Personalized vaccines for cancer

Abstract

To provide a patient-specific tumor therapy targeting individual expression patterns of tumor antigens, particularly common tumor antigens, and individual tumor mutations.SOLUTION: Provided is a method for preventing or treating cancer in a patient, comprising: (i) inducing a first immune response in the patient against one or more tumor antigens; and (ii) inducing a second immune response in the patient against one or more tumor antigens, the second immune response being specific to a cancer-specific somatic mutation present in a cancer cell of the patient.SELECTED DRAWING: None

Description

Technical Field

[0001] The present invention relates to patient-specific tumor treatment targeting individual expression patterns of tumor antigens, particularly common tumor antigens, and individual tumor mutations.

Background Art

[0002] Cancer is a major cause of death, accounting for one in four deaths. Cancer treatment has traditionally been based on the law of averages, i.e., what works best for the largest number of patients. However, due to molecular diversity in cancer, less than 25% of treated individuals often benefit from approved therapies. Personalized medicine based on patient-specific treatment is considered a potential solution to the low efficacy and high cost of pharmaceutical innovation.

[0003] Antigen-specific immunotherapy aims to enhance or induce a specific immune response in patients and has been successfully used to suppress cancer diseases. T cells play a central role in cell-mediated immunity in humans and animals. Recognition and binding of specific antigens are mediated by the T cell receptor (TCR) expressed on the surface of T cells. The T cell receptor (TCR) of T cells can bind to major histocompatibility complex (MHC) molecules and interact with immunogenic peptides (epitopes) presented on the surface of target cells. Specific binding of the TCR initiates a signal cascade within T cells, leading to proliferation and differentiation into mature effector T cells.

[0004] The identification of ever more pathogen - related antigens and tumor - associated antigens (TAAs) has led to a broad collection of suitable targets for immunotherapy. Cells presenting immunogenic peptides (epitopes) derived from these antigens can be specifically targeted by either active or passive immunization strategies. Active immunization tends to induce and expand antigen - specific T cells in patients that can specifically recognize and kill diseased cells. Various antigen forms can be used for tumor vaccination, including whole cancer cells, proteins, peptides, or immunological vectors such as RNA, DNA, or viral vectors that can be applied directly in vivo or in vitro by pulsing DCs after introduction into the patient.

[0005] Cancer is thought to arise from the accumulation of genomic mutations and epigenetic changes, some of which may play a causative role. In addition to tumor - associated antigens, human cancers carry on average between 100 and 120 non - synonymous mutations, many of which are targetable by vaccines. More than 95% of the mutations in tumors are private and patient - specific (Weide et al. 2008: J. Immunother. 31, 180 - 188). The number of somatic mutations that can generate tumor - specific T - cell epitopes and change proteins ranges from 30 to 400. It has been computationally predicted that there are HLA class I - restricted epitopes derived from 40 to 60 tumor - specific somatic mutations per patient (Azuma et al. 1993: Nature 366, 76 - 79). Furthermore, new immunogenic HLA class II - restricted epitopes are also likely to result from tumor - associated mutations, but their numbers are still unknown.

[0006] In particular, some non-synonymous mutations are involved in malignant transformation and are essential for maintaining the oncogenic phenotype (driver mutations), and may be a potential "Achilles' heel" of cancer cells. Mutations found in primary tumors can also be present in metastases. However, several studies have demonstrated that metastatic tumors in patients often acquire further clinically significant genetic mutations during individual tumor evolution (Suzuki et al. 2007: Mol. Oncol. 1(2), 172-180; Campbell et al. 2010: Nature 467(7319), 1109-1113). Furthermore, the molecular characteristics of many metastases also deviate significantly from those of primary tumors. [Prior art documents] [Non-patent literature]

[0007] [Non-Patent Document 1] Weide et al.2008:J.Immunother.31,180-188 [Non-Patent Document 2] Azuma et al.1993:Nature 366,76-79 [Non-Patent Document 3] Suzuki et al.2007:Mol.Oncol.1(2),172-180 [Non-Patent Document 4] Campbell et al.2010:Nature 467(7319),1109-1113 [Overview of the Initiative] [Problems that the invention aims to solve]

[0008] Tumor heterogeneity is considered a major obstacle to the effectiveness of currently available therapies. The underlying technical problem of this invention is to provide a highly effective cancer vaccination strategy that overcomes the obstacles of existing approaches related to tumor heterogeneity.

[0009] This invention relates to a personalized therapeutic concept that integrates an individual's disease genetics to create customized therapies in oncology. Human cancers express a variety of immunogenic common tumor antigens and carry dozens to hundreds of non-synonymous mutations, many of which are targetable by T cells. Since these do not undergo central immune tolerance, these mutations are ideal candidates for vaccine development. This invention uses personalized vaccines, particularly RNA vaccines, that target individual tumor antigen expression patterns and individual tumor mutations. The concept of this invention is to leverage genetic tumor changes for the benefit of the patient, rather than being hindered by them.

[0010] Instead of searching for a common molecular denominator across many patients for targeting, this invention utilizes the antigen target repertoire of each individual patient. Rather than accepting a trade-off by providing a treatment designed for the average, this invention provides the best combination for each individual patient. This is not merely a paradigm shift, but opens up new possibilities to solve critical problems in current cancer drug development, such as broad inter-individual variability and clonal heterogeneity within tumors. This invention enables the optimal utilization of the antigen repertoire in a patient.

[0011] Specifically, this application relates to the multispecific targeting of the entire individual tumor antigen repertoire found in each individual cancer patient, including both non-mutated and mutant tumor antigens. To target non-mutated tumor antigens, a tumor antigen target portfolio (warehouse) covering a large proportion of patients can be used. This warehouse is a drug storage location for pre-production vaccines in "stock" and for combining these with vaccine cocktails for use in individual patients. Combined with vaccines formulated based on mutanome analysis, this enables optimal utilization of the antigen repertoire in patients.

[0012] The present invention involves identifying patient-specific cancer mutations and targeting the "signatures" of individual cancer mutations in a patient. Identifying non-synonymous point mutations that result in amino acid changes presented in the patient's major histocompatibility complex (MHC) molecules provides novel epitopes (neoepitopes) that are specific to the patient's cancer but not found in the patient's normal cells. Collecting sets of mutations from cancer cells, such as circulating tumor cells (CTCs), makes it possible to provide vaccines that induce immune responses potentially targeting primary tumors, even in the case of genetically distinct subpopulations and tumor metastases. For vaccination, such neoepitopes identified according to the present application are preferably provided to the patient in the form of a polypeptide containing the neoepitope, and after appropriate processing and presentation by MHC molecules, the neoepitope is presented to the patient's immune system to stimulate appropriate T cells.

[0013] Preferably, according to the present invention, an immune response is induced in a patient by administering an immunogenic gene product, such as RNA encoding a peptide or polypeptide, which contains one or more immunogenic epitopes to which an immune response is to be induced. Such an immunogenic gene product may contain an entire tumor antigen to which an immune response is to be induced, or a portion thereof, such as a T cell epitope. Strategies for directly injecting in vitro transcription RNA (IVT-RNA) into patients via various immunization pathways have been tested and proven successful in various animal models. The RNA may be translated in transfected cells, and the expression product may be presented on MHC molecules on the cell surface after processing to induce an immune response.

[0014] The advantages of using RNA as a type of reversible gene therapy include its transient expression and non-transforming properties. RNA does not need to enter the nucleus to be expressed, and furthermore, it cannot be incorporated into the host genome, thus eliminating the risk of expression. The transfection rate achievable with RNA is relatively high. Moreover, the amount of protein achieved corresponds to that in physiological expression. [Means for solving the problem]

[0015] The present invention relates to methods for inducing an efficient and specific immune response in cancer patients by administering cancer vaccines that target individual expression patterns of tumor antigens, such as common tumor antigens, and by administering cancer vaccines that target individual tumor mutations. Preferably, the cancer vaccine administered to the patient according to the present invention provides a patient-tumor-specific MHC-presented epitope suitable for stimulating, priming, and / or promoting T cells against cells expressing antigens (common tumor antigens and antigens having patient-specific mutations) that are specific to the patient's tumor and derived from MHC-presented epitopes. Thus, the vaccines described herein can induce or promote a cellular response to cancer disease, preferably cytotoxic T cell activity, characterized by the presentation of one or more cancer-expressing antigens by class I MHC. The vaccines administered according to the present invention are also specific to the patient's tumor because they target cancer-specific mutations.

[0016] In one aspect, the present invention relates to a method for preventing or treating cancer in a patient, (i) a step of inducing a first immune response to one or more tumor antigens in a patient, and (ii) A method comprising the step of inducing a secondary immune response in a patient to one or more tumor antigens, wherein the secondary immune response is specific to cancer-specific somatic mutations present in the patient's cancer cells.

[0017] (i) the step of inducing a first immune response and (ii) the step of inducing a second immune response may be performed simultaneously or sequentially. When the steps are performed sequentially, step (ii) is preferably performed after step (i). When steps (i) and (ii) are performed simultaneously, the vaccine for inducing the first immune response is preferably administered at the same time as the vaccine for inducing the second immune response. When steps (i) and (ii) are performed sequentially, the vaccine for inducing the first immune response is preferably administered before the vaccine for inducing the second immune response.

[0018] In one embodiment, the first and / or second immune response is induced by administering one or more suitable vaccines, particularly RNA vaccines. In one embodiment, the tumor antigen is a tumor-associated antigen.

[0019] In one embodiment, the first and / or second immune response is a cellular response. In one embodiment, the first immune response includes a CD8+ T cell response. In one embodiment, the second immune response includes a CD4+ T cell response.

[0020] In one embodiment, the first immune response is specific to a tumor antigen expression pattern (i.e., a collection of tumor antigens) present in the patient's cancer cells. In this embodiment, the first immune response preferably comprises an immune response to a collection of tumor antigens expressed in the patient's cancer cells. In one embodiment, the first immune response is not specific to cancer-specific somatic mutations present in the patient's cancer cells. In one embodiment, the first immune response is induced to one or more tumor antigens that are common tumor antigens. In one embodiment, a vaccine for inducing the first immune response induces an immune response specific to tumor antigens expressed in the majority of cancer cells of a cancer patient, and the cancer patient has the same type of cancer or a different type of cancer, such as the cancer treated according to the present invention. According to the present invention, the first immune response may be induced by providing the patient with a collection of tumor antigens, such as a common tumor antigen or its epitopes, and the epitopes may be provided in the form of a polyepitope polypeptide (also referred to herein as a polyvalent polypeptide) containing these epitopes. The antigens or epitopes are preferably provided to the patient by administering nucleic acids, particularly RNA, that encode the antigens or epitopes. After appropriate processing and presentation by MHC molecules, the epitope is presented to the patient's immune system to stimulate appropriate T cells.

[0021] Epitopes may exist in polyepitope polypeptides in the form of vaccine sequences, i.e., they may exist in their natural sequence situations, for example, in naturally occurring proteins, adjacent to amino acid sequences that flank the epitope. Such flanking sequences may each contain 5 or more, 10 or more, 15 or more, 20 or more, and preferably up to 50, up to 45, up to 40, up to 35, or 30 amino acids, and may flank the epitope sequence at the N-terminus and / or C-terminus. Thus, vaccine sequences may contain 20 or more, 25 or more, 30 or more, 35 or more, 40 or more, and preferably up to 50, up to 45, up to 40, up to 35, or 30 amino acids. In one embodiment, the epitope and / or vaccine sequence are aligned in the polypeptide from head to tail. In one embodiment, the epitope and / or vaccine sequence are separated by a linker. Such linkers are described further below. When attempting to induce a primary immune response using a pre-manufactured polyepitope polypeptide, preferably administered without determining the individual tumor antigen expression pattern of the patient, it is preferable that the polyepitope polypeptide contains the epitopes most likely to induce an immune response targeting one or more tumor antigens expressed by the patient's cancer cells, based on experimental data. This can be achieved by including epitopes in the polyepitope polypeptide selected to target the maximum number of tumor samples of the same and / or different types. However, it is desirable to keep the number of epitopes as small as possible. For this purpose, the examples demonstrate that a specific set of only three different tumor antigens is sufficient to cover 88% of the melanoma metastasis patient samples being analyzed. In other words, 88% of melanoma metastasis patients express at least one antigen from the aforementioned specific set of only three different tumor antigens. Therefore, it is expected that a primary immune response will be induced in 88% of melanoma metastasis patients by including at least one epitope from each of the three different tumor antigens in the polyepitope polypeptide.It should be understood that these antigen collections do not necessarily need to be complementary to cover the largest number of tumor patients while keeping the number of epitopes in the polyepitope polypeptide as low as possible, and therefore the polyepitope polypeptide does not necessarily need to contain epitopes from antigens shared by the largest proportion of tumor patients. Rather, the epitopes in such a set are optimized with respect to (i) antigens shared by the largest proportion of tumor patients and (ii) antigens covering the largest number of tumor patients, while keeping the number of epitopes in the polyepitope polypeptide as low as possible.

[0022] In one embodiment, the vaccine product for inducing a first immune response contains (i) one or more peptides or polypeptides, each containing one or more tumor antigens; (ii) one or more peptides or polypeptides, each containing one or more T cell epitopes of one or more tumor antigens; or (iii) a nucleic acid, preferably RNA, encoding one or more peptides or polypeptides contained in (i) or (ii). In one embodiment, the polypeptide used for immunity contains up to 30 epitopes. In one embodiment, the epitopes are present in their natural sequence state to form a vaccine sequence. In one embodiment, the vaccine sequence is about 30 amino acids long. In one embodiment, the epitopes and / or vaccine sequence are aligned in a head-to-tail direction. In one embodiment, the epitopes and / or vaccine sequence are separated by a linker.

[0023] In one embodiment, a vaccine product administered to induce a primary immune response induces an immune response to a tumor antigen common in the cancer being treated and / or common in various cancers. Preferably, the tumor antigen involved in inducing the primary immune response is a common tumor antigen. In one embodiment, the patient is positive for one or more tumor antigens to which a primary immune response is induced. In one embodiment, the patient is positive for all tumor antigens to which a primary immune response is induced. According to the present invention, the term “the patient is positive for tumor antigens” means that the patient’s cancer cells express the tumor antigens.

[0024] In one embodiment, the first immune response is induced by administering one or more vaccine products selected from a set of pre-production vaccine products, particularly RNA encoding a peptide or polypeptide containing its immunogenic fragment, such as a tumor antigen or T cell epitope, each pre-production vaccine product inducing an immune response to a tumor antigen. In one embodiment, the set includes vaccine products that induce an immune response to a variety of tumor antigens, such as at least three tumor antigens, at least five tumor antigens, at least eight tumor antigens, at least ten tumor antigens, at least fifteen tumor antigens, at least 20 tumor antigens, at least 25 tumor antigens, at least 30 tumor antigens, or more.

[0025] In one embodiment, cancer-specific somatic mutations are present in the exomes of the patient's cancer cells. In one embodiment, the cancer-specific somatic mutations are non-synonymous mutations. In one embodiment, the cancer cells are circulating tumor cells in the blood. In one embodiment, the second immune response is induced by administering a polypeptide containing mutation-based neoepitopes, or a vaccine containing nucleic acid, preferably RNA, encoding the polypeptide. In one embodiment, the polypeptide contains up to 30 mutation-based neoepitopes. In one embodiment, the polypeptide further contains epitopes that do not contain cancer-specific somatic mutations expressed by cancer cells. In one embodiment, the epitopes are present in their native sequence context to form a vaccine sequence. In one embodiment, the vaccine sequence is approximately 30 amino acids long. In one embodiment, the neoepitopes, epitopes, and / or vaccine sequences are aligned in a head-to-tail direction. In one embodiment, the neoepitopes, epitopes, and / or vaccine sequences are separated by a linker.

[0026] According to the present invention, a vaccine for inducing the first immune response is (a) the process of identifying tumor antigens expressed in tumor specimens from cancer patients, particularly common tumor antigens; and (b) Preferably, the vaccine may be provided by a method comprising the step of providing a vaccine characterized by the tumor antigen profile obtained in step (a), in particular the common tumor antigen profile, by selecting a vaccine product from a set of pre-production vaccine products, each of which preferably induces an immune response to a common tumor antigen.

[0027] According to the present invention, the term “common tumor antigen” refers to a tumor antigen expressed by the majority of cancers, e.g., the majority of cancers of the same type, e.g., the majority of cancer types treated according to the present invention, and / or the majority of cancers of different types. Thus, the term “common tumor antigen” refers to a tumor antigen shared by the majority of various patients having the same and / or different cancer types. Preferably, such a tumor antigen is a tumor antigen carrying at least one immunogenic T cell epitope. The term “majority” preferably means at least 60%, more preferably at least 70%, more preferably at least 80%, more preferably at least 90%, and particularly at least 95%. By targeting such common tumor antigens, it is possible according to the present invention to use a limited number of vaccine products that can be applied to the majority of cancer patients. According to the present invention, the term “same type of cancer” refers to cancers of the same medical classification, such as cancers of the same organ or tissue. Furthermore, according to the present invention, the term “different type of cancer” refers to cancers of different medical classifications, such as cancers of different organs or tissues.

[0028] According to the present invention, "inducing an immune response to a tumor antigen" preferably relates to the ability to induce an immune response to a tumor antigen, such as a common tumor antigen, or an immune response, preferably a T-cell response, to cells, such as cancer cells, that express and / or present a tumor antigen, such as a common tumor antigen, when administered to a patient. Accordingly, a vaccine for inducing an immune response to a common tumor antigen may contain (i) a peptide or polypeptide containing a common tumor antigen, or (ii) a peptide or polypeptide containing one or more T-cell epitopes of a common tumor antigen. In one particularly preferred embodiment, the peptide or polypeptide containing a common tumor antigen according to the present invention, or (ii) a peptide or polypeptide containing one or more T-cell epitopes of a common tumor antigen according to the present invention, is administered to the patient in the form of RNA, preferably such as in vitro transcription RNA or synthetic RNA, which can be expressed in the patient's cells, such as antigen-presenting cells, to produce the peptide or polypeptide.

[0029] According to the present invention, the vaccine for inducing the first immune response is preferably selected from a pre-supply vaccine warehouse, such as a pre-supply RNA vaccine warehouse. This approach is also referred to herein as “stock.” Such a pre-supply vaccine warehouse relates to a set of pre-supply vaccine products, each of which induces an immune response to a tumor antigen, such as a common tumor antigen. According to the present invention, such a warehouse preferably comprises a limited number of vaccine products designed to be applicable to the majority of cancer patients with the same type of cancer and / or the majority of cancer patients with different types of cancer. Thus, a vaccine warehouse used in accordance with the present invention preferably comprises a set of vaccine products applicable to the majority of cancer patients. For example, if a set of common tumor antigens is known with respect to a particular type of cancer, it is possible to create such a pre-supply vaccine warehouse comprising a set of vaccine products, the vaccine products in the set induce an immune response to the aforementioned common tumor antigens. Since such vaccine warehouses are selected to be applicable to the majority of patients, it is possible to select one or more vaccine products from the pre-supply vaccine warehouse that target the tumor antigen profile of each patient without requiring the provision of further vaccine products specifically designed for the patient being treated, thereby inducing an immune response against one or more tumor antigens expressed in the cancer cells of the particular patient being treated. Such selection can be carried out by testing the patient for tumor antigen expression and then selecting an appropriate vaccine product from the pre-supply vaccine warehouse that targets the tumor antigens expressed by the patient's cancer cells. Such selection can be carried out based on transcriptome / peptideome analysis of the patient's tumor. For example, tumor samples from eligible patients can be analyzed for tumor antigen signatures. A common tumor antigen profile can be determined by quantitative multiplex RT-PCR and IHC, and the corresponding vaccine product can be selected from the warehouse.The selection of one or more vaccine products from the pre-supply vaccine warehouse that induce an immune response to one or more tumor antigens expressed in the cancer cells of a specific patient being treated, and thus target the tumor antigen profile of each patient, can also be carried out by randomly selecting vaccine products from the pre-supply vaccine warehouse that are most likely to target one or more tumor antigens expressed by the patient's cancer cells, based on experimental data.

[0030] According to the present invention, the pre-production vaccine product set is optimized with respect to the coverage of tumor samples and their tumor antigen expression patterns. In particular, the set includes vaccine products selected to target the maximum number of tumor samples of the same and / or different types while keeping the number of vaccine products in the set as low as possible. For this purpose, the examples demonstrate that a specific set of only three different tumor antigens is sufficient to cover 88% of melanoma metastasis patient samples to be analyzed. In other words, 88% of melanoma metastasis patients express at least one antigen from the specific set of only three different tumor antigens. It should be understood that such a set does not necessarily have to include antigens shared by the largest proportion of tumor patients, as these antigen collections do not necessarily have to be complementary to cover the largest number of tumor patients while keeping the number of vaccine products in the warehouse as low as possible. Rather, such a set of vaccine products is preferably optimized with respect to (i) antigens shared by the largest proportion of tumor patients and (ii) antigens covering the largest number of tumor patients, while keeping the number of vaccine products in the warehouse as low as possible.

