of the position of the mutation.

[0790] 99. A method for selecting A new epitopes for an individual suffering from or suspected of suffering from cancer, the method comprising the following steps:

[0791] a. Obtaining one or more new epitopes from the individual, each new epitope comprising at least one minimal epitope, wherein each new epitope comprises at least one mutation, such as an immunogenic mutation, compared to a reference sequence, wherein the minimal epitope consists of an amino acid number equal to or less than the amino acid number of the new epitope and comprises the at least one mutation; wherein preferably obtaining the new epitope comprises the step of identifying mutations in a nucleic acid sequence specific for the tumor;

[0792] b. Determining the MHC I and / or MHC II binding affinity for at least one minimal epitope within each of the new epitopes, such as at least two, three or four minimal epitopes, optionally, wherein the binding affinity is determined by computer prediction;

[0793] c. Selecting new epitopes comprising at least one minimal epitope predicted to bind to MHC I and / or MHC II, thereby obtaining new epitopes that bind MHC;

[0794] d. Ranking the new epitopes that bind MHC according to the likelihood of clinical utility;

[0795] e. Selecting A new epitopes from the new epitopes that bind MHC with the highest ranking,

[0796] thereby selecting A new epitopes that may have clinical utility,

[0797] wherein step d comprises ranking the new epitopes according to the number of minimal epitopes comprised by the new epitopes, wherein a higher number of minimal epitopes gives a higher ranking.

[0798] 100. A method for selecting a number A of new epitopes for an individual having or suspected of having cancer, the method comprising the steps of:

[0799] a. obtaining one or more new epitopes from the individual, each new epitope comprising at least one minimal epitope, wherein each new epitope comprises at least one mutation, such as an immunogenic mutation, compared to a reference sequence, wherein the minimal epitope consists of an amino acid number equal to or less than the amino acid number of the new epitope and comprises the at least one mutation; wherein preferably obtaining the new epitope comprises the step of identifying mutations in a nucleic acid sequence specific to the tumor;

[0800] b. determining the MHC I and / or MHC II binding affinity for at least one minimal epitope within each of the new epitopes, such as at least two, three or four minimal epitopes, optionally, wherein the binding affinity is determined by computer prediction;

[0801] c. selecting new epitopes comprising at least one minimal epitope predicted to bind to MHC I and / or MHC II, thereby obtaining new epitopes that bind to MHC;

[0802] d. ranking the new epitopes that bind to MHC according to the likelihood of clinical utility;

[0803] e. selecting A new epitopes from the highest-ranked new epitopes that bind to MHC,

[0804] thereby selecting A new epitopes that may have clinical utility,

[0805] wherein step d comprises ranking the new epitopes according to the number of samples in which the new epitopes are found, wherein new epitopes found in a higher number of samples are ranked higher than new epitopes found in a lower number of samples, preferably wherein the samples are samples from different lesions.

[0806] 101. The method according to any one of items 99 to 100, wherein the method further comprises any feature of the method according to any one of items 1 to 78.

Claims

1. A method for selecting A novel epitopes for an individual having or suspected of having cancer, the method comprising the steps of: a. obtaining one or more novel epitopes from the individual, each novel epitope comprising at least one minimal epitope, wherein each novel epitope comprises at least one mutation relative to a reference sequence, and wherein the minimal epitope is composed of an amino acid number equal to or less than the amino acid number of the novel epitope and comprises the at least one mutation; wherein obtaining the novel epitopes comprises the step of identifying mutations in a tumor-specific nucleic acid sequence; b. determining the MHC I and / or MHC II binding affinity for at least one minimal epitope and the corresponding reference peptide, wherein the binding affinity is determined by computer prediction by calculating a % rank score, wherein a low % rank score indicates strong binding affinity and a high % rank score indicates weak binding affinity, and wherein the corresponding reference peptide is obtained from the body fluid or healthy tissue of the same individual or a normal healthy population; c. selecting novel epitopes comprising at least one minimal epitope predicted to bind to MHC I and MHC II, thereby obtaining novel epitopes that bind MHC; d. ranking the novel epitopes that bind MHC according to the likelihood of clinical utility, wherein the ranking comprises: I. determining the number of minimal epitopes contained in the novel epitopes that bind MHC selected in step c, and wherein novel epitopes containing a large number of minimal epitopes are ranked higher than novel epitopes containing a smaller number of minimal epitopes; II. determining the MHC I binding difference of the novel epitopes that bind MHC I selected in step c, wherein the MHC I binding difference is given by: (reference % rank score / (% rank score of the novel epitope)), and wherein novel epitopes having a high MHC I binding difference are ranked higher than novel epitopes having a lower MHC I binding difference, and wherein among novel epitopes having a high MHC I binding difference, novel epitopes having a mutation at the anchor position are prioritized; III. prioritizing the novel epitopes that bind MHC I selected in step c according to the MHC I ranking, wherein novel epitopes having a low % rank score are ranked higher than novel epitopes having a higher % rank score, and wherein the % rank score of a novel epitope comprising one or more minimal epitopes predicted to bind to MHC I is equal to the lowest % rank score of the one or more minimal epitopes; and IV. determining the BLOSUM score, and prioritizing the novel epitopes that bind MHC I selected in step c in descending order according to the BLOSUM score of the novel epitope, wherein the BLOSUM score of the novel epitope is equal to the BLOSUM score of the best-ranked minimal epitope it contains, and wherein minimal epitopes with a BLOSUM score < 2 are prioritized; and e. selecting A novel epitopes from the novel epitopes that bind MHC ranked highest in step d, thereby selecting A novel epitopes that may have clinical utility.