[0031] In one embodiment, a vaccine warehouse of tumor antigens, preferably common tumor antigens, is suitable for targeting at least 60%, more preferably at least 70%, more preferably at least 80%, more preferably at least 90%, and especially at least 95% of patients having a particular tumor type.

[0032] According to the present invention, the term “tumor antigen profile” refers to a collection of tumor antigens present in a patient’s cancer cells, i.e., all tumor antigens, such as common tumor antigens, present (i.e., expressed and preferably presented) in one or more cancer cells of the patient, or a subset of tumor antigens present in one or more cancer cells of the patient. Preferably, such a tumor antigen profile or collection of tumor antigens includes two or more, three or more, four or more, five or more, six or more, seven or more, eight or more, nine or more, ten or more, and preferably up to 30, up to 20, or up to 15 tumor antigens. Thus, the present invention may include the identification of all common tumor antigens present in one or more cancer cells of the patient, or the identification of only a subset of common tumor antigens present in one or more cancer cells of the patient. Generally, the number of common tumor antigens can be identified in a tumor specimen from a cancer patient that provides a sufficient number of common tumor antigens to be targeted by the vaccine.

[0033] The vaccine for inducing the first immune response, when administered to the patient, preferably provides a collection of MHC-presented epitopes from a collection of tumor antigens, such as a common tumor antigen, for example, a collection of 2 or more, 3 or more, 4 or more, 5 or more, 6 or more, 7 or more, 8 or more, 9 or more, 10 or more, and preferably up to 30, up to 20, or up to 15 tumor antigens. Presentation of these epitopes by the patient's cells, particularly antigen-presenting cells, preferably generates T cells that target the patient's tumor, preferably the primary tumor and tumor metastases, expressing antigens from which the MHC-presented epitopes originate, which, when bound to MHC, target the epitope, and present the same epitope on the surface of tumor cells.

[0034] According to the present invention, a vaccine for inducing a second immune response is (a) the process of identifying cancer-specific somatic mutations in tumor specimens from cancer patients and providing a cancer mutation signature for the patient; and (b) A method may be provided that includes the step of providing a vaccine characterized by the cancer mutation signature obtained in step (a).

[0035] In one embodiment, the method of the present invention is i) A step of providing tumor specimens from cancer patients and, preferably, non-tumor-forming specimens derived from cancer patients; ii) A step of identifying sequence differences between the genome, exome, and / or transcriptome of a tumor specimen and the genome, exome, and / or transcriptome of a non-tumor-forming specimen; iii) A step of designing a peptide or polypeptide containing an epitope incorporating the sequence difference determined in step (ii); iv) A step of providing a peptide or polypeptide designed in step (iii) or a nucleic acid, preferably RNA, that encodes the peptide or polypeptide; and v) The step of providing a vaccine containing the peptide or polypeptide or nucleic acid provided in step (iv) may include the step of providing a vaccine.

[0036] According to the present invention, a tumor specimen relates to any specimen, such as a body specimen derived from a patient who has or is expected to have a tumor or cancer cells. The body specimen may be any tissue specimen, such as blood, a tissue specimen obtained from a primary tumor or tumor metastasis, or any other specimen containing tumor or cancer cells. Preferably, the body specimen is blood, and cancer-specific somatic mutations or sequence differences are determined in one or more circulating tumor cells (CTCs) contained in the blood. In another embodiment, a tumor specimen relates to one or more isolated tumor or cancer cells, such as circulating tumor cells (CTCs), or a specimen containing one or more isolated tumor or cancer cells, such as circulating tumor cells (CTCs).

[0037] Non-tumor-forming specimens relate to any specimen, such as a body specimen, derived from a patient or, preferably, another individual of the same species as the patient, preferably a healthy individual that does not contain or is not expected to contain tumors or cancer cells. The body specimen may be any tissue specimen, such as a specimen from blood or non-tumor-forming tissue.

[0038] According to the present invention, the term “cancer mutation signature” may represent all cancer mutations present in one or more cancer cells of a patient, or only a portion of cancer mutations present in one or more cancer cells of a patient. Therefore, the present invention may include the identification of all cancer-specific mutations present in one or more cancer cells of a patient, or the identification of only a portion of cancer-specific mutations present in one or more cancer cells of a patient. Generally, the present invention provides the identification of many mutations that provide a sufficient number of neoepitopes to be included in a vaccine. “Cancer mutation” refers to sequence differences between nucleic acids present in cancer cells and nucleic acids present in normal cells.

[0039] Preferably, the mutations identified according to the present invention are non-synonymous mutations, preferably non-synonymous mutations of proteins expressed in tumor or cancer cells.

[0040] In one embodiment, cancer-specific somatic mutations or sequence differences are determined in the genome of a tumor specimen, preferably the entire genome. Therefore, the present invention may include identifying cancer mutation signatures in the genome of one or more cancer cells, preferably the entire genome. In one embodiment, the step of identifying cancer-specific somatic mutations in a tumor specimen from a cancer patient includes identifying a whole-genome cancer mutation profile.

[0041] In one embodiment, cancer-specific somatic mutations or sequence differences are determined in the exome, preferably the entire exome, of a tumor specimen. The exome is part of an organism's genome, formed by exons, which are the coding portions of expressed genes. The exome provides the genetic blueprint used in the synthesis of proteins and other functional gene products. It is the functionally most important part of the genome and therefore most likely to contribute to an organism's phenotype. The exome of the human genome is estimated to account for 1.5% of the entire genome (Ng, PC et al., PLoS Gen., 4(8):1-15, 2008). Therefore, the present invention may include identifying the cancer mutation signature of the exome, preferably the entire exome, of one or more cancer cells. In one embodiment, the step of identifying cancer-specific somatic mutations in a tumor specimen from a cancer patient includes identifying a whole-exome cancer mutation profile.

[0042] In one embodiment, cancer-specific somatic mutations or sequence differences are determined in the transcriptome of a tumor specimen, preferably the entire transcriptome. The transcriptome is the set of all RNA molecules, including mRNA, rRNA, tRNA, and other non-coding RNAs, produced in a single cell or cell population. In relation to the present invention, the transcriptome means the set of all RNA molecules produced in a single cell, cell population, preferably cancer cell population, or all cells of a given individual at a particular point in time. Accordingly, the present invention may include identifying the cancer mutation signature of the transcriptome, preferably the entire transcriptome, of one or more cancer cells. In one embodiment, the step of identifying cancer-specific somatic mutations in a tumor specimen from a cancer patient includes identifying the whole transcriptome cancer mutation profile.

[0043] In one embodiment, the step of identifying cancer-specific somatic mutations or sequence differences includes single-cell sequencing of one or more cancer cells, preferably 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 or more. Thus, the present invention may include identifying cancer mutation signatures of the one or more cancer cells. In one embodiment, the cancer cells are circulating tumor cells. Cancer cells such as circulating tumor cells can be isolated before single-cell sequencing.

[0044] In one embodiment, the step of identifying cancer-specific somatic mutations or sequence differences includes using next-generation sequencing (NGS).

[0045] In one embodiment, the step of identifying cancer-specific somatic mutations or sequence differences includes sequencing the genomic DNA and / or RNA of a tumor specimen.

[0046] To identify cancer-specific somatic mutations or sequence differences, sequence information obtained from tumor specimens is compared with a reference, preferably sequence information obtained from sequencing nucleic acids such as DNA or RNA from normal non-cancerous cells, such as germline cells, which may be obtained from either the patient or a different individual. In one embodiment, normal genomic germline DNA is obtained from peripheral blood mononuclear cells (PBMCs).

[0047] Vaccines for inducing a secondary immune response, when administered to a patient, preferably provide a collection of MHC-presented epitopes incorporating sequence changes based on identified mutations or sequence differences, for example, a collection of 2 or more, 5 or more, 10 or more, 15 or more, 20 or more, 25 or more, 30 or more, and preferably up to 60, up to 55, up to 50, up to 45, up to 40, up to 35, or up to 30 MHC-presented epitopes. Such MHC-presented epitopes incorporating sequence changes based on identified mutations or sequence differences are also referred to herein as “neoepitopes.” Presentation of these epitopes by the patient’s cells, particularly antigen-presenting cells, preferably generates T cells that target the patient’s tumor, preferably primary tumors and tumor metastases, expressing antigens that, when bound to MHC, target the epitope, and thus present the same epitope on the surface of tumor cells.

[0048] To provide a vaccine for inducing a secondary immune response, the present invention may include arbitrarily incorporating a sufficient number of neoepitopes (preferably in the form of coding nucleic acids) into the vaccine, or may include a further step of determining the utility of identified mutations in epitopes for cancer vaccination. Thus, this further step may include one or more of the following: (i) evaluating whether the sequence change is located in a known or predicted MHC-presenting epitope; (ii) testing in vitro and / or in silico whether the sequence change is located in an MHC-presenting epitope, for example, testing whether the sequence change is part of a peptide sequence that is processed into and / or presented as an MHC-presenting epitope; and (iii) testing in vitro whether the assumed mutant epitope can stimulate patient T cells with desired specificity, particularly if it exists in their natural sequence context, for example, if it is adjacent to an amino acid sequence adjacent to the epitope even in naturally occurring proteins, and if expressed in antigen-presenting cells. Such flanking sequences may each contain three or more, five or more, ten or more, fifteen or more, twenty or more, and preferably up to fifty, up to 45, up to 40, up to 35, or up to 30 amino acids, and may be adjacent to an epitope sequence at the N-terminus and / or C-terminus.

[0049] Mutations or sequence differences determined according to the present invention can be ranked in terms of their usefulness as epitopes for cancer vaccination. Therefore, in one embodiment, the present invention provides a manual or computer-based analytical process for analyzing and selecting identified mutations in terms of their usefulness in each vaccine offered. In a preferred embodiment, the analytical process is based on a computer algorithm.

[0050] Preferably, the analytical process consists of the following steps: - For example, a process of identifying expressed protein modification mutations by analyzing transcripts; - The process of identifying potentially immunogenic mutations, i.e., identifying the obtained data by comparing it with available datasets of confirmed immunogenic epitopes, such as those contained in public immunoepitope databases, such as the IMMUNE EPITOPE DATABASE AND ANALYSIS RESOURCE at http: / / www.immunoepitope.org; This includes one or more, preferably all, of the following.

[0051] The step of identifying potentially immunogenic mutations may include determining and / or ranking epitopes according to predictions of MHC binding capacity, preferably MHC class I binding capacity.

[0052] In another embodiment, epitopes can be selected and / or ranked using further parameters such as protein influence, related gene expression, sequence specificity, predicted presentation potential, and association with oncogenes.

[0053] Multiple CTC analyses also enable the selection and prioritization of mutations. For example, mutations found in a larger proportion of CTCs may be given higher priority than mutations found in a smaller proportion of CTCs.

[0054] A collection of mutation-based neoepitopes, identified and provided by a vaccine for inducing a secondary immune response, preferably exists in the form of a polypeptide containing the neoepitope (polyepitope polypeptide) or a nucleic acid, particularly RNA, encoding the polypeptide. Furthermore, neoepitopes may also exist in polypeptides in the form of a vaccine sequence, adjacent to an amino acid sequence adjacent to the epitope, for example, in naturally occurring proteins, i.e., they may exist in their natural sequence context. Such flanking sequences may each contain 5 or more, 10 or more, 15 or more, 20 or more, and preferably up to 50, up to 45, up to 40, up to 35, or 30 amino acids, and may be adjacent to the epitope sequence at the N-terminus and / or C-terminus. Thus, vaccine sequences may contain 20 or more, 25 or more, 30 or more, 35 or more, 40 or more, and preferably up to 50, up to 45, up to 40, up to 35, or 30 amino acids. In one embodiment, the neoepitope and / or vaccine sequence are aligned in the polypeptide from head to tail.

[0055] In one embodiment, the neoepitope and / or vaccine sequence is separated by a linker, particularly a neutral linker. The term “linker” according to the present invention relates to a peptide added to two peptide domains, such as an epitope or vaccine sequence, to ligate the peptide domains. There are no particular restrictions on the linker sequence. However, it is preferable that the linker sequence reduces steric hindrance between the two peptide domains, is well translated, and supports or allows the processing of the epitope. Furthermore, the linker should have no immunogenic sequence elements or only a small amount. Preferably, the linker should not generate non-endogenous neoepitopes, such as those resulting from junctional sutures between adjacent neoepitopes, which may produce an undesirable immune response. Therefore, polyepitope vaccines should preferably include a linker sequence that can reduce the number of undesirable MHC-binding conjugating epitopes. Hoyt et al. (EMBO J.25(8),1720-9,2006) and Zhang et al. (J.Biol.Chem.,279(10),8635-41,2004) showed that glycine-rich sequences reduce proteasome processing, and therefore the use of glycine-rich linker sequences works to minimize the number of peptides contained in the linker that can be processed by the proteasome. Furthermore, glycine was observed to inhibit strong binding at the MHC binding groove site (Abastado et al.,J.Immunol.151(7),3569-75,1993). Schlessinger et al. (Proteins,61(1),115-26,2005) found that the amino acids glycine and serine included in the amino acid sequence result in a more flexible protein that is more efficiently translated and processed by the proteasome, allowing better access to the encoded neoepitope. Each linker may contain 3 or more, 6 or more, 9 or more, 10 or more, 15 or more, 20 or more, and preferably up to 50, up to 45, up to 40, up to 35, or up to 30 amino acids.Preferably, the linker is enriched with glycine and / or serine amino acids. Preferably, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, or at least 95% of the amino acids in the linker are glycine and / or serine. In one preferred embodiment, the linker consists substantially of the amino acids glycine and serine. In one embodiment, the linker is an amino acid sequence (GGS). a (GSS) b (GGG) c (SSG) d (GSG) e This includes, where a, b, c, d, and e are independently selected numbers from 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20, and a+b+c+d+e is not 0, but preferably 2 or greater, 3 or greater, 4 or greater, or 5 or greater. In one embodiment, the linker includes the sequences described herein, including the linker sequences described in the examples, such as the sequence GGSGGGGSG.

[0056] In another embodiment, the collection of mutation-based neoepitopes identified and provided by a vaccine for inducing a secondary immune response may exist in the form of a collection of peptides containing the neoepitopes on various peptides, each containing one or more neoepitopes that may also overlap, or a collection of nucleic acids, particularly RNA, encoding the peptides.

[0057] Vaccine administration for inducing a secondary immune response may provide an MHC class II presenting epitope that can induce a CD4+ helper T cell response against cells expressing an antigen derived from the MHC presenting epitope. Alternatively, in addition, vaccine administration for inducing a secondary immune response may provide an MHC class I presenting epitope that can induce a CD8+ T cell response against cells expressing an antigen derived from the MHC presenting epitope. Furthermore, vaccine administration for inducing a secondary immune response may provide one or more neoepitopes (including known neoepitopes and neoepitopes identified according to the present invention), as well as one or more epitopes that do not include cancer-specific somatic mutations but are expressed by cancer cells and preferably induce an immune response against cancer cells, preferably a cancer-specific immune response. In one embodiment, administration of a vaccine to induce a secondary immune response provides a neoepitope that can induce a CD4+ helper T cell response in cells expressing an antigen that is an MHC class II presenting epitope and / or derived from an MHC presenting epitope, and a cancer-specific somatic mutation-free epitope that can induce a CD8+ T cell response in cells expressing an antigen that is an MHC class I presenting epitope and / or derived from an MHC presenting epitope. In one embodiment, the cancer-specific somatic mutation-free epitope is derived from a tumor antigen. In one embodiment, the neoepitope and the cancer-specific somatic mutation-free epitope have a synergistic effect in the treatment of cancer. Preferably, the vaccine to induce a secondary immune response is useful for polyepitope stimulation of cytotoxic responses and / or helper T cell responses.

[0058] In one particularly preferred embodiment, a peptide or protein for vaccination, such as a polyepitope polypeptide according to the present invention, is administered to the patient in the form of a nucleic acid, preferably RNA such as in vitro transcription RNA or synthetic RNA, which can be expressed in the patient's cells, such as antigen-presenting cells, to produce the peptide or protein. The present invention also envisions the administration of one or more multiepitope polypeptides, which for the purposes of the present invention are encompassed in the term "polyepitope polypeptide," preferably in the form of a nucleic acid, preferably RNA such as in vitro transcription RNA or synthetic RNA, which can be expressed in the patient's cells, such as antigen-presenting cells, to produce one or more polypeptides. When two or more multiepitope polypeptides are administered, the neoepitopes provided by the different multiepitope polypeptides may be different or partially overlapping. Once present in the patient's cells, such as antigen-presenting cells, the peptide or protein is processed to produce an immunogenic epitope, such as a neoepitope.

[0059] Particularly preferred embodiments of the present invention include (i) an in vitro transcribed polynucleotide RNA vaccine cocktail containing “stock” RNA from a pre-supply RNA warehouse targeting a common tumor antigen profile of each patient, and (ii) an on-demand-produced RNA vaccine encoding a neoepitope derived from a patient-specific mutation.

[0060] The vaccines described herein may contain a pharmaceutically acceptable carrier and may optionally contain one or more adjuvants, stabilizers, etc. The vaccines may be in the form of therapeutic vaccines or prophylactic vaccines.

[0061] In a further embodiment, the present invention provides a vaccine described herein for use in the therapeutic methods described herein, and more particularly for use in the treatment or prevention of cancer.

[0062] The cancer treatments described herein may be combined with surgical resection and / or radiation and / or conventional chemotherapy.

[0063] Other features and advantages of the present invention will become apparent from the following detailed description and claims. [Brief explanation of the drawing]

[0064] [Figure 1] Figure 1, top: Steps for identifying and prioritizing potential immunogenic somatic mutations in bulk tumor samples. Figure 1, bottom: Steps applied to B16 and Black6 systems. [Figure 2] Figure 2: Examples of demonstrated mutations in Kif18b. Mutations identified in the Kif18b gene by NGS exome sequencing, confirmed by Sanger sequencing. In wild-type cells, the sequence is T / T. In tumor cells, this sequence is a mixture of T / G. [Figure 3] Figure 3: Immunoreactivity to mutant sequences. Mice (n=5) were immunized twice (day 0 and day 7) with a mutant peptide sequence (100 μg + PolyI:C 50 μg; sc). Mice were sacrificially killed on day 12, and splenocytes were collected. IFNγ ELISpot was performed using 5 × 10⁵ splenocytes / well as effectors and 5 × 10⁴ bone marrow dendritic cells loaded with the peptide (2 μg / ml at 37°C and 5% CO₂ for 2 hours) as target cells. Effector splenocytes were tested against the mutant peptide, wild-type peptide, and control peptide (bullous stomatitis virus nucleoprotein, VSV-NP, amino acids 52-59). The mean number of spots measured for VSV-NP, after subtracting background spots for all mice, is shown (white circles: mice immunized with wild-type peptide; black squares: mice immunized with mutant peptide). Data are shown for each mouse and expressed as mean ± SEM. [Figure 4]Figure 4: Survival benefits of mice vaccinated with newly identified mutant peptide sequences. B16F10 cells (7.5 × 10⁴) were subcutaneously inoculated on day 0. Mice were vaccinated with peptide 30 (Jerini Peptide Technologies (Berlin); peptide 100 μg + PolyI:C 50 μg sc (Invivogen)) on days -4, +2, and +9. The control group received only Poly I:C (50 μg sc). Tumor growth was observed up to day +16, and the log-rank (Mantel-Cox) test* showed p<0.05. [Figure 5] Figure 5A: Example of enhanced protein expression by RNA optimized for stability and translation efficiency (left: eGFP, right: luciferase). Figure 5B: Example of polyepitope expansion of antigen-specific CD8+ and CD4+ T cells by RNA optimized for effective antigen pathways (see reference, Kreiter, Konrad, Sester et al, Cancer Immunol. Immunother. 56:1577-1587, 2007). Figure 5C: Example of preclinical demonstration of antitumor effect in a B16 black model using an RNA vaccine encoding a single epitope (OVA-SIINFEKL). Survival data were obtained for mice treated with the vaccine alone or in combination with an adjuvant. Figure 5D: Design of an individualized polyneoepitope vaccine. Functional elements for increased expression and optimized immunogenicity are incorporated into the vaccine vehicle. Up to 30 mutant epitopes isolated by a linker can be incorporated for each molecule in their native sequence context. [Figure 6]Figure 6: Construct design. Figure 6A: Schematic diagram of the RNA polyepitope construct. Cap; Cap analogue; 5'UTR: 5' untranslated region; L: Linker; Sequence 1: RNA sequence encoding a peptide containing a mutant amino acid; 3'UTR: 3' untranslated sequence; PolyA: PolyA tail. Figure 6B: Sequence of the RNA construct encoding two amino acid sequences containing a mutant amino acid from B16F10. The start and end codons, as well as the signal peptide and MITD sequences, are not part of the schematic diagram and are represented by the symbol "....". [Figure 7] Figure 7: Functionality of RNA polyepitopes. IFNγ ELISpot data using 5 × 10⁵ splenocytes / well as effectors and 5 × 10⁴ BMDCs as target cells. BMDCs were transfected with RNA (20 μg) by peptide loading (2 μg / ml at 37°C and 5% CO₂ for 2 hours) or electroporation. Control RNAs were eGFP (left panel) or RNA constructs encoding two unrelated peptides containing mutant amino acids isolated by a linker. Data are shown mean ± SEM. Figure 7A: Data for RNA encoding mutant peptide 30, wild-type peptide 30, and mutants 30 and 31. Figure 7B: Data for RNA encoding mutant peptide 12, wild-type peptide 12, and mutants 12 and 39. Figure 7C: Representative ELISpot scan from a single mouse readout as shown in Figure 7B. [Figure 8] Figure 8: Two embodiments of RNA polyneoepitope vaccines showing conjugation epitopes. RNA vaccines can be constructed with a linker between the mutation-encoding peptides (top) or without a linker (bottom). Good epitopes include those containing somatic mutations ("*") that bind to MHC molecules. Poor epitopes include epitopes that bind to MHC molecules but contain either part of one of the two peptides (bottom) or part of the peptides and a linker sequence (top). [Figure 9A]Figure 9: Discovery and characterization of the "T-cell druggable mutanome". Figure 9A: The flowchart outlines the experimental procedure from B16F10 and C57BL / 6 samples to ELISPOT readout. [Figure 9B] Figure 9B: Shows the number of hits for each evaluation stage and the process of selecting mutations for DNA validation and immunogenicity testing. The mutations selected for validation and immunogenicity testing were immunogenic and predicted to be expressed in the gene with RPKM > 10. [Figure 9C] Figure 9C: T-cell druggable mutantomes mapped to the B16F10 genome. The outer-to-inner ring represents the following subsets: (1) present in all triplets, (2) having an FDR < 0.05, (3) located within a protein-coding region, (4) causing non-synonymous mutations, (5) localized within expressed genes, and (6) present in the demonstrated sets. Mouse chromosomes (outer circle), gene density (green), gene expression (green (low) / yellow / red (high)), and somatic mutations (orange). [Figure 10A] Figure 10: In vivo immune response induced by vaccination of mice with long synthetic peptides representing mutations. Figure 10A, B: IFN-γ ELISPOT analysis of T cell effectors from mice vaccinated with mutation-coding peptides. Columns represent the mean (±SEM) of 5 mice per group. Asterisks indicate statistically significant differences in responsiveness to the mutation peptide and the wild-type peptide (Student's t-test; p-value < 0.05). Figure 10A: Splenocytes from vaccinated mice were restimulated with BMDCs transfected with the mutation-coding peptide used for vaccination, the corresponding wild-type peptide, and an unrelated control peptide (VSV-NP). [Figure 10B] Figure 10B: To analyze T cell responsiveness to endogenously processed mutations, vaccinated mouse splenocytes were restimulated with BMDCs transfected with control RNA (eGFP) or RNA encoding the indicated mutation. [Figure 10C]Figure 10C: Mutation 30 (gene Kif18B, protein Q6PFD6, mutation p.K739N). Sanger sequencing trace and mutation sequence (top). Protein domain and mutation location (bottom). [Figure 11] Figure 11: Antitumor activity of mutant peptide vaccines in mice with aggressively growing B16F10 tumors. Figure 11A: C57BL / 6 mice (n=7) were subcutaneously inoculated with 7.5 × 10⁴ B16F10 cells into the flanks of the mice. On days 3 and 10 after tumor inoculation, the mice were vaccinated with 100 μg of MUT30 or MUT44 peptide + 50 μg of poly(I:C) or the adjuvant alone. Figure 11B: C57BL / 6 mice (n=5) received a single immunization with 100 μg of MUT30 peptide + 50 μg of poly(I:C) on day -4. On day 0, 7.5 × 10⁴ B16F10 cells were subcutaneously inoculated into the flanks of the mice. Booster immunization with MUT30 peptide (+ poly(I:C)) was performed on days 2 and 9. Kaplan-Meier survival blot (left). Tumor growth dynamics (right). [Figure 12] Figure 12: Vaccination with mutation-encoding RNA evokes CD4+ and CD8+ T cell responses. Intracellular cytokine staining analysis data for IFN-γ in CD4+ and CD8+ T cell effectors from mice vaccinated with mutation-encoding RNA. The RNA encoded one (monoepitope, top row), two (biepitope, middle row), or 16 (polyepitope, bottom row) different mutations. Dots represent the average of three mice per group. Asterisks indicate statistically significant differences in responsiveness to the mutant peptide and control peptide (VSV-NP) (Student's t-test; p-value < 0.05). FACS plots show the effectors from the animals with the highest IFN-γ secretion for each mutation, indicating the phenotype of the T cell response. [Figure 13]Figure 13: Vaccination with mutation-encoding polyepitope RNA evokes T cell responses to several mutations. IFN-γ ELISPOT analysis of T cell effectors from mice vaccinated with mutation-encoding polyepitope RNA containing 16 different mutations. Columns represent the mean (±SEM) of 3 mice per group. The photograph shows a triple well of cells from one exemplary animal restimulated with the indicated peptide. [Figure 14] Figure 14: Vaccination with five different model epitopes encoded by a single RNA molecule elicits an immune response to all encoded epitopes. Figure 14A: IFN-γ ELISPOT analysis in T cell effectors from mice vaccinated with mutation-encoding model polyepitopes, including five different model epitopes (SIINFEKL, Trp2, VSV-NP, Inf-NP, OVA class II). Splenocytes were restimulated with the indicated peptides. Spots represent the mean of triple wells from five mice per group. Figure 14B: Pentamer staining of blood lymphocytes from one control mouse and one mouse immunized with a model polyepitope. CD8+ cells stained with the Inf-NP pentamer are specific to the Inf-NP peptide. [Figure 15A] Figure 15: Mutation-inducing CD4+ T cells can induce potent antitumor activity against B16F10 melanoma through synergistic action with CD8+ T cell epitopes. 1 × 10⁵ B16F10 cells were subcutaneously inoculated into the flanks of C57BL / 6 mice (n=8). On days 3, 10, and 17 after tumor inoculation, mice were vaccinated with 100 μg of MUT30, Trp2, or both peptides + 50 μg of poly(I:C). Figure 15A: Mean tumor growth dynamics for each group. On day 28, the mean values ​​between the single-treatment group and the untreated and combined treatment groups were statistically different (Mann-Whitney test, p < 0.05). [Figure 15B] Figure 15B: Kaplan-Meier survival plots for various groups. The survival curves for mice vaccinated with MUT30 and MUT30+Trp2 are statistically different (log-rank test, p-value = 0.0029). [Figure 16] Figure 16: Overview of the process for finding somatic mutations in B16. The number of steps for each individual step is shown as an example for one B16 sample compared to one black6 sample. "Exon" refers to the exon coordinates defined by all RefSeq transcripts encoding the protein. [Figure 17] Figure 17: Venn diagrams showing the number of somatic mutations in protein-coding exons found by individual, two, or all three software tools. The numbers were calculated after filtering and represent the consensus of all three samples. [Figure 18A] Figure 18A: Examples of single nucleotide mutations found: somatic mutations found in all three B16 samples (left), non-somatic mutations found in all B16 and black6 samples (center), and a mutation found in only one black6 sample (right). [Figure 18B] Figure 18B: Computed FDR distributions for a dataset of demonstrated mutations; the distributions are visualized as mean-estimated ROC curves, with gray bars indicating 95% confidence intervals for the mean of both dimensions at uniform sampling locations. The mean was obtained from the distribution of estimated ROC curves of FDR for all 18 possible combinations (see text). [Figure 19] Figure 19A: Estimated ROC curves for comparison of three different software tools (doublet, 38x coverage). Figure 19B: Estimated ROC curves for comparison of various average sequencing depths (samtools, no replication). 38x represents experimentally obtained coverage, while other coverages were downsampled from this data. Figure 19C: Estimated ROC curves visualizing the effect of experimental replication (38x coverage, samtools). Figure 19D: Estimated ROC curves for various sequencing protocols (samtools, no replication). Curves were calculated using results from a 2×100 nucleotide library. [Figure 20]Figure 20A: Ten demonstrated mutations with the lowest FDR, selected using the optimal parameter set from the final set of 2396 mutations. None of these mutations are present in dbSNP (version 128; genome assembly mm9). Figure 20B: Relative amounts of mutations found in the same dataset as A with respect to a given FDR cutoff value, plotted separately for all mutations in the dataset and demonstrated mutations. Only values ​​between 0 and 10% FDR are shown for visual clarity. [Figure 21] Figure 21: Antitumor activity of a polyepitope RNA vaccine encoding mutations. C57BL / 6 mice (n=10) were subcutaneously inoculated with 1 × 10⁵ B16F10 cells into the flank of the mice. On days 3, 6, 10, 17, and 21 after tumor inoculation, the mice were vaccinated with a liposomal RNA transfection reagent containing polytope RNA. The control group was given liposomes without RNA. The figure shows Kaplan-Meier survival plots for the various groups. The survival curves are statistically different (log-rank test, p-value = 0.0008). [Figure 22] Figure 22: Selection of tumor antigen combinations as targets for cancer treatment. A combination of only three tumor antigens, DCT, TYR, and TPTE, was sufficient to represent 88% of the melanoma metastasis samples analyzed. [Modes for carrying out the invention]

[0065] The present invention will be described in detail below, but it should be understood that the present invention is not limited to the specific methods, protocols, and reagents described herein, and that these may vary. Furthermore, it should be understood that the terms used herein are intended solely to describe specific embodiments and are not intended to limit the scope of the present invention, and that the scope of the present invention is limited only by the accompanying claims. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art.

[0066] The elements of the present invention are described below. These elements are listed together with specific embodiments, but it should be understood that additional embodiments can be created by combining them in any way and in any number. The various examples and preferred embodiments described should not be construed as limiting the present invention to only the embodiments described herein. It should be understood that this description supports and encompasses embodiments in which the explicitly described embodiments are combined with many disclosed and / or preferred elements. Furthermore, any order and combination of all elements described herein should be considered disclosed by this description unless specifically indicated by the context. For example, if in one preferred embodiment the RNA comprises a poly(A) tail consisting of 120 nucleotides, and in another preferred embodiment the RNA molecule comprises a 5' cap analog, then in the preferred embodiment the RNA comprises a poly(A) tail consisting of 120 nucleotides and a 5' cap analog.

[0067] Preferably, the terms used herein are defined as those described in "A multilingual glossary of biotechnological terms: (IUPAC Recommendations)", H.G. W. Heuenberger, B. Nagel, and H. Kolbl, Eds., Helvetica Chimica Acta, CH-4010 Basel, Switzerland, (1995).

[0068] The implementation of this invention will, unless otherwise specified, utilize conventional methods of biochemistry, cell biology, immunology, and recombinant DNA technology as described in the literature of the art (e.g., Molecular Cloning: A Laboratory Manual, 2 nd Edition, J. Sambrook et al. eds., Cold Spring Harbor Laboratory Press, Cold Spring Harbor 1989).

[0069] Throughout this specification and the following claims, unless otherwise specifically required by context, the word “includes” and variations such as “includes” mean the inclusion of the member, integer or process or group of members, integers or processes described, and not the exclusion of any other member, integer or process or group of members, integers or processes; however, in some embodiments, such other members, integers or processes or groups of members, integers or processes may be excluded, i.e., it is understood that the subject matter lies in the inclusion of the member, integer or process or group of members, integers or processes described. The terms “one” and “it” and similar references used in connection with the description of the invention (particularly in connection with the claims) should be interpreted as including both singular and plural unless otherwise specifically indicated herein or unless otherwise clearly inconsistent with the context. Enumerations of numerical ranges in this specification are intended to serve as a simplified way of simply referring to each separate numerical value belonging to the range individually. Unless otherwise specifically indicated herein, each separate numerical value is incorporated herein as if it were enumerated individually.

[0070] All methods described herein may be carried out in any suitable order, unless otherwise specifically indicated herein or if it is clearly inconsistent with the context. The use of any examples or exemplary language provided herein (e.g., "etc.") is intended solely to better illustrate the invention and not to limit the scope of the invention or the claims. No language herein should be construed as indicating that any unclaimed element is essential for carrying out the invention.

[0071] Several sources are referenced throughout this specification. Each of the sources referenced above or below (including all patents, patent applications, academic publications, manufacturer specifications, instructions, etc.) is incorporated herein by reference in its entirety. Nothing in this specification should be construed as an acknowledgment that the present invention has no prior rights to such disclosures on the grounds of prior art.

[0072] The vaccines described herein are preferably recombinant vaccines.

[0073] In relation to the present invention, the term “recombinant” means “produced through genetic manipulation.” Preferably, in relation to the present invention, a “recombinant entity,” such as a recombinant polypeptide, is the result of a combination of entities, such as amino acid sequences or nucleic acid sequences, that do not exist in nature and preferably do not combine in nature. For example, in relation to the present invention, a recombinant polypeptide may comprise several amino acid sequences, such as a neoepitope or a vaccine sequence derived from different proteins or different parts of the same protein fused together, for example, by peptide bonds or a suitable linker.

[0074] As used herein, the term “naturally occurring” refers to the fact that a substance can be found in nature. For example, peptides or nucleic acids that are present in living organisms (including viruses), can be isolated from natural sources, and have not been intentionally modified by human hands in a laboratory are considered naturally occurring.

[0075] According to the present invention, the term "vaccine" refers to a pharmaceutical preparation (pharmaceutical composition) or product that, upon administration, induces an immune response, particularly a cellular immune response, that recognizes and attacks pathogens or abnormal cells such as cancer cells. Vaccines may be used for the prevention or treatment of disease. The term "personalized cancer vaccine" refers to a specific cancer patient, meaning that the cancer vaccine is adapted to the needs or specific circumstances of that individual cancer patient.

[0076] The term "immune response" refers to the integrated bodily response to an antigen, preferably a cellular immune response or a cellular and humoral immune response. An immune response can be protective / preventive / protective and / or therapeutic.

[0077] "Inducing an immune response" may mean that no immune response to a particular antigen existed before induction, but it may also mean that a certain level of immune response to a particular antigen existed before induction, and that the immune response is enhanced after induction. Therefore, "inducing an immune response" also includes "enhancing an immune response." Preferably, after inducing an immune response in a subject, the subject is protected from developing a disease such as cancer, or the disease state is improved by inducing an immune response. For example, an immune response to a tumor-expressing antigen can be induced in a patient with cancer or a subject at risk of developing cancer. In this case, inducing an immune response may mean that the subject's disease state is improved, the subject does not develop metastases, or a subject at risk of developing cancer does not develop cancer.

[0078] According to the present invention, the term "immune response to tumor antigen" refers to an immune response such as a cellular response to a tumor antigen or a cell that presents a tumor antigen, and includes an immune response to cells such as cancer cells that express and present a tumor antigen.

[0079] The terms "cellular immune response," "cellular response," "cellular response to antigen," or similar terms are intended to include cellular responses to cells characterized by the presentation of antigens by MHC class I or class II. The cellular response relates to cells called T cells or T lymphocytes that act as either "helper" or "killer" cells. Helper T cells (CD4 + T cells (also known as CTLs) play a central role in regulating the immune response, while killer cells (also known as cytotoxic T cells, cytolytic T cells, CD8+ T cells, or CTLs) kill abnormal cells such as cancer cells and prevent the production of further abnormal cells. In preferred embodiments, the present invention includes stimulating an antitumor CTL response against tumor cells that express one or more tumor expression antigens, preferably presenting such tumor expression antigens together with class I MHC.

[0080] The term "antigen" in the present invention encompasses any substance that induces an immune response. In particular, "antigen" refers to any substance, preferably a peptide or protein, that specifically reacts with an antibody or T lymphocyte (T cell). According to the present invention, the term "antigen" encompasses any molecule comprising at least one epitope. Preferably, in relation to the present invention, an antigen is a molecule that, after processing, preferably induces an immune response specific to the antigen (including cells expressing the antigen). According to the present invention, any suitable antigen that is a candidate for an immune response may be used, where the immune response is preferably a cellular immune response. In relation to embodiments of the present invention, the antigen is preferably presented by cells, preferably by antigen-presenting cells including abnormal cells, particularly cancer cells, in relation to an MHC molecule, which results in an immune response to the antigen. The antigen is preferably a product corresponding to or derived from a naturally occurring antigen. Such naturally occurring antigens include tumor antigens.

[0081] In preferred embodiments, the antigen is a tumor antigen, i.e., a part of a tumor cell such as a protein or peptide expressed in a tumor cell, which may originate from the cytoplasm, cell surface, or cell nucleus, and is particularly present primarily as an intracellular or surface antigen in the tumor cell. For example, tumor antigens include carcinoembryonic antigens, α1-fetoprotein, isoferritin and fetal sulfoglycoprotein, α2-H-ferrous protein and γ-fetoprotein. According to the present invention, tumor antigens preferably include any antigen expressed in tumors or cancers and in tumor cells or cancer cells, and which may be specific to tumors or cancers and specific to tumor cells or cancer cells with respect to type and / or expression level. In one embodiment, the terms “tumor antigen” or “tumor-associated antigen” refer to proteins that, under normal conditions, are specifically expressed in a limited number of tissues and / or organs or at a particular developmental stage, for example, tumor antigens may, under normal conditions, be specifically expressed in gastric tissue, preferably in the gastric mucosa, in reproductive organs, e.g., in the testes, in trophoblast tissue, e.g., in the placenta, or in germline cells, and may be expressed or abnormally expressed in one or more tumor or cancer tissues. In this context, “limited number” means preferably 3 or less, more preferably 2 or less. In relation to the present invention, tumor antigens include, for example, differentiation antigens, preferably cell type-specific differentiation antigens, i.e., proteins specifically expressed in specific cell types at specific developmental stages under normal conditions, cancer / testicular antigens, i.e., proteins specifically expressed in the testes and sometimes in the placenta under normal conditions, and germline-specific antigens. Preferably, tumor antigens or abnormal expression of tumor antigens identify cancer cells. In relation to the present invention, tumor antigens expressed by cancer cells in a subject, e.g., a patient suffering from a cancerous disease, are preferably autologous proteins in the subject. In preferred embodiments, in relation to the present invention, tumor antigens are specifically expressed in tissues or organs that are not essential under normal conditions, i.e., tissues or organs that, if damaged by the immune system, do not result in the death of the subject, or in organs or structures of the body that are not accessed or are hardly accessed by the immune system.

[0082] According to the present invention, the terms "tumor antigen," "tumor expression antigen," "cancer antigen," and "cancer expression antigen" are equivalent and are used interchangeably herein.

[0083] The term "immunogenicity" refers to the relative effectiveness of an antigen in inducing an immune response.

[0084] The “antigen peptide” according to the present invention preferably relates to a portion or fragment of an antigen that can stimulate an immune response, preferably a cellular response, to an antigen, or to an antigen characterized by antigen expression, preferably to abnormal cells, particularly cancer cells or other cells characterized by antigen presentation. Preferably, the antigen peptide can stimulate a cellular response to cells characterized by antigen presentation by class I MHC, and preferably to antigen-responsive cytotoxic T lymphocytes (CTLs). Preferably, the antigen peptide according to the present invention is an MHC class I and / or class II presenting peptide, or can be processed to produce an MHC class I and / or class II presenting peptide. Preferably, the antigen peptide comprises an amino acid sequence substantially corresponding to the amino acid sequence of an antigen fragment. Preferably, the aforementioned fragment of the antigen is an MHC class I and / or class II presenting peptide. Preferably, the antigen peptide according to the present invention comprises an amino acid sequence substantially corresponding to the amino acid sequence of such a fragment and is processed to produce such a fragment, i.e., an antigen-derived MHC class I and / or class II presenting peptide.

[0085] When a peptide is presented directly, i.e., without processing and especially without cleavage, it has a length suitable for binding to MHC molecules, particularly class I MHC molecules, preferably 7 to 20 amino acids long, more preferably 7 to 12 amino acids long, more preferably 8 to 11 amino acids long, and especially 9 or 10 amino acids long.

[0086] When the peptide is part of a larger entity containing an additional sequence, such as a vaccine sequence or polypeptide, and is presented after processing, particularly after cleavage, the peptide produced by processing has a length suitable for binding to MHC molecules, particularly class I MHC molecules, preferably 7 to 20 amino acids long, more preferably 7 to 12 amino acids long, more preferably 8 to 11 amino acids long, particularly 9 or 10 amino acids long. Preferably, the sequence of the peptide presented after processing is derived from the amino acid sequence of the antigen, i.e., its sequence substantially corresponds to, and preferably is completely identical to, a fragment of the antigen. Thus, in one embodiment, the antigen peptide or vaccine sequence according to the present invention comprises a sequence of 7 to 20 amino acids long, more preferably 7 to 12 amino acids long, more preferably 8 to 11 amino acids long, particularly 9 or 10 amino acids long, which substantially corresponds to, and preferably is completely identical to, a fragment of the antigen, and forms the peptide presented after processing of the antigen peptide or vaccine sequence. According to the present invention, such a peptide produced by processing contains the identified sequence change.