2. The method according to claim 1, wherein at least one minimal epitope predicted to bind to MHC I is selected in step c.

3. The method according to any one of claims 1 or 2, wherein in step d.I, the number of minimal epitopes contained in the newly identified MHC I-binding epitope is determined.

4. The method according to any one of claims 1 or 2, wherein A is an integer from 1 to 100.

5. The method according to any one of claims 1 or 2, wherein A is an integer from 5 to 50.

6. The method according to any one of claims 1 or 2, wherein A is an integer from 5 to 30.

7. The method according to any one of claims 1 or 2, wherein step a comprises identifying mutations in nucleic acid sequences in one or more samples from an individual compared to a reference sequence.

8. The method according to claim 7, wherein the one or more samples are samples of tumor cells from an individual.

9. The method according to claim 7, wherein the reference sequence is obtained from healthy tissue of an individual in need of treatment or from a healthy individual.

10. The method according to claim 7, wherein, the mutation is found in at least two samples of tumor cells from the individual.

11. The method according to any one of claims 1 or 2, wherein the newly identified epitope comprises 1 to 50 minimal epitopes.

12. The method according to claim 11, wherein the newly identified epitope comprises 1 to 40 minimal epitopes.

13. The method according to claim 11, wherein the newly identified epitope comprises 1 to 30 minimal epitopes.

14. The method according to claim 11, wherein the newly identified epitope comprises 1 to 20 minimal epitopes.

15. The method according to any one of claims 1 or 2, wherein step c comprises selecting a newly identified epitope that comprises at least one minimal epitope predicted to bind MHC I and at least one minimal epitope predicted to bind MHC II.

16. The method according to any one of claims 1 or 2, wherein step d.III comprises prioritizing newly identified epitopes having a % rank score below 2.

0.

17. The method according to any one of claims 1 or 2, wherein in step d.III, newly identified MHC I-binding epitopes having a % rank score higher than 2 are excluded or their prioritization is reduced.

18. The method according to any one of claims 1 or 2, wherein in step d.III, newly identified MHC I-binding epitopes having a % rank score higher than 10 are excluded or their prioritization is reduced.

19. The method according to any one of claims 1 or 2, wherein newly identified epitopes found in more than one sample from tumor cells are ranked higher than newly identified epitopes found in only one sample from tumor cells.

20. The method according to any one of claims 1 or 2, wherein the mutation is identified by at least two different variant callers.

21. The method according to any one of claims 1 or 2, further comprising the step of determining the RNA expression level of the newly identified epitope, wherein newly identified epitopes with undetectable RNA expression are excluded or their prioritization is reduced.

22. The method according to any one of claims 1 or 2, further comprising the step of determining the RNA expression level of the new epitope, wherein a new epitope having a detectable normalized transcript level expression higher than 0 transcripts per million is selected.

23. The method according to any one of claims 1 or 2, wherein step d further comprises the step of determining the allelic frequency of mutations present in the new epitopes that bind MHC, wherein the new epitopes that bind MHC with a high allelic frequency are ranked higher than the new epitopes that bind MHC with a lower allelic frequency, and wherein the high allelic frequency is an allelic frequency greater than 0.

05.

24. The method according to any one of claims 1 or 2, wherein the length of the new epitope is from 7 to 40 amino acids.

25. The method according to any one of claims 1 or 2, wherein step d further comprises the step of de-prioritizing or excluding new epitopes that do not contain any minimal epitopes predicted to bind MHC I.

26. The method according to any one of claims 1 or 2, wherein new epitopes containing at least 5 minimal epitopes that bind MHC are ranked higher than new epitopes containing fewer than 5 minimal epitopes that bind MHC.

27. The method according to claim 1, wherein the new epitopes of step d, II. and III. are ranked by assigning to each new epitope the highest score of the minimal epitopes it contains, wherein for each minimal epitope contained in the new epitope, it is determined whether the mutation is at an anchor position or a non-anchor position of the minimal epitope, and wherein the ranking of the new epitopes is carried out in descending order as follows in 1) to 7): 1) If the mutation is at an anchor position and the binding difference of the minimal epitope is equal to or greater than 20, the score of the minimal epitope is the highest; 2) If the mutation is at a non-anchor position and the binding difference of the minimal epitope is equal to or greater than 20, the score of the minimal epitope is lower than the score of the minimal epitope in 1); 3) If the mutation is at an anchor position and the binding difference of the minimal epitope is less than 20 and equal to or greater than 3, the score of the minimal epitope is lower than the score of the minimal epitope in 2); 4) If the mutation is at a non-anchor position and the binding difference of the minimal epitope is less than 20 and equal to or greater than 3, the score of the minimal epitope is lower than the score of the minimal epitope in 3); 5) If the mutation is at an anchor position and the binding difference of the minimal epitope is less than 3 and equal to or greater than 1, the score of the minimal epitope is lower than the score of the minimal epitope in 4); 6) If the mutation is at a non-anchor position and the binding difference of the minimal epitope is less than 3 and equal to or greater than 1, the score of the minimal epitope is lower than the score of the minimal epitope in 5); 7) If the minimal epitope has a mutation with a binding difference lower than 1, the score of the minimal epitope is the lowest, regardless of the position of the mutation.

28. A method for preparing a cancer vaccine comprising a new epitope, the method comprising the step of selecting a number A of new epitopes using the method according to any one of claims 1 or 2.

29. The method according to claim 28, wherein the cancer vaccine comprises: - a targeting unit - a dimerization unit; and - An antigen unit, wherein the antigen unit contains A neoepitopes.

30. The method according to claim 29, wherein A is an integer from 3 to 50.

31. The method according to claim 29, wherein A is an integer from 10 to 20.

Images