[0087] According to the present invention, an antigenic peptide or epitope may be present in a vaccine as part of a larger entity such as a vaccine sequence and / or polypeptide comprising two or more antigenic peptides or epitopes. The presented antigenic peptide or epitope is generated after appropriate processing.

[0088] A peptide having an amino acid sequence substantially corresponding to the sequence of a peptide presented by a class I MHC may differ in one or more residues that are not essential for TCR recognition of the peptide presented by the class I MHC or for peptide binding to the MHC. Such substantially corresponding peptides may also stimulate antigen-responsive CTLs and may be considered immunologically equivalent. A peptide having an amino acid sequence different from the presented peptide in residues that do not affect TCR recognition but improve the stability of binding to the MHC may improve the immunogenicity of the antigen peptide and may be referred to herein as an "optimized peptide." A reasonable approach to designing substantially corresponding peptides can be used by utilizing existing knowledge of which of these residues is more likely to affect binding to either the MHC or the TCR. The resulting functional peptide is intended to be the antigen peptide.

[0089] Antigen peptides should be recognizable by T cell receptors when presented by MHC. Preferably, when recognized by T cell receptors, antigen peptides can induce clonal proliferation of T cells carrying T cell receptors that specifically recognize the antigen peptide, in the presence of appropriate co-stimulatory signals. Preferably, when presented in conjunction with MHC molecules, antigen peptides can stimulate an immune response, preferably a cellular response, against cells characterized by the antigen from which they originate, or by the expression of the antigen, preferably by the presentation of the antigen. Preferably, antigen peptides can stimulate a cellular response against cells characterized by the presentation of the antigen by class I MHC, and preferably stimulate antigen-responsive CTLs. Such cells are preferably target cells.

[0090] "Antigen processing" or "processing" refers to the breakdown of peptides or proteins, such as polypeptides or antigens, into processing products, which are fragments of the peptide or protein (e.g., breakdown of polypeptides into peptides), and the association (e.g., by binding) of MHC molecules with one or more of these fragments for presentation to specific T cells by cells, preferably antigen-presenting cells.

[0091] Antigen-presenting cells (APCs) are cells that display peptide fragments of protein antigens associated with MHC molecules on their cell surface. Some APCs can activate antigen-specific T cells.

[0092] Professional antigen-presenting cells are highly efficient at internalizing antigens either through phagocytosis or receptor-mediated endocytosis, and then presenting antigen fragments bound to MHC class II molecules on their membranes. T cells recognize and interact with the antigen-MHC class II molecule complex on the membrane of the antigen-presenting cell. Subsequently, further costimulatory signals are generated by the antigen-presenting cell, leading to T cell activation. The expression of costimulatory molecules is a defining characteristic of professional antigen-presenting cells.

[0093] The main types of professional antigen-presenting cells are dendritic cells, macrophages, B cells, and certain activated epithelial cells, which have the broadest range of antigen presentation and are perhaps the most important antigen-presenting cells.

[0094] Dendritic cells (DCs) are a population of white blood cells that present antigens captured in peripheral tissues to T cells via both MHC class II and class I antigen presentation pathways. It is well known that dendritic cells are potent inducers of the immune response, and that the activation of these cells is an essential step in inducing antitumor immunity.

[0095] Dendritic cells are conveniently classified as “immature” and “mature” cells, and this can be used as a simple way to distinguish between two well-characterized phenotypes. However, this nomenclature should not be interpreted as excluding all possible intermediate stages of differentiation.

[0096] Immature dendritic cells are characterized as antigen-presenting cells with a high capacity for antigen uptake and processing, a capacity that correlates with high expression of Fcγ receptors and mannose receptors. Mature phenotypes are typically characterized by lower expression of these markers, but are also characterized by high expression of cell surface molecules responsible for T cell activation, such as MHC classes I and II, adhesion molecules (e.g., CD54 and CD11), and costimulatory molecules (e.g., CD40, CD80, CD86, and 4-1BB).

[0097] Dendritic cell maturation is referred to as the dendritic cell activation state in which such antigen-presenting dendritic cells lead to T cell priming, while presentation by immature dendritic cells results in tolerance. Dendritic cell maturation is primarily triggered by biomolecules with microbial characteristics detected by innate receptors (bacterial DNA, viral RNA, endotoxins, etc.), pro-inflammatory cytokines (TNF, IL-1, IFN), ligation of CD40 on the dendritic cell surface by CD40L, and substances released from cells that have undergone stress-induced cell death. Dendritic cells can be induced by culturing bone marrow cells in vitro with cytokines such as granulocyte-macrophage colony-stimulating factor (GM-CSF) and tumor necrosis factor α.

[0098] Nonprofessional antigen-presenting cells do not constitutively express MHC class II proteins necessary for interaction with naive T cells; these are expressed only after stimulation of nonprofessional antigen-presenting cells by certain cytokines such as IFNγ.

[0099] "Antigen-presenting cells" can be loaded with MHC class I-presenting peptides by transduction of a nucleic acid, preferably RNA, encoding a peptide or polypeptide containing the peptide to be presented, such as a nucleic acid encoding an antigen.

[0100] In some embodiments, a pharmaceutical composition containing a gene delivery vehicle targeting dendritic cells or other antigen-presenting cells may be administered to a patient to induce in vivo transfection. In vivo transfection of dendritic cells can generally be carried out using any method known in the art, such as the gene gun approach described by Mahvi et al., Immunology and Cell Biology 75:456-460, 1997, for example, as described in International Publication No. 97 / 24447.

[0101] According to the present invention, the term "antigen-presenting cell" also includes target cells.

[0102] "Target cells" means cells that are the target of an immune response, such as a cellular immune response. Target cells include cells that present an antigen or antigen epitope, i.e., a peptide fragment derived from an antigen, and include any undesirable cells, such as cancer cells. In preferred embodiments, target cells are cells that express the antigens described herein, and preferably present the antigen together with a class I MHC.

[0103] The term "epitope" refers to an antigenic determinant in a molecule such as an antigen, i.e., a portion or fragment of a molecule that is recognized by the immune system, for example, by T cells, when presented in relation to an MHC molecule. Epitopes of proteins such as tumor antigens preferably consist of a continuous or discontinuous portion of the protein and are preferably 5 to 100, preferably 5 to 50, more preferably 8 to 30, and most preferably 10 to 25 amino acids long. For example, an epitope may preferably be 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, or 25 amino acids long. In relation to the present invention, it is particularly preferable that the epitope be a T cell epitope.

[0104] According to the present invention, an epitope can bind to an MHC molecule, such as an MHC molecule on the surface of a cell, and is therefore an "MHC-binding peptide" or "antigen peptide." The term "MHC-binding peptide" refers to a peptide that binds to MHC class I and / or MHC class II molecules. For class I MHC / peptide complexes, the binding peptide is typically 8-10 amino acid long, but longer or shorter peptides may be effective. For class II MHC / peptide complexes, the binding peptide is typically 10-25 amino acid long, particularly 13-18 amino acid long, but longer and shorter peptides may be effective.

[0105] The terms “epitope,” “antigen peptide,” “antigen epitope,” “immunogenic peptide,” and “MHC-binding peptide” are used interchangeably herein and preferably relate to an incomplete form of an antigen that can induce an immune response to or to cells that express or contain the antigen, preferably presenting the antigen. Preferably, this term relates to the immunogenic portion of an antigen. Preferably, this is a portion of the antigen that is recognized (i.e., specifically bound) by a T cell receptor, particularly when presented in relation to an MHC molecule. A preferred immunogenic portion binds to an MHC class I or class II molecule. As used herein, an immunogenic portion is said to “bind” to an MHC class I or class II molecule if such binding is detectable using any assay known in the art.

[0106] As used herein, the term “neoepitope” refers to an epitope that is not present in reference cells, such as normal non-cancerous cells or germline cells, but is found in cancer cells. This includes, in particular, situations in which a corresponding epitope is present in normal non-cancerous cells or germline cells, but one or more mutations in the cancer cell alter the sequence of the epitope to produce a neoepitope.

[0107] The term “portion” refers to a fraction. With respect to an amino acid sequence or a particular structure such as a protein, the term “portion” can refer to a continuous or discontinuous fraction of the structure. Preferably, a portion of an amino acid sequence contains at least 1%, at least 5%, at least 10%, at least 20%, at least 30%, preferably at least 40%, preferably at least 50%, more preferably at least 60%, more preferably at least 70%, even more preferably at least 80%, and most preferably at least 90% of the amino acids of the amino acid sequence. Preferably, if the portion is a discontinuous fraction, the discontinuous fraction consists of two, three, four, five, six, seven, eight or more parts of the structure, each part being a continuous element of the structure. For example, a discontinuous fraction of the amino acid sequence may consist of 2, 3, 4, 5, 6, 7, 8 or more, preferably 4 or fewer, portions of the amino acid sequence, where each portion preferably contains at least 5 consecutive amino acids, at least 10 consecutive amino acids, preferably at least 20 consecutive amino acids, and preferably at least 30 consecutive amino acids.

[0108] The terms “part” and “fragment” are used interchangeably herein and refer to a continuous element. For example, a part of a structure such as an amino acid sequence or a protein refers to a continuous element of the said structure. A part, part, or fragment of a structure preferably contains one or more functional properties of the said structure. For example, a part, part, or fragment of an epitope, peptide, or protein is preferably immunologically equivalent to the epitope, peptide, or protein from which it is derived. In relation to the present invention, a “part” of a structure such as an amino acid sequence preferably comprises, and preferably consists of, at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 85%, at least 90%, at least 92%, at least 94%, at least 96%, at least 98%, or at least 99% of the whole structure or the whole amino acid sequence.

[0109] In the context of the present invention, the term "immunoreactive cell" relates to cells that exert effector functions during an immune response. An "immunoreactive cell" can preferably bind to cells characterized by the presentation of an antigen or an antigen peptide derived from an antigen, and can mediate an immune response. For example, such cells secrete cytokines and / or chemokines, secrete antibodies, recognize cancerous cells, and optionally eliminate such cells. For example, immunoreactive cells include T cells (cytotoxic T cells, helper T cells, tumor infiltrating T cells), B cells, natural killer cells, neutrophils, macrophages, and dendritic cells. Preferably, in the context of the present invention, the "immunoreactive cell" is a T cell, preferably CD4 + and / or CD8 + T cell.

[0110] Preferably, an "immunoreactive cell" recognizes with a certain degree of specificity an antigen or an antigen peptide derived from an antigen, particularly when presented in association with MHC molecules on the surface of abnormal cells such as antigen-presenting cells or cancer cells. Preferably, said recognition enables the cell that recognizes the antigen or the antigen peptide derived from said antigen to become responsive or reactive. When the cell is a helper T cell (CD4 + T cell) carrying a receptor that recognizes an antigen or an antigen peptide derived from an antigen in association with MHC class II molecules, such responsiveness or reactivity is the release of cytokines and / or CD8 +This may include activation of lymphocytes (CTLs) and / or B cells. If the cell is a CTL, such responsiveness or reactivity may include elimination of cells presented in relation to MHC class I molecules, i.e., cells characterized by antigen presentation by class I MHCs, for example, by apoptosis or perforin-mediated cytolysis. According to the present invention, CTL responsiveness may include sustained calcium flow, cell division, production of cytokines such as IFN-γ and TNF-α, upregulation of activation markers such as CD44 and CD69, and specific cytolytic death of target cells expressing the antigen. CTL responsiveness may also be determined by using an artificial reporter that accurately indicates CTL responsiveness. CTLs that recognize an antigen or antigen peptide derived from an antigen and are responsive or reactive are also referred to herein as “antigen-responsive CTLs.” If the cell is a B cell, such responsiveness may include the release of immunoglobulins.

[0111] The terms “T cell” and “T lymphocyte” are used interchangeably herein and include T helper cells (CD4+ T cells) and cytotoxic T cells (CTLs, CD8+ T cells), including cytolytic T cells.

[0112] T cells belong to the group of white blood cells known as lymphocytes and play a central role in cell-mediated immunity. They can be distinguished from other lymphocyte types, such as B cells and natural killer cells, by the presence of a specialized receptor on their cell surface called the T cell receptor (TCR). The thymus is the main organ responsible for the maturation of T cells. Several different subsets of T cells have been discovered, each with a distinct function.

[0113] T helper cells, among many other functions, assist other leukocytes in immunological processes, including the maturation of B cells into plasma cells and the activation of cytotoxic T cells and macrophages. These cells are also known as CD4+ T cells because they express the CD4 protein on their surface. Helper T cells are activated when presented with peptide antigens by MHC class II molecules expressed on the surface of antigen-presenting cells (APCs). Once activated, they rapidly divide and secrete small proteins called cytokines that modulate or assist the active immune response.

[0114] Cytotoxic T cells destroy virus-infected cells and tumor cells and are also involved in graft rejection. These cells express the CD8 glycoprotein on their surface and are therefore also known as CD8+ T cells. These cells recognize their targets by binding to antigens associated with MHC class I, which are present on the surface of almost every cell in the body.

[0115] Most T cells possess a T cell receptor (TCR), which exists as a complex of several proteins. The actual T cell receptor is produced from independent T cell receptor alpha and beta (TCRα and TCRβ) genes and consists of two separate peptide chains called the α-TCR chain and the β-TCR chain. Gamma delta T cells (γδT cells) are a small subset of T cells that have a different T cell receptor (TCR) on their surface. However, in γδT cells, the TCR consists of one γ chain and one δ chain. This group of T cells is far rarer than αβT cells (2% of all T cells).

[0116] The initial signal in T cell activation is given when the T cell receptor binds to a short peptide presented by the major histocompatibility complex (MHC) on another cell. This ensures that only T cells with a TCR specific to that peptide are activated. The partner cell is usually a professional antigen-presenting cell (APC), and in the case of a naive response, it is usually a dendritic cell, although B cells and macrophages can also be important APCs. Peptides presented to CD8+ T cells by MHC class I molecules are typically 8-10 amino acids long; peptides presented to CD4+ T cells by MHC class II molecules are typically longer, as the binding groove ends of MHC class II molecules are open.

[0117] According to the present invention, the T cell receptor has significant affinity to a predetermined target in a standard assay, and when it binds to the predetermined target, it can bind to the predetermined target. "Affinity" or "binding affinity" is often expressed as the equilibrium dissociation constant (K). D ) is measured by ). If a T cell receptor does not have significant affinity for the target in a standard assay and does not significantly bind to the target, it is unable to (substantially) bind to the target.

[0118] A T cell receptor is preferably capable of specifically binding to a predetermined target. A T cell receptor is specific to the predetermined target if it can bind to a predetermined target but cannot (substantially) bind to other targets; that is, if it does not have significant affinity for other targets in a standard assay and does not significantly bind to other targets.

[0119] Cytotoxic T lymphocytes can be generated in vivo by incorporating an antigen or antigen peptide into antigen-presenting cells in vivo. The antigen or antigen peptide may exist as a protein, as DNA (e.g., in a vector), or as RNA. Antigens can be processed to produce peptide partners for MHC molecules, but their fragments can be presented without requiring further processing, especially if they can bind to MHC molecules. Generally, administration to patients is possible by intradermal injection. However, injection can also be performed by intranodular injection into lymph nodes (Maloy et al. (2001), Proc Natl Acad Sci USA 98:3299-303). The resulting cells present a complex of interest, which is recognized by autocytotoxic T lymphocytes, which then proliferate.

[0120] Specific activation of CD4+ or CD8+ T cells can be detected in various ways. Methods for detecting specific T cell activation include detecting T cell proliferation, cytokine (e.g., lymphokine) production, or the development of cytolytic activity. For CD4+ T cells, a preferred method for detecting specific T cell activation is the detection of T cell proliferation. For CD8+ T cells, a preferred method for detecting specific T cell activation is the detection of the development of cytolytic activity.

[0121] The term "major histocompatibility complex" and the abbreviation "MHC" refer to a complex of genes present in all vertebrates, including MHC class I and MHC class II molecules. MHC proteins or molecules are crucial for signaling between lymphocytes and antigen-presenting or abnormal cells in immune responses, where they bind to peptides and present them for recognition by T cell receptors. Proteins encoded by MHC are expressed on the surface of cells and present both self-antigens (peptide fragments from the cell itself) and non-self-antigens (e.g., fragments of invading microorganisms) to T cells.

[0122] The MHC region is divided into three subgroups: Class I, Class II, and Class III. MHC Class I proteins contain the α chain and β2 microglobulin (which is not part of the MHC encoded by chromosome 15). These present antigen fragments to cytotoxic T cells. On most immune system cells, particularly antigen-presenting cells, MHC Class II proteins contain the α and β chains and present antigen fragments to T helper cells. The MHC Class III region encodes other immune components, such as complement components and some components encoding cytokines.

[0123] In humans, genes in the MHC region of the cell surface that encode antigen-presenting proteins are called human leukocyte antigen (HLA) genes. However, the abbreviation MHC is often used to refer to HLA gene products. HLA genes include nine so-called classical MHC genes: HLA-A, HLA-B, HLA-C, HLA-DPA1, HLA-DPB1, HLA-DQA1, HLA-DQB1, HLA-DRA, and HLA-DRB1.

[0124] In one preferred embodiment of all aspects of the present invention, the MHC molecule is an HLA molecule.

[0125] The terms "cells characterized by antigen presentation" or "cells that present antigens" or similar expressions refer to abnormal cells, such as cancer cells or antigen-presenting cells, that present antigens expressed by the cell or fragments derived from such antigens, in relation to MHC molecules, particularly MHC class I molecules, for example, by antigen processing. Similarly, the term "diseases characterized by antigen presentation" refers to diseases involving cells characterized by antigen presentation, particularly by class I MHCs. Antigen presentation by cells can be carried out by transfecting the cells with nucleic acids, such as RNA, that encode the antigen.

[0126] The phrase "fragment of the antigen to be presented" or similar expression means that the fragment may be presented by MHC class I or class II, preferably MHC class I, if, for example, it is directly added to an antigen-presenting cell. In one embodiment, the fragment is a fragment that is naturally presented by a cell expressing the antigen.

[0127] The term "immunologically equivalent" means that immunologically equivalent molecules, such as immunologically equivalent amino acid sequences, exhibit the same or essentially the same immunological properties and / or exert the same or essentially the same immunological effects with respect to the type of immunological action, such as induction of humoral and / or cellular immune responses, the strength and / or duration of the induced immune response, or the specificity of the induced immune response. In connection with the present invention, the term "immunologically equivalent" is preferably used with respect to the immunological action or properties of peptides used for immunity. For example, if an amino acid sequence induces an immune response that has specificity to react with a reference amino acid sequence when exposed to the immune system of a subject, then the amino acid sequence is immunologically equivalent to the reference amino acid sequence.

[0128] In relation to the present invention, the term “immune effector function” encompasses any function mediated by components of the immune system that results in inhibition of tumor growth and / or tumor development, including, for example, the death of tumor cells or the inhibition of tumor dissemination and metastasis. Preferably, in relation to the present invention, the immune effector function is a T cell-mediated effector function. Such functions include helper T cells (CD4 + In the case of T cells, recognition of antigens or antigen-derived antigen peptides associated with MHC class II molecules by T cell receptors, cytokine release, and / or CD8 +This includes activation of lymphocytes (CTLs) and / or B cells, and in the case of CTLs, it includes recognition of antigens or antigen peptides associated with MHC class I molecules by T cell receptors, elimination of cells presented in relation to MHC class I molecules, i.e., cells characterized by antigen presentation by class I MHC, for example by apoptosis or perforin-mediated cytolysis, production of cytokines such as IFN-γ and TNF-α, and specific cytolytic death of target cells expressing the antigen.

[0129] The term "genome" refers to the total amount of genetic information in the chromosomes of an organism or cell. The term "exome" refers to the coding region of the genome. The term "transcriptome" refers to the set of all RNA molecules.

[0130] According to the present invention, "nucleic acid" is preferably deoxyribonucleic acid (DNA) or ribonucleic acid (RNA), more preferably RNA, most preferably in vitro transcription RNA (IVT RNA) or synthetic RNA. According to the present invention, nucleic acids include genomic DNA, cDNA, mRNA, recombinantly produced molecules, and chemically synthesized molecules. According to the present invention, nucleic acids can exist as single-stranded or double-stranded linear or covalently bound closed cyclic molecules. According to the present invention, nucleic acids can be isolated. According to the present invention, the term "isolated nucleic acid" means that the nucleic acid has been (i) amplified in vitro, for example by polymerase chain reaction (PCR), (ii) recombinantly produced by cloning, (iii) purified, for example by cleavage and separation by gel electrophoresis, or (iv) synthesized, for example by chemical synthesis. Nucleic acids can be used for introduction into cells, i.e., cell transfection, in particular in the form of RNA which can be prepared by in vitro transcription from a DNA template. The RNA can be further modified before application by sequence stabilization, capping, and polyadenylation.

[0131] The term "genetic material" refers to an isolated nucleic acid, either DNA or RNA, a portion of a double helix, a portion of a chromosome, or the entire genome of an organism or cell, particularly its exome or transcriptome.

[0132] The term "mutation" refers to a change or difference (substitution, addition, or deletion of nucleotides) in a nucleic acid sequence compared to a reference. "Somatic mutations" can occur in any cell of the body except germ cells (sperm and egg cells) and are therefore not transmitted to offspring. These changes can (but not always) cause cancer or other diseases. Preferably, the mutation is a non-synonymous mutation. The term "non-synonymous mutation" refers to a mutation that results in an amino acid change, such as an amino acid substitution, in the translation product, preferably a nucleotide substitution.

[0133] According to the present invention, the term "mutation" encompasses point mutations, indels, fusions, chromothripsis, and RNA editing.

[0134] According to this invention, the term "indel" refers to a specific class of mutations defined as those resulting in colocalized insertions and deletions and a net increase or decrease in nucleotides. In the coding regions of the genome, if the length of an indel is not a multiple of 3, this results in a frameshift mutation. Indels can be contrasted with point mutations; while indels insert and delete nucleotides from a sequence, point mutations are a form of substitution that replaces one nucleotide.

[0135] Fusion can produce a hybrid gene formed from two previously separate genes. This can occur as a result of translocation, intermediate deletion, or chromosomal inversion. Often, fusion genes are oncogenes. Oncogenic fusion genes can yield gene products with new or different functions from their two fusion partners. Alternatively, a proto-oncogene may fuse with a strong promoter, thereby initiating oncogenic function through the upregulation of the upstream fusion partner by the strong promoter. Oncogenic fusion transcripts can also be caused by trans-splicing or read-through events.

[0136] According to the present invention, the term "chromothripsis" refers to a genetic phenomenon in which a specific region of the genome is disrupted by a single disruption event and then sutured together.

[0137] According to the present invention, the term "RNA editing" refers to a molecular process in which the information content in an RNA molecule is modified through chemical changes in its base composition. RNA editing includes nucleoside modifications such as deamination from cytidine (C) to uridine (U) and adenosine (A) to inosine (I), as well as non-template nucleotide addition and insertion. RNA editing in mRNA effectively alters the amino acid sequence of the encoded protein to differ from that predicted by the genomic DNA sequence.

[0138] The term "cancer mutation signature" refers to a set of mutations present in cancer cells compared to non-cancerous reference cells.

[0139] According to the present invention, a “reference” can be used to relate and compare results obtained from tumor specimens using the method of the present invention. Typically, a “reference” can be obtained based on one or more normal specimens, particularly specimens unaffected by cancer, obtained from a patient or one or more different individuals, preferably healthy individuals, especially individuals of the same species. A “reference” can be determined empirically by testing a sufficiently large number of normal specimens.

[0140] Any suitable sequencing method can be used in accordance with the present invention, and next-generation sequencing (NGS) technology is preferred. Third-generation sequencing methods may replace NGS technology in the future to expedite the sequencing process of the method. For clarification, in relation to the present invention, the terms “next-generation sequencing” or “NGS” mean all novel high-throughput sequencing technologies that read nucleic acid templates in parallel and randomly along the entire genome by dividing the entire genome into small pieces, as opposed to “conventional” sequencing methods known as the Sanger method. Such NGS technologies (also known as massively parallel sequencing technologies) can deliver nucleic acid sequence information of the whole genome, exome, transcriptome (all transcription sequences of the genome), or methylome (all methylated sequences of the genome) in a very short period of time, for example, within 1 to 2 weeks, preferably within 1 to 7 days, or most preferably within 24 hours, and in principle enable a single-cell sequencing approach. Several commercially available or literature-referenced NGS platforms, such as those detailed in Zhang et al. 2011: The impact of next-generation sequencing on genomics. J. Genet Genomics 38(3), 95-109; or Voelkerding et al. 2009: Next generation sequencing: From basic research to diagnostics. Clinical Chemistry 55, 641-658, can be used in connection with the present invention. Non-limiting examples of such NGS technologies / platforms are as follows: 1) For example, the "synthetic sequencing" technique known as a pyrosequencing method performed in the GS-FLX 454 Genome Sequencer® of Roche-affiliated 454 Life Sciences (Branford, Connecticut), first described in Ronaghi et al. 1998: A sequencing method based on real-time pyrophosphate. Science 281(5375), 363-365. This technique uses emulsion PCR, in which single-stranded DNA-binding beads are encapsulated in aqueous micelles containing PCR reaction products surrounded by oil for emulsion PCR amplification by vigorous vortexing. During the pyrosequencing process, as polymerase synthesizes DNA strands, light emitted from phosphate molecules during nucleotide incorporation is recorded. 2) A “synthetic sequencing” approach developed by Solexa (now part of Illumina Inc., San Diego, California), based on reversible dye-terminators, performed in, for example, the Illumina / Solexa Genome Analyzer® and the Illumina HiSeq 2000 Genome Analyzer®. In this technique, all four nucleotides are simultaneously added to an oligoprimed cluster fragment in a flow cell channel along with DNA polymerase. Crosslinking amplification extends the cluster chain, which has all four fluorescently labeled nucleotides, for sequencing. 3) For example, the “ligation-based sequencing” approach performed on the SOLid® platform of Applied Biosystems (now Life Technologies Corporation, Carlsbad, California). In this technique, a pool of all possible oligonucleotides of fixed length is labeled according to the sequenced positions. The oligonucleotides are annealed and ligated; selective ligation with DNA ligase to match the sequences yields a signal that gives information about the nucleotides at that position. Before sequencing, the DNA is amplified by emulsion PCR. The resulting beads, each containing only one copy of the same DNA molecule, are placed on a glass slide. As a second example, the Polonator® G.007 platform of Dover Systems (Salem, New Hampshire) also employs a “ligation-based sequencing” approach by amplified DNA fragments for parallel sequencing using emulsion PCR based on randomly placed beads. 4) Single-molecule sequencing technologies, such as those performed on the PacBio RS system at Pacific Biosciences (Menlo Park, California) or the HeliScope® platform at Helicos Biosciences (Cambridge, Massachusetts). A key feature of this technology is its ability to sequence a single DNA or RNA molecule without amplification, defined as single-molecule real-time (SMRT) DNA sequencing. For example, HeliScope uses a highly sensitive fluorescence detection system that directly detects each nucleotide as it is synthesized. A similar approach based on fluorescence resonance energy transfer (FRET) has been developed by Visigen Biotechnology (Houston, Texas). Other fluorescence-based single-molecule technologies come from USGenomics (GeneEngine®) and Genovoxx (AnyGene®). 5) Nanotechnology for single-molecule sequencing, for example, using various nanostructures placed on a chip to observe the movement of polymerase molecules on a single strand during replication. Non-limiting examples of nanotechnology-based approaches include the GridON® platform from Oxford Nanopore Technologies (Oxford, UK), the Hybridization-Assisted Nanopore Sequencing (HANS®) platform developed by Nabsys (Providence, Rhode Island), and a patented ligase-based DNA sequencing platform using DNA nanoball (DNB) technology called Combinatorial Probe Anchor Ligation (cPAL®). 6) Electron microscopy-based techniques for single-molecule sequencing, such as those developed by LightSpeed ​​Genomics (Sunnyvale, California) and Halcyon Molecular (Redwood City, California). 7) Ion semiconductor sequencing based on the detection of hydrogen ions released during DNA polymerization. For example, Ion Torrent Systems (San Francisco, California) uses a high-density array of micro-machined wells to carry out this biochemical process in ultra-parallel. Each well holds a different DNA template. Beneath the wells is an ion-sensitive layer, and below that is a patented ion sensor.

[0141] Preferably, DNA and RNA preparations serve as starting materials for NGS. Such nucleic acids can be readily obtained from samples such as biological materials, for example, from fresh, rapidly frozen, or formalin-fixed paraffin-embedded tumor tissue (FFPE), or from newly isolated cells, or from cytotoxic tumor cells (CTCs) present in the peripheral blood of patients. Normal, non-mutant genomic DNA or RNA can be extracted from normal body tissue, but germline cells are preferred in relation to the present invention. Germline DNA or RNA is extracted from peripheral blood mononuclear cells (PBMCs) in patients with non-hematological malignancies. Nucleic acids extracted from FFPE tissue or fresh, isolated single cells are highly fragmented, but they are suitable for NGS application.

[0142] Several targeted NGS methods for exome sequencing have been described in the literature (see, for example, Teer and Mullikin 2010: Human Mol Genet 19(2), R145-51 for a review), and all of them can be used in conjunction with the present invention. Many of these methods utilize hybridization techniques (described, for example, as genome capture, genome distribution, genome enrichment, etc.) and include array-based (e.g., Hodges et al. 2007: Nat. Genet. 39, 1522-1527) and liquid-based (e.g., Choi et al. 2009: Proc. Natl. Acad. Sci USA 106, 19096-19101) hybridization approaches. Commercial kits for DNA sample preparation and subsequent exome capture are also available; for example, Illumina Inc. (San Diego, California) offers the TruSeq® DNA Sample Preparation Kit and the TruSeq® Exome Enrichment Kit.

[0143] For example, when comparing the sequence of a tumor sample with the sequence of a reference sample, such as a germline sample, it is preferable to determine the sequences in one or both of these sample types to reduce the number of false positive findings when detecting cancer-specific somatic mutations or sequence differences. Therefore, it is preferable to determine the sequence of the reference sample, such as a germline sample, two, three, or more times. Alternatively, the sequence of the tumor sample may also be determined two, three, or more times. Furthermore, it may be possible to determine the sequence of the reference sample, such as a germline sample, and / or the sequence of the tumor sample two or more times by determining the sequence in the genomic DNA at least once and the sequence in the RNA of the reference sample and / or the tumor sample at least once. For example, by determining the mutations between replicas of a reference sample, such as a germline sample, the expected false positive rate (FDR) of somatic mutations as a statistical quantity can be estimated. Technical replication of one sample should produce the same result, and any mutation detected in this "same vs. same comparison" is a false positive. In particular, to determine the false detection rate for somatic mutation detection in tumor samples compared to a reference sample, technical replicates of a reference sample can be used as a reference to estimate the number of false positives. Furthermore, various quality-related metrics (e.g., coverage or SNP quality) can be combined into a single quality score using a machine learning approach. For a given somatic mutation, all other mutations with a higher quality score can be counted, which allows for the ranking of all mutations in the dataset.

[0144] According to the present invention, a high-throughput whole-genome single-cell genotyping method can be applied.

[0145] In one embodiment of high-throughput whole-genome single-cell genotyping, the Fluidigm platform may be used. Such an approach may include the following steps: 1. Obtain tumor tissue / cells and healthy tissue from the given patient. 2. Genetic material is extracted from cancerous and healthy cells, and then its exome (DNA) is sequenced using a standard next-generation sequencing (NGS) protocol. The NGS coverage is sufficient to detect heterozygous alleles with a frequency of at least 5%. Transcriptome (RNA) is also extracted from cancer cells, converted to cDNA, sequenced, and used to determine which genes are expressed by the cancer cells. 3. Identify the expressed non-synonymous single nucleotide variants (SNVs) as described herein. Filter out the SNP sites in the healthy tissue. 4. Select N=96 mutations from (3) at varying frequencies. Design and synthesize fluorescence detection-based SNP genotyping assays for these mutations (examples of such assays include the TaqMan-based SNP assay by Life Technologies or the SNPtype assay by Fluidigm). The assays include specific targeted amplification (STA) primers for amplifying the amplicon containing a given SNV (this is standard in TaqMan and SNPtype assays). 5. Individual cells are isolated from tumor and healthy tissue by either laser microdissection (LMD) or separation into single-cell suspension, and then sorted as previously described (Dalerba P. et al. (2011) Nature Biotechnology 29:1120-1127). Cells can be selected without pre-selection (i.e., unbiased), or cancer cells can be enriched. Enrichment methods include specific staining, sorting by cell size, and histological examination in LMD. 6. Isolate individual cells in a PCR tube containing the master mix and STA primers, and amplify the amplicons containing SNVs. Alternatively, amplify the genome of a single cell by whole-genome amplification (WGA) as previously described (Frumkin D. et al. (2008) Cancer Research 68:5924). Achieve cell lysis by either a heating step at 95°C or a dedicated lysis buffer. 7. Dilute the STA-amplified sample and load it onto the Fluidigm genotyping array. 8. Use samples from healthy tissue as a positive control to determine homozygous allele clusters (no mutations). Since NGS data indicate that homozygous mutations are extremely rare, typically only two clusters are predicted: XX and XY, where X = healthy. 9. There is no limit to the number of arrays that can be run; in practice, it is possible to test up to approximately 1000 single cells (approximately 10 arrays). When performed with 384 plates, sample preparation can be reduced to a few days. 10. Next, determine the SNV for each cell.

[0146] Another embodiment of high-throughput whole-genome single-cell genotyping may utilize an NGS platform. Such an approach may include the following steps: 1. Steps 1 through 6 above are identical except that N (the number of SNVs being tested) may be much greater than 96. In the case of WGA, several cycles of STA are then performed. The STA primers contain two universal tag sequences on each primer. 2. After STA, the barcode primers are PCR-amplified into amplicons. The barcode primers contain a unique barcode sequence and the universal tag sequence described above. Therefore, each cell contains a unique barcode. 3. Mix the amplicons from all cells and sequence them by NGS. The practical limit to the number of cells that can be duplicated is the number of plates that can be prepared. Since the sample can be prepared in 384 plates, the practical limit is approximately 5000 cells. 4. Detect SNVs (or other structural abnormalities) in individual cells based on sequence data.

[0147] To determine antigen prioritization, tumor phylogenetic reconstruction based on single-cell genotyping ("phylogenetic antigen prioritization") may be used in accordance with the present invention. In addition to antigen prioritization based on criteria such as expression, mutation type (non-synonymous vs. other mutations), and MHC binding properties, further dimensions of prioritization designed to address intratumoral and intertumoral heterogeneity and biopsy bias may be used, for example, as described below.

[0148] 1. Identification of the most abundant antigens Based on the single-cell assay described above in relation to high-throughput whole-genome single-cell genotyping, the frequency of each SNV can be accurately estimated, and the most abundant SNVs present can be selected to provide personalized cancer vaccines (IVACs).

[0149] 2. Identification of primary basal antigens based on rooted tree analysis NGS data from tumors suggest that homozygous mutations (hits in both alleles) are a rare event. Therefore, haplotyping is not necessary, and a phylogenetic tree of tumor somatic mutations can be constructed from single-cell SNV datasets. The phylogenetic tree is rooted using germline sequences. Algorithms for reconstructing ancestral sequences are used to reconstruct sequences from nodes near the root of the phylogenetic tree. These sequences contain the earliest mutations predicted to be present in the primary tumor (defined herein as primary basal mutations / antigens). Because the probability of two mutations occurring in the same allele at the same location on the genome is low, mutations in the ancestral sequences are predicted to be fixed in the tumor.

[0150] Prioritizing the primary basal antigen is not equivalent to prioritizing the most frequent mutation in the biopsy (although the primary basal mutation is expected to be one of the most frequent in the biopsy). The reason is as follows: For example, if two SNVs are thought to be present in all cells from the biopsy (and therefore have the same frequency, i.e., 100%), but one mutation is basal and the other is not, the basal mutation should be selected for IVAC. This is because the basal mutation is likely to be present in all areas of the tumor, while the latter mutation may be a more recent mutation that happened to be fixed in the area from which the biopsy was taken. In addition, the basal antigen is likely to be present in metastatic tumors originating from the primary tumor. Therefore, prioritizing the basal antigen for IVAC can greatly increase the likelihood that IVAC can eradicate not just a part of the tumor but the entire tumor.

[0151] If secondary tumors are present and collected, it is possible to estimate the evolutionary tree of all tumors. This can improve the robustness of the phylogenetic tree and enable the detection of underlying mutations in all tumors.

[0152] 3. Identification of antigens that span the tumor to the greatest extent possible. Another approach to obtaining antigens that provide maximum coverage of all tumor sites is to take several biopsies from the tumor. One strategy is to select antigens identified as present in all biopsies by NGS analysis. To improve the probability of identifying basal mutations, phylogenetic analysis based on single-cell mutations can be performed from all biopsies.

[0153] In cases of metastasis, biopsies can be obtained from all tumors, and mutations common to all tumors, identified by NGS, can be selected.

[0154] 4. Use of CTCs to prioritize antigens that inhibit metastasis Metastatic tumors are thought to originate from single cells. Therefore, by genotyping individual cells extracted from various tumors in a given patient, as well as genotyping circulating tumor cells (CTCs) in the patient's blood, it is possible to reconstruct the evolutionary history of cancer. It is expected that metastatic tumors evolving from the original tumor can be observed through the clade of CTCs originating from the primary tumor.

[0155] The following (unbiased method for identifying, counting, and genetically probing CTCs) describes an extension of the high-throughput whole-genome single-cell genotyping method described above for unbiased isolation and genomic analysis of CTCs. Using the analysis described above, a phylogenetic tree can then be reconstructed of the primer tumor, CTCs, and secondary tumors (if any) arising from metastases. Based on this phylogenetic tree, mutations (passenger or driver) that occurred at or immediately after the initial separation of the CTC from the primary tumor can be identified. The genomes of CTCs arising from primary tumors are expected to be more evolutionarily similar to the primary tumor genome than the secondary tumor genomes. Furthermore, the genomes of CTCs arising from primary tumors are expected to contain unique mutations that are fixed in the secondary tumor, or are likely to be fixed if secondary tumors form in the future. These unique mutations can be prioritized for IVAC targeting (or preventing) metastases.

[0156] The advantage of prioritizing CTC mutations over primary basal mutations is that antigens derived from CTCs can recruit T cells to specifically target metastases, thus becoming an independent weapon (using different antigens) from T cells targeting primary tumors. In addition, if secondary tumors are few (or none) present, the probability of tumor evasion should correspond to the number of cancer cells carrying a given antigen, so the likelihood of immune evasion from CTC-derived antigens is expected to be lower.

[0157] 5. Identification of co-occurring antigens on the same cell ("cocktail" IVAC) Tumors are thought to evolve to suppress mutations resulting from the immune system and therapeutic selection pressures. Cancer vaccines targeting multiple antigens that co-occur on the same cells and are frequently present in tumors have a greater potential to neutralize tumor evasion mechanisms and thus reduce the likelihood of recurrence. Such “cocktail vaccines” are analogous to antiretroviral combination therapy for HIV-positive patients. Co-occurring mutations can be identified by phylogenetic analysis or by examining SNV alignments of all cells.

[0158] Furthermore, according to the present invention, an unbiased method can be used to identify, count, and gene probe CTCs. Such an approach may include the following steps: 1. Obtain a biopsy of the tumor and determine the somatic mutation map. 2. Option 1: Select mutations with N ≥ 96 for further consideration based on the previously established prioritization scheme. Option 2: Perform a single-cell assay (see the high-throughput whole-genome single-cell genotyping method described above), followed by phylogenetic analysis, selecting primary basal mutations with N≧96 and, if applicable, recent mutations to maximize diversity. The former mutations are useful for identifying CTCs (see below), and the latter for performing phylogenetic analysis (see the chapter "Identification of Co-occurring Antigens on the Same Cell ("Cocktail" IVAC)"). 3. Obtain whole blood from a cancer patient. 4. Dissolve the red blood cells. 5. CD45 + Leukocytes are removed by depleting the cells (for example, by sorting, using magnetic beads bound to anti-CD45 antibodies, etc.), and CTCs are concentrated. 6. Remove free DNA by DNAase digestion. The source of free DNA can be DNA present in the blood or DNA from dead cells. 7. The remaining cells are sorted into PCR tubes and subjected to STA (based on selected mutations), followed by screening with Fluidigm (the high-throughput whole-genome single-cell genotyping method described above). CTCs should generally be positive for multiple SNVs. 8. Next, based on the screened panel of SNVs, cancerous cells (=CTCs) can be further analyzed phylogenetically (see the chapter "Identification of Co-occurring Antigens on the Same Cell ("Cocktail" IVAC)").

[0159] Furthermore, this method can be combined with previously established methods for isolated CTCs. For example, EpCAM+ cells or cells positive for cytokeratin can be selected (Rao CG. et al. (2005) International Journal of Oncology 27:49; Allard WJ. et al. (2004) Clinical Cancer Research 10:6897-6904). These putative CTCs can then be validated / profiled using Fluidigm / NGS to derive their mutations.

[0160] This method can be used to count CTCs. This method is an unbiased way to detect and count CTCs because it is based on the patient's unique mutation profile of cancer somatic mutations, rather than relying on a single specific marker that may or may not be expressed by cancer cells.

[0161] According to the present invention, an approach including tumor phylogenetic reconstruction based on single-cell genotyping ("phylogenetic filtering") may be used to enrich driver mutations.

[0162] In one embodiment of this approach, a pantumor phylogenetic analysis is performed to recover the driver mutation.

[0163] For example, driver mutations can be detected from a tumor with n=1.

[0164] The chapter "Identification of Primary Basal Antigens Based on Rooted Tree Analysis" describes a method for identifying cells that have sequences that restore ancestral sequences and / or sequences close to the roots of a tree. Since these are, by definition, sequences close to the roots of a tree, the number of mutations in these sequences is expected to be significantly less than the number of mutations in the bulk sample of cancer. Therefore, by selecting sequences close to the roots of the tree, many passenger mutations are expected to be "phylogenetically filtered out". This procedure has the potential to greatly enrich driver mutations. Driver mutations can then be used to identify / select treatments for patients or as clues for novel therapies.

[0165] In another example, driver mutations can be detected from n>1 tumors of a given type.

[0166] Reconstructing primary basal mutations from a large number of tumors of a specific type can significantly increase the likelihood of detecting driver mutations. Since basal sequences near the root of the tree filter out many passenger mutations, the signal-to-noise ratio is expected to increase significantly when detecting driver mutations. Therefore, this method has the potential to (1) detect less frequent driver mutations and (2) detect high-frequency driver mutations from fewer samples.

[0167] Another embodiment of the approach, which includes tumor phylogenetic reconstruction based on single-cell genotyping to enrich driver mutations ("phylogenetic filtering"), involves performing phylogenetic analysis to recover metastases that cause driver mutations.

[0168] The chapter "Use of CTCs to Prioritize Antigens that Inhibit Metastasis" describes a method for detecting CTC-associated mutations. This method can also be used to enrich driver mutations that lead to metastasis. For example, by mapping the combined phylogeny of primer tumors, secondary tumors, and CTCs, CTCs originating from primary tumors should bridge the gap between primary and secondary tumor clades. Such phylogenetic analysis can help pinpoint intrinsic mutations in this transition between primer tumors and secondary tumors. One fraction of these mutations may be driver mutations. Furthermore, by comparing intrinsic CTC mutations from different cases of the same cancer (i.e., tumors n>1), the intrinsic driver mutations that cause metastasis can be further enriched.

[0169] According to the present invention, phylogenetic analysis can be used to identify primary tumors versus secondary tumors.

[0170] In cases of metastasis, if all tumors are collected, the rooted tree can be used to predict the chronological order in which the tumors appeared, i.e., which tumor is the primary tumor (the node closest to the root of the tree) and which is the most recent. This can be useful when it is difficult to determine which tumor is primary.

[0171] In connection with the present invention, the term “RNA” refers to a molecule comprising, and preferably entirely or substantially, ribonucleotide residues. “Ribonucleotide” refers to a nucleotide having a hydroxyl group at the 2' position of a β-D-ribofuranosyl group. The term “RNA” includes isolated RNA such as double-stranded RNA, single-stranded RNA, partially or completely purified RNA, essentially pure RNA, synthetic RNA, and recombinant RNA such as modified RNA that differs from naturally occurring RNA by the addition, deletion, substitution and / or alteration of one or more nucleotides. Such alterations may include, for example, the addition of non-nucleotide substances to or within one or both ends of RNA, for example, one or more nucleotides of RNA. Nucleotides in RNA molecules may also include non-standard nucleotides, such as nucleotides that do not exist in nature, or chemically synthesized nucleotides or deoxynucleotides. These altered RNAs may be referred to as analogs or analogs of naturally occurring RNA.

[0172] According to the present invention, the term "RNA" encompasses and preferably relates to "mRNA". The term "mRNA" means "messenger RNA" and relates to a "transcript" that is produced using a DNA template and codes for a peptide or protein. Typically, mRNA contains a 5'-UTR, a protein-coding region, and a 3'-UTR. mRNA has a limited half-life in cells and in vitro. In relation to the present invention, mRNA can be produced from a DNA template by in vitro transcription. Methods of in vitro transcription are known to those skilled in the art. For example, various in vitro transcription kits are commercially available.

[0173] According to the present invention, the stability and translation efficiency of RNA can be modified as needed. For example, RNA can be stabilized and its translation efficiency increased by one or more modifications that have a stabilizing effect on RNA and / or enhance its translation efficiency. Such modifications are described, for example, in PCT / EP2006 / 009448, which is incorporated herein by reference. In order to increase the expression of RNA used in accordance with the present invention, the RNA can be modified to enhance mRNA stability by increasing the GC content, performing codon optimization, and thus enhancing translation in cells, preferably without altering the sequence of the expressed peptide or protein, within the coding region, i.e., within the sequence encoding the expressed peptide or protein.

[0174] In relation to RNA used in this invention, the term "modification" encompasses any modification of the RNA that is not naturally present in the RNA.

[0175] In one embodiment of the present invention, the RNA used according to the present invention does not have uncapped 5'-triphosphates. Removal of such uncapped 5'-triphosphates can be achieved by treating the RNA with a phosphatase.

[0176] The RNA according to the present invention may have ribonucleotides modified to enhance its stability and / or reduce its cytotoxicity. For example, in one embodiment, cytidine is partially or completely, preferably completely, substituted with 5-methylcytidine in the RNA used according to the present invention. Or, in addition, in one embodiment, uridine is partially or completely, preferably completely, substituted with pseudouridine in the RNA used according to the present invention.

[0177] In one embodiment, the term “modification” relates to providing RNA with a 5'-cap or a 5'-cap analogue. The term “5'-cap” refers to a cap structure found at the 5' end of an mRNA molecule, generally consisting of a guanosine nucleotide linked to the mRNA by a distinctive 5'-5' triphosphate bond. In one embodiment, this guanosine is methylated at position 7. The term “conventional 5'-cap” refers to a naturally occurring RNA 5'-cap, preferably a 7-methylguanosine cap (m 7 G) refers to the present invention. In relation to the present invention, the term “5'-cap” includes 5'-cap analogues that are similar to RNA cap structures and are modified to have the ability to stabilize RNA and / or enhance RNA translation when bound to RNA, preferably in vivo and / or in cells.

[0178] Preferably, the 5' end of the RNA has the following general formula: [ka] [In the formula, R1 and R2 are independently hydroxyl or methoxy, and W - , X - and Y - [These are independently oxygen, sulfur, selenium, or BH3] It includes a cap structure having a hydroxyl group. In a preferred embodiment, R1 and R2 are hydroxyl groups, and W - , X - and Y - is oxygen. In a further preferred embodiment, one of R1 and R2, preferably R1 is hydroxyl and the other is methoxy, and W - , X - and Y - is oxygen. In a further preferred embodiment, R1 and R2 are hydroxyl, and W - , X - and Y - One of them, preferably X -is sulfur, selenium, or BH3, preferably sulfur, and otherwise oxygen. In a further preferred embodiment, one of R1 and R2, preferably R2 is hydroxyl and the other is methoxy, and W - , X - and Y - One of them, preferably X - The other is sulfur, selenium, or BH3, preferably sulfur, and the other is oxygen.

[0179] In the above formula, the nucleotide on the right is attached to the RNA chain via its 3' group.

[0180] W - , X - and Y - Cap structures having at least one sulfurous component, i.e., a phosphorothioate moiety, exist in various diastereoisomer forms, all of which are incorporated herein. Furthermore, the present invention encompasses all tautomers and stereoisomers of the above formula.

[0181] For example, if R1 is methoxy and R2 is hydroxy, then X - is sulfur, and W - and Y - The cap structure having the above structure in which is oxygen exists in two diastereoisomer forms (Rp and Sp). These can be separated by reversed-phase HPLC and are named D1 and D2 according to the elution order from the reversed-phase HPLC column. According to the present invention, m2 7,2'-O Gpp S The D1 isomer of pG is particularly preferred.

[0182] Providing a 5'-cap or 5'-cap analogue to RNA can be achieved by in vitro transcription of a DNA template in the presence of the 5'-cap or 5'-cap analogue, by co-transcribement of the 5'-cap into the prepared RNA strand, or by preparing the RNA, for example, by in vitro transcription, and then binding the 5'-cap to the RNA after transcription using a capping enzyme, such as the capping enzyme of vaccinia virus.

[0183] The RNA may undergo further modifications. For example, further modifications of the RNA used in the present invention may include changes to the 5'UTR or 3'UTR, such as elongation or termination of the naturally occurring poly(A) tail, or introduction of an untranslated region (UTR) unrelated to the coding region of the RNA, such as the exchange of one or more copies, preferably two copies, of the existing 3'UTR with a globin gene, such as α2-globin, α1-globin, β-globin, preferably β-globin, more preferably human β-globin, or the insertion of one or more copies, preferably two copies, of the 3'UTR derived from the globin gene.

[0184] RNA with an unmasked poly(A) sequence is translated more efficiently than RNA with a masked poly(A) sequence. The terms "poly(A) tail" or "poly(A) sequence" typically refer to the sequence of adenyl (A) residues located at the 3' end of an RNA molecule, while an "unmasked poly(A) sequence" means that the poly(A) sequence at the 3' end of the RNA molecule ends with an A, and is not followed by any non-A nucleotides downstream of the 3' end of the poly(A) sequence. Furthermore, a long poly(A) sequence of approximately 120 base pairs results in optimal transcript stability and translation efficiency for RNA.

[0185] Therefore, in order to increase the stability and / or expression of the RNA used according to the present invention, the RNA may be modified to be present with a polyA sequence having a length of preferably 10 to 500, more preferably 30 to 300, even more preferably 65 to 200, and especially 100 to 150 adenosine residues. In a particularly preferred embodiment, the polyA sequence has a length of approximately 120 adenosine residues. In order to further increase the stability and / or expression of the RNA used according to the present invention, the polyA sequence may be demasked.

[0186] In addition, the incorporation of a 3' untranslated region (UTR) into another 3' untranslated region (UTR) of an RNA molecule can lead to increased translation efficiency. Synergistic effects can be achieved by incorporating two or more such 3' untranslated regions. The 3' untranslated regions may be self- or heterologous to the RNA into which they are introduced. In one particular embodiment, the 3' untranslated region is derived from the human β-globin gene.

[0187] The aforementioned modifications, namely the incorporation of poly(A) sequences, the demasking of poly(A) sequences, and the incorporation of one or more 3' untranslated regions, have a synergistic effect on increasing RNA stability and translation efficiency.

[0188] The term "stability" of RNA refers to its "half-life." Half-life refers to the time required to remove half of the activity, quantity, or number of molecules. In relation to this invention, the half-life of RNA is an indicator of its stability. The half-life of RNA can affect the "duration of expression" of RNA. RNA with a long half-life can be expected to be expressed for a long period.

[0189] Needless to say, according to the present invention, when it is desirable to reduce the stability and / or translation efficiency of RNA, it is possible to modify the RNA in such a way as to interfere with the function of the aforementioned elements that increase the stability and / or translation efficiency of RNA.

[0190] The term "expression" is used in its most general sense according to the present invention and includes, for example, the production of RNA and / or peptides or polypeptides by transcription and / or translation. With respect to RNA, the terms "expression" or "translation" particularly refer to the production of peptides or polypeptides. It also includes partial expression of nucleic acids. Furthermore, expression can be transient or stable.

[0191] According to the present invention, the term "expression" also includes "ectopic expression" or "abnormal expression." According to the present invention, "ectopic expression" or "abnormal expression" means that the expression is altered, preferably increased, compared to a reference, e.g., a subject without disease associated with ectopic or abnormal expression of a particular protein, e.g., a tumor antigen. Increased expression refers to an increase of at least 10%, particularly at least 20%, at least 50%, or at least 100% or more. In one embodiment, expression is observed only in affected tissue, and expression in healthy tissue is suppressed.

[0192] The term "specifically expressed" means that a protein is expressed essentially only in a specific tissue or organ. For example, a tumor antigen specifically expressed in the gastric mucosa means that the protein is primarily expressed in the gastric mucosa and not expressed in other tissues or to a significant degree in other tissue or organ types. Therefore, a protein that is exclusively expressed in gastric mucosal cells and expressed to a significantly lower degree in other tissues such as the testes is specifically expressed in gastric mucosal cells. In some embodiments, a tumor antigen may also be specifically expressed in two or more tissue types or organs, for example, in two or three tissue types or organs, but preferably three or fewer different tissue or organ types, under normal conditions. In this case, the tumor antigen is specifically expressed in these organs. For example, if a tumor antigen is expressed to approximately the same degree in the lungs and stomach, preferably under normal conditions, then the tumor antigen is specifically expressed in the lungs and stomach.

[0193] In relation to the present invention, the term “transcription” refers to the process by which the genetic code in a DNA sequence is transcribed into RNA. The RNA can then be translated into a protein. According to the present invention, the term “transcription” includes “in vitro transcription,” where “in vitro transcription” refers to the process by which RNA, particularly mRNA, is synthesized in vitro in a cell-free system, preferably using a suitable cell extract. Preferably, a cloning vector is applied to the production of the transcript. These cloning vectors are generally referred to as transcription vectors and, according to the present invention, are encompassed in the term “vector.” According to the present invention, the RNA used in the present invention is preferably in vitro transcription RNA (IVT RNA), which can be obtained by in vitro transcription of a suitable DNA template. The promoter for controlling transcription can be any promoter for any RNA polymerase. Specific examples of RNA polymerases are T7, T3, and SP6 RNA polymerases. Preferably, in vitro transcription according to the present invention is controlled by a T7 or SP6 promoter. A DNA template for in vitro transcription can be obtained by cloning a nucleic acid, particularly cDNA, and introducing it into a suitable vector for in vitro transcription. cDNA can be obtained by reverse transcription of RNA.

[0194] The term "translation" in this invention relates to a process in a cell's ribosome in which a chain of messenger RNA instructs the assembly of an amino acid sequence to produce a peptide or polypeptide.

[0195] According to the present invention, the expression regulatory or regulatory sequence that can functionally link to a nucleic acid may be homogeneous or heterogeneous with respect to the nucleic acid. The coding sequence and the regulatory sequence are "functionally" linked together if they are covalently bonded together such that the transcription or translation of the coding sequence is under the control or influence of the regulatory sequence. When the functional linkage of the coding sequence and the regulatory sequence results in the coding sequence being translated into a functional protein, the induction of the regulatory sequence results in the transcription of the coding sequence without causing a reading frame shift of the coding sequence or making it impossible for the coding sequence to be translated into the desired protein or peptide.

[0196] The terms “expression regulatory sequence” or “regulatory sequence” include, according to the present invention, promoters, ribosome-binding sequences, and other regulatory elements that control the transcription of nucleic acids or the translation of induced RNA. In certain embodiments of the present invention, regulatory sequences can be controlled. The exact structure of regulatory sequences may vary depending on the species or cell type, but generally include 5' non-transcription sequences and 5' and 3' non-translating sequences involved in the initiation of transcription or translation, such as TATA boxes, capping sequences, CAAT sequences, etc. In particular, 5' non-transcription regulatory sequences include promoter regions containing promoter sequences for the transcriptional control of functionally bound genes. Regulatory sequences may also include enhancer sequences or upstream activating sequences.

[0197] Preferably, according to the present invention, RNA to be expressed in the cell is introduced into the cell. In one embodiment of the method according to the present invention, the RNA to be introduced into the cell is obtained by in vitro transcription of a suitable DNA template.

[0198] According to the present invention, terms such as “expressible RNA” and “coding RNA” are used interchangeably herein and mean that, with respect to a particular peptide or polypeptide, the RNA can be expressed to produce the peptide or polypeptide when present in a suitable environment, preferably within a cell. Preferably, the RNA according to the present invention can interact with the cellular translation mechanism to provide the peptide or polypeptide that the RNA can express.

[0199] Terms such as “import,” “introduce,” or “transfect” are used interchangeably herein and relate to the introduction of nucleic acids, particularly exogenous or heterologous nucleic acids, especially RNA, into cells. According to the present invention, cells may form organs, tissues, and / or parts of organisms. According to the present invention, the administration of nucleic acids is achieved as naked nucleic acids or in combination with an administration reagent. Preferably, the administration of nucleic acids is in the form of naked nucleic acids. Preferably, RNA is administered in combination with a stabilizing substance such as an RNase inhibitor. The present invention also envisions repeated introduction of nucleic acids into cells to enable long-term sustained expression.

[0200] RNA can be transfected using any carrier that can bind to RNA, for example, by forming a complex with other RNA or by forming a vesicle in which RNA is encapsulated or enclosed, resulting in increased stability of the RNA compared to naked RNA. Useful carriers according to the present invention include, for example, lipid-containing carriers, such as cationic lipids, liposomes, particularly cationic liposomes, micelles, and nanoparticles. Cationic lipids can form complexes with negatively charged nucleic acids. Any cationic lipid can be used according to the present invention.

[0201] Preferably, the introduction of RNA encoding a peptide or polypeptide into cells, particularly cells present in vivo, results in the expression of the peptide or polypeptide within the cells. In certain embodiments, targeting of nucleic acids to specific cells is preferred. In such embodiments, the carrier (e.g., retrovirus or liposome) applied to the administration of nucleic acids to cells represents the targeting molecule. For example, molecules such as antibodies specific to surface membrane proteins on target cells or ligands for receptors on target cells may be incorporated into or bound to the nucleic acid carrier. When nucleic acids are administered by liposomes, proteins that bind to surface membrane proteins associated with endocytosis may be incorporated into the liposomal formulation to enable targeting and / or uptake. Such proteins include capsid proteins or fragments specific to a particular cell type, antibodies against proteins to be internalized, proteins that target intracellular locations, and the like.

[0202] According to the present invention, the term "peptide" refers to a substance containing two or more, preferably three or more, preferably four or more, preferably six or more, preferably eight or more, preferably ten or more, preferably thirteen or more, preferably sixteen or more, preferably twenty-one or more, and preferably eight, ten, twenty-five, or fifty, and particularly up to 100 amino acids, which are covalently linked by peptide bonds. The terms "polypeptide" or "protein" refer to larger peptides, preferably peptides having more than 100 amino acid residues, but generally the terms "peptide," "polypeptide," and "protein" are synonymous and are used interchangeably herein.

[0203] According to the present invention, the term “sequence change” in relation to peptides or proteins refers to amino acid insertion mutants, amino acid addition mutants, amino acid deletion mutants, and amino acid substitution mutants, preferably amino acid substitution mutants. All of these sequence changes according to the present invention can potentially generate new epitopes.

[0204] Amino acid insertion mutants involve the insertion of one or more amino acids in a specific amino acid sequence.

[0205] Amino acid addition mutants include amino-terminus and / or carboxyl-terminus fusions of one or more amino acids, such as 1, 2, 3, 4, or 5 or more amino acids.

[0206] Amino acid deletion mutants are characterized by the removal of one or more amino acids from a sequence, for example, the removal of one, two, three, four, or five or more amino acids.

[0207] An amino acid substitution mutant is characterized by the removal of at least one residue in the sequence and the insertion of another residue in its place.

[0208] The term “derived” means, according to the present invention, that a particular entity, in particular a particular sequence, is present in the object from which it is derived, in particular a living organism or molecule. In the case of an amino acid sequence, in particular a particular sequence region, “derived” means in particular that the relevant amino acid sequence is derived from the amino acid sequence in which it is present.

[0209] The terms “cell” or “host cell” preferably refer to an intact cell, i.e., a cell with an intact membrane from which its normal intracellular components, such as enzymes, organelles, or genetic material, have not been released. An intact cell is preferably a viable cell, i.e., a living cell capable of performing its normal metabolic functions. Preferably, the terms refer to any cell that can be transformed or transfected with exogenous nucleic acids according to the present invention. The term “cell” according to the present invention includes prokaryotic cells (e.g., Escherichia coli (E. coli)) or eukaryotic cells (e.g., dendritic cells, B cells, CHO cells, COS cells, K562 cells, HEK293 cells, HELA cells, yeast cells, and insect cells). Exogenous nucleic acids may be found (i) freely dispersed by themselves within a cell, (ii) incorporated into a recombinant vector, or (iii) incorporated into the host cell genome or mitochondrial DNA. Mammalian cells, such as those derived from humans, mice, hamsters, pigs, goats, and primates, are particularly preferred. The cells may originate from many tissue types and include primary cells and cell lines. Specific examples include keratinocytes, peripheral blood leukocytes, bone marrow stem cells, and embryonic stem cells. In further embodiments, the cells are antigen-presenting cells, particularly dendritic cells, monocytes, or macrophages.

[0210] Cells containing nucleic acid molecules preferably express peptides or polypeptides encoded by the nucleic acid.

[0211] The term "clonal proliferation" refers to the process by which a particular entity increases. In connection with the present invention, this term is preferably used in relation to an immune response in which lymphocytes are stimulated by an antigen, proliferate, and amplify specific lymphocytes that recognize the antigen. Preferably, clonal proliferation results in the differentiation of lymphocytes.

[0212] Terms such as “reduce” or “inhibit” relate to the ability to produce an overall reduction of a level of preferably 5% or more, 10% or more, 20% or more, more preferably 50% or more, and most preferably 75% or more. “Inhibit” or similar phrases include complete or virtually complete inhibition, i.e., reduction to zero or virtually zero.

[0213] Terms such as “increase,” “enhance,” “promote,” or “extend” preferably relate to an increase, enhancement, promotion, or extension of approximately 10%, preferably at least 20%, preferably at least 30%, preferably at least 40%, preferably at least 50%, preferably at least 80%, preferably at least 100%, preferably at least 200%, and especially at least 300%. These terms may also relate to an increase, enhancement, promotion, or extension from zero or an unmeasurable or undetectable level to a level above zero or a measurable or detectable level.

[0214] The active substances, compositions, and methods described herein can be used to treat subjects having diseases, such as diseases characterized by the presence of abnormal cells that express antigens and present antigenic peptides. Cancer is a particularly preferred disease. The active substances, compositions, and methods described herein can also be used for immunization or vaccination to prevent the diseases described herein.

[0215] According to the present invention, the term "disease" refers to any pathological condition, including cancerous diseases, and in particular the forms of cancerous diseases described herein.

[0216] The term "normal" refers to a healthy state or condition in a healthy subject or tissue, i.e., a non-pathological state, where "healthy" preferably means non-cancerous.

[0217] According to the present invention, a “disease involving cells expressing an antigen” means that the expression of an antigen is detected in cells of an abnormal tissue or organ. The expression in cells of an abnormal tissue or organ may be increased compared to the state in healthy tissue or organ. The increase refers to an increase of at least 10%, particularly at least 20%, at least 50%, at least 100%, at least 200%, at least 500%, at least 1000%, at least 10000%, or even greater. In one embodiment, the expression is observed only in the affected tissue, and the expression in healthy tissue is suppressed. According to the present invention, diseases involving or related to cells expressing an antigen include cancerous diseases.

[0218] Cancer (medical term: malignant neoplasm) is a class of diseases characterized by uncontrolled growth (division beyond normal limits), invasion (invasion and destruction of adjacent tissues), and sometimes metastasis (spread to other parts of the body via the lymphatic system or bloodstream). These three malignant characteristics of cancer distinguish it from benign tumors, which are self-limiting and do not invade or metastasize. Most cancers form tumors, but some, such as leukemia, do not.

[0219] Malignant tumors are essentially synonymous with cancer. Malignant diseases, malignant neoplasms, and malignant tumors are essentially synonymous with cancer.

[0220] According to the present invention, the terms “tumor” or “tumor disease” preferably refer to the abnormal proliferation of cells (called neoplastic cells, tumor-forming cells, or tumor cells) that form swelling or lesions. “Tumor cells” means abnormal cells that grow by rapid, uncontrolled cell proliferation and continue to grow even after the stimulus that initiated new growth has ceased. Tumors exhibit a partial or complete lack of structural mechanisms and functional coordination with normal tissue, and typically form a distinct tissue mass that can be benign, premalignant, or malignant.

[0221] A benign tumor is a tumor that lacks all three malignant characteristics of cancer. Therefore, by definition, a benign tumor does not grow in an unrestrained, aggressive manner, does not invade surrounding tissues, and does not spread (metastasize) to non-adjacent tissues.

[0222] A neoplasm is an abnormal tissue mass resulting from neoplasia. Neoplasia (from the Greek word for new growth) is the abnormal proliferation of cells. The growth of cells outpaces the growth of the surrounding normal tissue and does not coordinate with that tissue. The growth continues in the same excessive manner even after the cessation of stimulation. This usually results in a lump or tumor. Neoplasms can be benign, premalignant, or malignant.

[0223] The term "tumor growth" or "tumor development" according to the present invention relates to the tendency of a tumor to increase in size and / or the tendency of tumor cells to proliferate.

[0224] For the purposes of this invention, the terms "cancer" and "cancer disease" are used interchangeably with the terms "tumor" and "tumor disease."

[0225] Cancers are classified by the type of cells that resemble the tumor, and therefore by the tissue from which the tumor is presumed to originate. These are histology and location, respectively.

[0226] The term "cancer" according to the present invention includes leukemia, seminoma, melanoma, teratoma, lymphoma, neuroblastoma, glioma, rectal cancer, endometrial cancer, kidney cancer, adrenal cancer, thyroid cancer, hematological cancer, skin cancer, brain cancer, cervical cancer, intestinal cancer, liver cancer, colon cancer, stomach cancer, intestinal cancer, head and neck cancer, gastrointestinal cancer, lymph node cancer, esophageal cancer, colorectal cancer, pancreatic cancer, ear, nose and throat (ENT) cancer, breast cancer, prostate cancer, uterine cancer, ovarian cancer and lung cancer, and their metastases. Examples include lung carcinoma, breast carcinoma, prostate carcinoma, colon carcinoma, renal cell carcinoma, cervical carcinoma or metastases of the cancer types or tumors mentioned above. The term "cancer" according to the present invention also includes cancer metastases and cancer recurrence.

[0227] The main types of lung cancer are small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC). NSCLC has three main subtypes: squamous cell carcinoma, adenocarcinoma, and large cell lung cancer. Adenocarcinoma accounts for approximately 10% of lung cancers. This cancer is usually found in the peripheral parts of the lungs, whereas both SCLC and squamous cell carcinoma tend to be located more centrally.

[0228] Skin cancer is a malignant growth on the skin. The most common types of skin cancer are basal cell carcinoma, squamous cell carcinoma, and melanoma. Malignant melanoma is a serious type of skin cancer. Malignant melanoma results from the uncontrolled growth of pigment cells called melanocytes.

[0229] According to the present invention, a "carcinoma" is a malignant tumor derived from epithelial cells. This group includes the most common cancers, including common forms of breast cancer, prostate cancer, lung cancer, and colon cancer.

[0230] "Bronchioloalveolar carcinoma" is a carcinoma of the lung thought to be derived from the epithelium of the terminal bronchioles, where the neoplastic tissue grows along the alveolar walls and forms nodules within the alveoli. Mucin may be revealed in part of the cells and in the substances within the alveoli, which also include exfoliated cells.

[0231] "Adenocarcinoma" is cancer that develops from glandular tissue. This tissue is also part of a larger tissue category known as epithelial tissue. Epithelial tissue includes the skin, glands, and various other tissues that line the body cavities and internal organs. Epithelium is embryologically derived from the ectoderm, endoderm, and mesoderm. To be classified as adenocarcinoma, cells do not necessarily have to be part of a gland as long as they have secretory properties. This form of carcinoma can occur in some higher mammals, including humans. Well-differentiated adenocarcinomas tend to resemble the glandular tissue from which they arise, while poorly differentiated ones may not. By staining cells from a biopsy, a pathologist can determine whether a tumor is an adenocarcinoma or some other type of cancer. Adenocarcinoma can occur in many tissues of the body due to the ubiquity of glands within the body. Each gland does not necessarily secrete the same substance, but as long as there is an exocrine function to the cells, they are considered glands and thus their malignant form is named adenocarcinoma. Malignant adenocarcinomas invade other tissues and often metastasize if given enough time. Ovarian adenocarcinoma is the most common type of ovarian cancer. This includes serous and mucinous adenocarcinomas, clear cell adenocarcinomas, and endometrioid adenocarcinomas.

[0232] Renal cell carcinoma, also known as renal cell adenocarcinoma, is a type of kidney cancer that develops in the inner layer of the proximal tubule, which is a very small tube in the kidney that filters blood and removes waste products. Renal cell carcinoma is the most common type of kidney cancer, excluding cases in adults, and is the most lethal of all urogenital tumors. The different subtypes of renal cell carcinoma are clear cell renal cell carcinoma and papillary renal cell carcinoma. Clear cell renal cell carcinoma is the most common form of renal cell carcinoma. When viewed under a microscope, the cells that make up clear cell renal cell carcinoma appear very pale or transparent. Papillary renal cell carcinoma is the second most common subtype. These cancers form small finger-like projections (called papillae) in part, if not mostly, of the tumor.

[0233] Lymphoma and leukemia are malignant diseases that originate from hematopoietic (blood-forming) cells.

[0234] A blastocytic tumor, or blastoma, is a tumor (usually malignant) that resembles immature or embryonic tissue. Yes, they exist. Many of these tumors are most common in children.

[0235] "Metastasis" refers to the spread of cancer cells from their original site to another part of the body. The formation of metastasis is a very complex process, depending on the separation of malignant cells from the primary tumor, invasion of the extracellular matrix, penetration of the endothelial basement membrane to enter body cavities and blood vessels, and then, after being carried by the blood, invasion of the target organ. Finally, the growth of a new tumor at the target site, i.e., a secondary or metastatic tumor, depends on angiogenesis. Tumor metastasis often occurs even after the removal of the primary tumor, because tumor cells or tumor components may remain and exhibit metastatic potential. In one embodiment, the term "metastasis" according to the present invention refers to "distant metastasis," which refers to metastasis far from the primary tumor and the regional lymph node system.

[0236] The cells of a secondary or metastatic tumor are similar to the cells in the original tumor. This means, for example, that if ovarian cancer metastasizes to the liver, the secondary tumor will consist of abnormal ovarian cells rather than abnormal liver cells. In that case, the tumor in the liver is called metastatic ovarian cancer, not liver cancer.

[0237] In ovarian cancer, metastasis can occur in the following ways: by direct contact or expansion; by invading adjacent tissues or organs located near or around the ovary, such as the fallopian tubes, uterus, bladder, rectum, etc; by dissemination or shedding into the abdominal cavity, which is the most common way ovarian cancer spreads; by cancer cells breaking through the surface of the ovarian mass and "dropping" into other structures in the abdomen, such as the liver, stomach, colon, or diaphragm; by detaching from the ovarian mass and entering the lymphatic system, then migrating to other parts of the body or distant organs such as the lungs or liver; by detaching from the ovarian mass and entering the bloodstream, then migrating to other parts of the body or distant organs.

[0238] According to the present invention, metastatic ovarian cancer includes cancer of the fallopian tube, cancer of abdominal organs such as the intestine, uterus, bladder, rectum, liver, stomach, colon, diaphragm, lung, the inner lining of the abdomen or pelvis (peritoneum), and brain cancer. Similarly, metastatic lung cancer refers to cancer that has spread from the lungs to distant and / or several other sites in the body, including liver cancer, adrenal gland cancer, bone cancer, and brain cancer.

[0239] The term "circulating tumor cells" or "CTCs" refers to cells isolated from a primary tumor or tumor metastasis that circulate in the bloodstream. CTCs can constitute the seeds for the subsequent growth of further tumors (metastases) in different tissues. Circulating tumor cells are found in patients with metastatic disease at a frequency of approximately 1 to 10 CTCs per mL of whole blood. Diagnostic methods have been developed to isolate CTCs. Several diagnostic methods for isolating CTCs have been described in this field, including techniques that utilize the fact that epithelial cells commonly express the cell adhesion protein EpCAM, which is not present in normal blood cells. The immunomagnetic bead-based capture method involves treating a blood sample with an antibody against EpCAM bound to magnetic particles, and then separating the labeled cells with a magnetic field. The isolated cells are then stained with antibodies against cytokeratin, another epithelial marker, and CD45, a common leukocyte marker, to distinguish rare CTCs from contaminated leukocytes. This robust, semi-automated approach identifies CTCs with an average yield of approximately 1 CTC / mL and a purity of 0.1% (Allard et al., 2004: Clin Cancer Res 10, 6897-6904). A second method for isolating CTCs uses a microfluidics-based CTC capture device, which involves flowing whole blood through a chamber embedded with 80,000 microposts functionalized by coating with an antibody against EpCAM. The CTCs are then stained with cytokeratin or a tissue-specific marker, e.g., a secondary antibody against PSA in prostate cancer or HER2 in breast cancer, and visualized by automated scanning of the microposts in multiple planes along a three-dimensional coordinate system. The CTC chip can identify cytokeratin-positive circulating tumor cells in the patient with an average yield of 50 cells / mL and a purity ranging from 1 to 80% (Nagrath et al., 2007: Nature 450, 1235-1239). Another possibility for isolating CTCs is to use the CellSearch® Circulating Tumor Cell (CTC) Test from Veridex, LLC (Raritan, NJ), which captures, identifies, and counts CTCs in a blood tube.The CellSearch® system is a U.S. Food and Drug Administration (FDA) approved method for counting circulating tumor cells (CTCs) in whole blood, based on a combination of immunomagnetic labeling and automated digital microscopy. Other methods for isolating CTCs described in the literature exist, all of which can be used in conjunction with the present invention.

[0240] Recurrence or relapse occurs when a person becomes ill again with a condition they have had in the past. For example, if a patient has had a tumor and has been successfully treated for the disease, but then develops the disease again, the newly developed disease may be considered a recurrence or relapse. However, according to the present invention, recurrence or relapse of a tumor may occur at the site of the original tumor, but not necessarily. For example, if a patient has had an ovarian tumor and has been successfully treated for it, recurrence or relapse may be the development of an ovarian tumor or the development of a tumor at a site different from the ovary. Tumor recurrence or relapse includes situations in which the tumor occurs at a site different from the original tumor site, as well as situations in which it occurs at the original tumor site. Preferably, the original tumor that the patient has been treated for is a primary tumor, and the tumor at a site different from the original tumor site is a secondary or metastatic tumor.

[0241] "To treat" means administering the compounds or compositions described herein to a subject in order to prevent or eliminate a disease, including by reducing the size or number of tumors in the subject; to stop or slow the progression of a disease in a subject; to inhibit or delay the onset of a new disease in a subject; to reduce the frequency or severity of symptoms and / or recurrences in a subject who currently has or has previously had a disease; and / or to prolong, i.e., increase, the survival time of the subject. In particular, the term "treatment of disease" includes curing, shortening the duration, improving, preventing, slowing or inhibiting the progression or worsening of a disease or its symptoms, or preventing or delaying its onset.

[0242] "At risk" refers to subjects, or patients, who are identified as having a higher-than-usual likelihood of developing a disease, particularly cancer, compared to the general population. In addition, subjects who have had or currently have a disease, particularly cancer, are at high risk of developing the disease because they continue to be at risk of developing it. Subjects who currently have or have had cancer are also at high risk of cancer metastasis.

[0243] The term "immunotherapy" relates to treatments that involve the activation of a specific immune response. In relation to the present invention, terms such as "protective," "preventive," "preventive," "protective," or "protective" relate to the prevention or treatment of the onset and / or transmission of a disease in a subject, or both, in particular to minimizing the likelihood that the subject will develop a disease or delaying the onset of a disease. For example, as mentioned above, a person at risk of tumors is a candidate for a treatment to prevent tumors.

[0244] Prophylactic administration of immunotherapy, such as prophylactic administration of the compositions described herein, preferably protects the recipient from the onset of the disease. Therapeutic administration of immunotherapy, such as therapeutic administration of the compositions described herein, may result in the inhibition of disease progression / growth. This preferably includes slowing of disease progression / growth, in particular cessation of disease progression, resulting in the elimination of the disease.

[0245] Immunotherapy may be carried out using any of the various techniques in which the active ingredients provided herein function to remove abnormal cells from a patient. Such removal may occur as a result of enhancing or inducing an immune response in the patient that is specific to the antigen or cells expressing the antigen.

[0246] In certain embodiments, immunotherapy may be active immunotherapy, in which case the treatment is based on in vivo stimulation of the endogenous host immune system that responds to abnormal cells by administration of immune response modulochemicals (such as polypeptides and nucleic acids provided herein).

[0247] The agents and compositions provided herein can be used alone or in combination with conventional treatment regimens, such as surgery, radiation, chemotherapy, and / or bone marrow transplantation (autologous, syngeneic, allogeneic, or unrelated).

[0248] The terms "immunization" or "vaccination" refer to the process of treating a subject for therapeutic or prophylactic reasons to induce an immune response.

[0249] The term "in vivo" relates to the situation in a subject.

[0250] The terms "subject", "individual", "organism", or "patient" are used interchangeably and relate to vertebrates, preferably mammals. For example, in the context of the present invention, mammals include humans, non-human primates, domestic animals such as dogs, cats, sheep, cows, goats, pigs, horses, etc., laboratory animals such as mice, rats, rabbits, guinea pigs, etc., and captive animals such as zoo animals. The term "animal" as used herein also includes humans. The term "subject" can also include a patient, i.e., an animal, preferably a human having a disease, preferably a human having a disease described herein.

[0251] The term "autologous" is used to denote something that is derived from the same subject. For example, "autologous transplantation" refers to the transplantation of tissue or an organ derived from the same subject. Such procedures are advantageous as they overcome the immunological barriers that would otherwise result in rejection.

[0252] The term "heterologous" is used to denote something that consists of multiple different elements. As an example, the transfer of bone marrow from one individual to a different individual constitutes a heterologous transplantation. A heterologous gene is a gene that is derived from a source other than the subject.

[0253] Preferably, one or more active ingredients as described herein are administered as part of a composition for immunization or vaccination, together with one or more adjuvants to induce or enhance an immune response. The term “adjuvant” refers to a compound that prolongs, enhances, or promotes an immune response. The compositions of the present invention preferably exert their effects without the addition of adjuvants. Nevertheless, the compositions of this application may still contain any known adjuvants. Adjuvants include a heterogeneous group of compounds such as oily emulsions (e.g., Freund's adjuvants), inorganic compounds (e.g., alum), bacterial products (e.g., Bordetella pertussis toxin), liposomes, and immunostimulant complexes. Examples of adjuvants include monophosphoryl lipid A (MPL SmithKline Beecham), saponins such as QS21 (SmithKline Beecham), DQS21 (SmithKline Beecham; International Publication No. 96 / 33739), QS7, QS17, QS18 and QS-L1 (So et al., 1997, Mol. Cells 7:178-186), incomplete Freund's adjuvant, complete Freund's adjuvant, vitamin E, montanide, alum, CpG oligonucleotides (Krieg et al., 1995, Nature 374:546-549), and various water-in-oil emulsions prepared from biodegradable oils such as squalene and / or tocopherol.

[0254] Other substances that stimulate the patient's immune response may also be administered. For example, cytokines can be used in vaccination due to their regulatory properties on lymphocytes. Such cytokines include interleukin-12 (IL-12), GM-CSF, and IL-18, which have been shown to enhance the protective effect of vaccines (see Science 268:1432-1434, 1995).

[0255] There are many compounds that enhance the immune response and can therefore be used in vaccination. These compounds include costimulatory molecules provided in the form of proteins or nucleic acids, such as B7-1 and B7-2 (CD80 and CD86, respectively).

[0256] According to the present invention, a “tumor specimen” is a specimen containing tumor cells or cancer cells, such as circulating tumor cells (CTCs), particularly tissue specimens containing bodily fluids, and / or cell specimens. According to the present invention, a “non-tumor-forming specimen” is a specimen that does not contain tumor cells or cancer cells, such as circulating tumor cells (CTCs), particularly tissue specimens containing bodily fluids, and / or cell specimens. Such specimens can be obtained by conventional methods, for example, by tissue biopsy including punch biopsy, and by collecting blood, bronchial aspirate, sputum, urine, feces, or other bodily fluids. According to the present invention, the term “specimen” also includes processed specimens, such as fractions or isolates of biological specimens, for example, nucleic acids or cell isolates.

[0257] The therapeutic agents, vaccines, and compositions described herein may be administered by any conventional route, including injection or infusion. Administration may be carried out, for example, orally, intravenously, intraperitoneally, intramuscularly, subcutaneously, or percutaneously. In one embodiment, administration is carried out intranodally, such as by injection into a lymph node. Other forms of administration envision in vitro transfection of antigen-presenting cells, such as dendritic cells, with nucleic acids described herein, followed by administration of the antigen-presenting cells.

[0258] The active substances described herein are administered in effective doses. “Effective dose” means the amount, alone or in combination with further doses, that achieves the desired response or effect. In the case of treating a particular disease or condition, the desired response preferably relates to inhibiting the progression of the disease. This includes slowing the progression of the disease, in particular preventing or reversing its progression. The desired response in the treatment of a disease or condition may also be a delay in the onset of the disease or condition or a prevention of its onset.

[0259] The effective dose of the active ingredients described herein depends on the individual patient's parameters, including the condition being treated, the severity of the disease, age, physiological state, size, and weight, the duration of treatment, the type of concomitant therapy (if any), the specific route of administration, and similar factors. Therefore, the dose administered of the active ingredients described herein may depend on these various parameters. If the response in the patient is insufficient with the initial dose, a higher dose (or an effectively higher dose achieved by a different, more localized route of administration) may be used.

[0260] The pharmaceutical compositions described herein are preferably sterile and contain an effective amount of therapeutic active substance to produce the desired reaction or effect.

[0261] The pharmaceutical compositions described herein are generally administered in pharmaceutically acceptable amounts and in pharmaceutically acceptable preparations. The term "pharmaceutically acceptable" refers to non-toxic substances that do not interact with the action of the active ingredients of the pharmaceutical composition. Such preparations may typically contain salts, buffers, preservatives, carriers, adjuvants, and other supplemental immunostimulants such as CpG oligonucleotides, cytokines, chemokines, saponins, GM-CSF and / or RNA, and, where appropriate, other therapeutically active compounds. When used in a drug, salts should be pharmaceutically acceptable. However, non-pharmaceutically acceptable salts may be used to prepare pharmaceutically acceptable salts and are included in the present invention. Such pharmacokinetic and pharmaceutically acceptable salts include, but are not limited to, those prepared from the following acids: hydrochloric acid, hydrobromic acid, sulfuric acid, nitric acid, phosphoric acid, maleic acid, acetic acid, salicylic acid, citric acid, formic acid, malonic acid, succinic acid, etc. Pharmacologically suitable salts can also be prepared as alkali metal salts or alkaline earth metal salts, such as sodium salts, potassium salts, or calcium salts.

[0262] The pharmaceutical compositions described herein may contain pharmaceutically suitable carriers. The term “carrier” refers to a natural or synthetic organic or inorganic component to which an active ingredient is combined to facilitate application. According to the present invention, the term “pharmaceutically suitable carrier” includes one or more suitable solid or liquid fillers, diluents, or encapsulating materials suitable for administration to a patient. The components of the pharmaceutical compositions described herein are generally not subject to interactions that substantially impair the desired pharmaceutically effective effect.

[0263] The pharmaceutical compositions described herein may contain suitable buffering agents, such as acetic acid in salt form, citric acid in salt form, boric acid in salt form, and phosphoric acid in salt form.

[0264] The pharmaceutical composition may also contain, where appropriate, suitable preservatives, such as benzalkonium chloride, chlorobutanol, parabens, and thimerosal.

[0265] Pharmaceutical compositions are typically provided in unit dose forms and can be manufactured by methods known to the public. Pharmaceutical compositions described herein may be, for example, in the form of capsules, tablets, lozenges, solutions, suspensions, syrups, elixirs, or emulsions.

[0266] Compositions suitable for parenteral administration typically contain sterile aqueous or non-aqueous preparations of the active compound, preferably isotonic with the recipient's blood. Examples of suitable carriers and solvents include Ringer's solution and isotonic sodium chloride solution. In addition, sterile fixative oil is usually used as the solution or suspension medium.

[0267] The present invention will be described in detail with reference to the following drawings and examples, which are for illustrative purposes only and are not intended to limit the invention. Further embodiments, similarly included in the present invention, will be accessible to those skilled in the art. Examples of embodiments are provided below. 1. A method for preventing or treating cancer in a patient, (i) a step of inducing a first immune response to one or more tumor antigens in the patient, and (ii) A step of inducing a secondary immune response in the patient to one or more tumor antigens, wherein the secondary immune response is specific to cancer-specific somatic mutations present in the patient's cancer cells. A method that includes this. 2. The method according to 1., wherein the first and / or second immune response is a cellular response. 3. The method according to 1. or 2., wherein the first immune response comprises a CD8+ T cell response. 4. The method according to any one of 1 to 3, wherein the second immune response includes a CD4+ T cell response. 5. The method according to any one of 1 to 4, wherein the first immune response is not specific to cancer-specific somatic mutations present in the patient's cancer cells. 6. The method according to any one of 1 to 5, wherein the first immune response is induced by administering one or more vaccine products selected from a set of pre-manufactured vaccine products, each of which pre-manufactured vaccine products induces an immune response to a tumor antigen, and the set preferably comprises vaccine products that induce immune responses to different tumor antigens. 7. The method according to 6, wherein the administered vaccine product induces an immune response against tumor antigens common in the treated cancer. 8. The method according to 6. or 7., wherein the set comprises a vaccine product that induces an immune response to tumor antigens common in various cancers. 9. The method according to any one of 1 to 8, wherein the patient is positive for one or more tumor antigens. 10. The method according to any one of 1 to 9, wherein the cancer-specific somatic mutation is an exome present in the cancer cells of the patient and / or a non-synonymous mutation. 11. The method according to any one of claims 1 to 10, wherein the second immune response is induced by administering a vaccine containing a polypeptide comprising a mutation-based neoepitope or a nucleic acid encoding the polypeptide, wherein the polypeptide preferably comprises up to 30 mutation-based neoepotopes. 12. The method according to 11, wherein the polypeptide further comprises an epitope that does not contain cancer-specific somatic mutations expressed by cancer cells. 13. The method according to 11. or 12., wherein the epitopes are present in their natural sequence context so as to form a vaccine sequence, and the vaccine sequence is preferably about 30 amino acids long. 14. The method according to any one of 11 to 13, wherein the neoepitope, epitope and / or vaccine sequence are aligned in the head-to-tail direction and / or separated by a linker. 15. The method according to any one of 1 to 14, wherein the first and / or second immune response is induced by administration of an RNA vaccine. 16. The method according to any one of 1 to 15, wherein the tumor antigen is a tumor-associated antigen. [Examples]

[0268] The techniques and methods used herein are described herein or are known in themselves, as well as, for example, Sambrook et al., Molecular Cloning: A Laboratory Manual, 2 nd The procedure shall be carried out as described in the Edition (1989) Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. All methods, including the use of kits and reagents, shall be carried out according to the manufacturer's instructions unless otherwise specified.

[0269] (Example 1) Mutation detection and prioritization We first reveal sequence profiling of tumor and normal samples for identifying somatic mutations in an unbiased manner. We not only reveal this for bulk tumor samples, but also demonstrate, for the first time, the ability to identify mutations from individual circulating tumor cells. Next, we prioritize mutations for inclusion in polyneoepitope vaccines based on their predicted immunogenicity and demonstrate that the identified mutations are indeed immunogenic.

[0270] Mutation detection Rationale for using CTCs: Detection of circulating tumor cells (CTCs) in the peripheral blood of cancer patients is a widely recognized independent prognostic marker of the clinical course of a tumor (Pantel et al, Trends Mol Med 2010;16(9):398-406). Over the years, the clinical significance of CTCs has been the subject of enthusiastic academic and clinical research in oncology. Detection of CTCs in the blood of patients with metastatic breast cancer, prostate cancer, and colorectal cancer has been shown to be prognostic and to provide additional information to conventional imaging techniques and other prognostic tumor biomarkers. Sequential blood samples taken from patients before, during, and after therapeutic intervention (systemic or targeted) provide information on the response / failure of the intervention. Molecular analysis of drug-resistant CTCs may provide further insight into the resistance mechanisms in individual patients (e.g., mutations in specific signaling pathways or loss of target expression). Further potential from profiling and genetic characterization of CTCs is the identification of novel cancer targets for the development of new targeted therapies. This new diagnostic strategy is called "liquid tumor biopsy." This profiling method is rapid, repeatable, requires only the patient's blood, and does not involve surgery, thus providing a "real-time" view of the tumor's condition.

[0271] Mutations from Tumor Cells: We demonstrate our ability to identify mutations using B16 melanoma cells, exome capture for protein-coding region extraction, next-generation sequencing with our HiSeq 2000, and then bioinformatics analysis using our "iCAM" software pipeline (Figure 1). We identified 2448 nonsynonymous mutations and selected 50 for confirmation. We were able to confirm all 50 somatic mutations.

[0272] The following is an example of the protein effects of somatic mutations found in B16 melanoma cells: [ka]

[0273] Mutations from individual circulating tumor cells (CTCs): Next, we were able to identify tumor-specific somatic mutations from NGS profiling of RNA from a single CTC. Labeled B16 melanoma cells were intravenously injected into the tail of mice, the mice were sacrificially killed, blood was collected from the heart, cells were sorted to recover labeled circulating B16 cells (CTCs), RNA was extracted, cDNA synthesis and nonspecific amplification were performed based on SMART, followed by NGS RNA-Seq assays and subsequence data analysis (see below).

[0274] We profiled eight individual CTCs and identified somatic mutations. Furthermore, we identified previously identified somatic mutations in all eight cells. In multiple cases, the data showed heterogeneity at the individual cell level. For example, at position 144078227 (assembly mm9) on chromosome 2, in the gene Snx15, two cells showed the reference nucleotide (C) and two cells showed the mutant nucleotide (T).

[0275] This demonstrates that we can profile individual CTCs to identify somatic mutations, which is a fundamental pathway to "real-time" iVAC (individualized vaccines), where patients are profiled repeatedly, and the results reflect the patient's current state rather than their state at an earlier point in time. Furthermore, this reveals that we can identify heterogeneous somatic mutations present in subsets of tumor cells, enabling the assessment of mutation frequencies for purposes such as identifying major and rare mutations.

[0276] (method) Samples: For profiling experiments, samples included 5-10 mm tail specimens from C57BL / 6 mice ("Black6") and highly aggressive B16F10 mouse melanoma cells ("B16") originally derived from Black6 mice.

[0277] Circulating tumor cells (CTCs) were generated using fluorescently labeled B16 melanoma cells. B16 cells were resuspended in PBS, and an equal volume of freshly prepared CFSE solution (5 μM in PBS) was added to the cells. The sample was gently mixed by vortexing and then incubated at room temperature for 10 minutes. To stop the labeling reaction, an equal volume of PBS containing 20% ​​FSC was added to the sample and gently mixed by vortexing. After incubation at room temperature for 20 minutes, the cells were washed twice with PBS. Finally, the cells were resuspended in PBS and injected intravenously (iv) into mice. After 3 minutes, the mice were euthanized and blood was collected.

[0278] Red blood cells from blood samples were dissolved by adding 1.5 ml of freshly prepared PharmLyse Solution (Beckton Dickinson) per 100 μl of blood. After one washing step, 7-AAD was added to the sample and incubated at room temperature for 5 minutes. Following incubation, two more washings were performed, and the sample was resuspended in 500 μl of PBS.

[0279] CFSE-labeled circulating B16 cells were sorted using an Aria I cell sorter (BD). Single cells were sorted on a 96-well v-bottom plate prepared with 50 μl / well of RLT buffer (Quiagen). After sorting was complete, the plate was stored at -80°C until nucleic acid extraction and sample preparation were initiated.

[0280] Nucleic acid extraction and sample preparation: Nucleic acids (DNA and RNA) from B16 cells and Black6 tail tissue (DNA) were extracted using the Qiagen DNeasy Blood and Tissue kit (DNA) and the Qiagen RNeasy Micro kit (RNA).

[0281] RNA was extracted from each selected CTC, and cDNA synthesis and nonspecific amplification were performed based on SMART. RNA from the selected CTC cells was extracted using the RNeasy Micro Kit (Qiagen, Hilden, Germany) according to the supplier's instructions. A modified BD SMART protocol was used for cDNA synthesis: Mint Reverse Transcriptase (Evrogen, Moscow, Russia) was combined with TS-short (Eurogentec SA, Seraing, Belgium), which introduced a long oligo(dT)-T primer for priming the first-chain synthesis reaction and an oligo(riboG) sequence to enable the generation of an extended template and template switching via the terminal transferase activity of the reverse transcriptase [Chenchik, A., Y. et al. 1998. Generation and use of high quality cDNA from small amounts of total RNA by SMART PCR. In Gene Cloning and Analysis by RT-PCR. PLJSiebert, ed. BioTechniques Books, MA, Natick. 305-319]. First-strand cDNA synthesized according to the manufacturer's instructions was amplified for 35 cycles using PfuUltra Hotstart High-Fidelity DNA Polymerase 5U (Stratagene, La Jolla, CA) and 0.48 μM TS-PCR primers in the presence of 200 μM dNTPs (cycling conditions: 2 min at 95°C, 30 sec at 94°C, 30 sec at 65°C, 1 min at 72°C, and 6 min for final extension at 72°C). The success of CTC gene amplification was controlled by specific primers that observe actin and GAPDH.

[0282] Next-generation sequencing, DNA sequencing: In this case, exome capture for DNA resequencing was performed using a capture assay based on Agilent Sure-Select solution designed to capture all mouse protein-coding regions [Gnirke A et al: Solution hybrid selection with ultra-long oligonucleotides for massively parallel targeted sequencing. Nat Biotechnol 2009, 27:182-189].

[0283] Briefly, 3 μg of purified genomic DNA was fragmented into 150-200 bp fragments using a Covaris S2 sonicator. The gDNA fragments were repaired at the ends using T4 DNA polymerase and Krenow DNA polymerase, and 5' phosphorylated using T4 polynucleotide kinase. The blunt-end gDNA fragments were 3' adenylated (3'-5' exo-minus) using Krenow fragments. A single 3' T overhang Illumina paired-end adapter was ligated to the gDNA fragments in a 10:1 molar ratio of adapter to genomic DNA insert using T4 DNA ligase. The adapter-ligated gDNA fragments were enriched before capture, and flow cell-specific sequences were added using Illumina PE PCR primers 1.0 and 2.0 and Herculase II polymerase (Agilent) in four PCR cycles.

[0284] 500 ng of adapter-linked, PCR-enriched gDNA fragments were hybridized to Agilent's SureSelect biotinylated mouse whole exome RNA library bait at 65°C for 24 hours. The hybridized gDNA / RNA bait complex was extracted using streptavidin-coated magnetic beads. The gDNA / RNA bait complex was washed, and the RNA bait was cleaved during elution in SureSelect elution buffer to retain the captured adapter-linked, PCR-enriched gDNA fragments. The captured gDNA fragments were PCR-amplified using Herculase II DNA polymerase (Agilent) and SureSelect GA PCR primers for 10 cycles.

[0285] All purification was performed using 1.8x volume AMPure XP magnetic beads (Agencourt). All quality control was performed using Invitrogen's Qubit HS assay, and fragment sizes were determined using Agilent's 2100 Bioanalyzer HS DNA assay.

[0286] Exome-enriched gDNA libraries were clustered using cBot with the Truseq SR cluster kit v2.5 at 7 pM, and 50 bp of each cluster were sequenced using the Truseq SBS kit-HS 50 bp on an Illumina HiSeq2000.

[0287] Next-generation sequencing, RNA sequencing (RNA-Seq): Barcoded mRNA-seq cDNA libraries were prepared from 5 μg of total RNA using a modified Illumina mRNA-seq protocol. mRNA was isolated using Seramag Oligo (dT) magnetic beads (Thermo Scientific). The isolated mRNA was fragmented using divalent cations and heat to produce fragments ranging from 160 to 220 bp. The fragmented mRNA was converted to cDNA using random primers and SuperScript II (Invitrogen), and the second strand was then synthesized using DNA polymerase I and RNase H. The cDNA was repaired at the ends using T4 DNA polymerase and Krenow DNA polymerase, and 5' phosphorylated using T4 polynucleotide kinase. The blunt-end cDNA fragments were 3' adenylated (3'-5' exo-minus) using Krenow fragments. A single 3'T overhang Illumina multiplex-specific adapter was ligated using T4 DNA ligase in a 10:1 molar ratio of adapter to cDNA insert.

[0288] cDNA libraries were purified and size-selected to 200-220 bp using E-Gel 2% SizeSelect gel (Invitrogen). Concentration, Illumina 6-nucleotide index sequences, and flow cell-specific sequences were added by PCR using Phusion DNA polymerase (Finnzymes). All cleansing was performed using 1.8x volume AgincourtAMPure XP magnetic beads. All quality control was performed using Invitrogen's Qubit HS assay, and fragment sizes were determined using Agilent's 2100 Bioanalyzer HS DNA assay.

[0289] Barcoded RNA-Seq libraries were clustered using cBot with the Truseq SR cluster kit v2.5 at 7 pM, and 50 bp of each were sequenced using the Truseq SBS kit-HS 50 bp on an Illumina HiSeq2000.

[0290] For RNA-Seq profiling of CTCs, a modified version of this protocol was used, in which 500-700 ng of SMART-amplified cDNA was ligated using paired-end adapters and PCR enrichment was performed using Illumina PE PCR primers 1.0 and 2.0.

[0291] NGS data analysis, gene expression: To determine expression levels, sequence reads from RNA samples output from Illumina HiSeq 2000 were preprocessed according to the Illumina standard protocol. This included filtering and demultiplexing of low-quality reads. For RNA-Seq transcriptome analysis, bowtie (version 0.12.5) [Langmead B. et al. Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome Biol 10:R25] was used, with the "-v2-best" parameter for genome alignment and default parameters for transcript alignment, to align sequence reads to the reference genome sequence [Mouse Genome Sequencing Consortium. Initial sequencing and comparative analysis of the mouse genome. Nature, 420, 520-562 (2002)]. Alignment coordinates were compared with the exon coordinates of RefSeq transcripts [Pruitt KD. et al. NCBI Reference Sequence (RefSeq): a curated non-redundant sequence database of genomes, transcripts and proteins. Nucleic Acids Res. 2005 Jan 1;33 (Database issue): D501-4], and the overlap alignment count was recorded for each transcript. Sequence reads that could not be aligned to the genome sequence were aligned to a database of all possible exon-exon junction sequences of RefSeq transcripts.The count of reads aligned at the splice junction was summed with the count of each transcript obtained in the previous step, and normalized for each transcript to RPKM (number of reads which map per kilobase of exon model per million mapped reads) [Mortazavi, A. et al. (2008). Mapping and quantifying mammalian transcriptomes by rna-seq. Nat Methods, 5(7):621-628]). Both gene expression values ​​and exon expression values ​​were calculated based on the normalized number of reads overlapping each gene or exon, respectively.

[0292] Mutation detection, bulk tumors: 50-nucleotide single-end reads from Illumina HiSeq 2000 were aligned to a reference mouse genome assembly mm9 using bwa (version 0.5.8c) [Li H. and Durbin R. (2009) Fast and accurate short read alignment with Burrows-Wheeler Transform. Bioinformatics, 25:1754-60] with default options. Ambiguous reads—reads mapping to multiple locations in the genome were removed, the remaining alignments were screened, indexed, converted to Binary Compression Format (BAM), and read quality scores were converted from Illumina standard phred+64 to standard Sanger quality score using a shell script.

[0293] For each sequencing lane, mutations were identified using three software programs: samtools (version 0.1.8) [Li H. Improving SNP discovery by base alignment quality. Bioinformatics. 2011 Apr 15;27(8):1157-8. Epub 2011 Feb 13], GATK (version 1.0.4418) [McKenna A. et al. The Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data. Genome Res. 2010 Sep;20(9):1297-303. Epub 2010 Jul 19], and SomaticSniper (http: / / genome.wustl.edu / software / somaticsniper). For samtools, the author's recommended options and filtering criteria, including a first filtering and a maximum coverage of 200, were used. For the second filtering in samtools, the minimum indel quality score was 50, and the minimum point mutation quality was 30. For GATK mutation calling, we followed the best practice guidelines designed by the creators as presented in the GATK user manual (http: / / www.broadinstitute.org / gsa / wiki / index.php / The_Genome_Analysis_Toolkit). We omitted the variant score recalibration step and replaced it with the hard filtering option. For SomaticSniper mutation calling, we used the default option and further examined only predicted mutations with a “somatic cell score” of 30 or higher.

[0294] For mutation detection, following the CTC:bulk tumor iCAM process, 50-nucleotide single-ended reads from Illumina HiSeq 2000 were aligned to a reference mouse genome assembly mm9 using bwa (version 0.5.8c[5]) with default options. Since the CTC NGS reads were derived from RNA-Seq assays, the reads were also aligned to the transcriptome sequence, including exon-exon junctions, using bowtie (above). All alignments were used to compare the nucleotide sequences from the reads to both the reference genome and B16 mutations derived from the bulk tumor. Identified mutations were manually evaluated using a Perl script, as well as the samtools software program and IGV (Integrated Genome Viewer) to image the results.

[0295] The output of "Mutation Discovery" is the identification of somatic mutations in tumor cells, from NGS data from the sample to a list of mutations. In the B16 sample, we identified 2448 somatic mutations using exome resequencing.

[0296] Prioritizing mutations Next, we explore the potential of a mutation prioritization pipeline for inclusion in vaccines. Called the “Individual Cancer Mutation Detection Pipeline” (iCAM), this method identifies and prioritizes somatic mutations through a series of steps incorporating multiple state-of-the-art algorithms and bioinformatics methods. The output of this process is a list of somatic mutations prioritized based on their high immunogenicity.

[0297] Identification of somatic mutations: Mutations were identified using three different algorithms for both B16 and Black6 samples (Mutation Discovery, above). The first iCAM step was to combine the output lists from each algorithm to create a high-confidence list of somatic mutations. GATK and samtools report mutations in one sample compared to a reference genome. Mutations identified in all replicates were selected to select high-confidence mutations with few false positives for a given sample (i.e., tumor or normal). Next, mutations present in tumor samples but not in normal samples were selected. SomaticSniper automatically reports potential somatic mutations from pairs of tumor and normal data. We further filtered the results through intersection of results obtained from replicates. To remove as many false-positive cells as possible, we intersected the lists of mutations derived from the use of all three algorithms and all replicates. The final step for each somatic mutation was to assign a confidence value (p-value) to each mutation based on coverage depth, SNP quality, consensus quality, and mapping quality.

[0298] Impact of Mutations: The impact of filtered consensus somatic mutations is determined by scripting within the iCAM mutation pipeline. First, sequence reads aligned at multiple locations are excluded, thus excluding mutations occurring in non-specific genomic regions within the genome, such as those occurring for certain protein paralogs and pseudogenes. Second, it is determined whether the mutation occurs in the transcript. Third, it is determined whether the mutation occurs within the protein-coding region. Fourth, the transcript sequence is translated with and without the mutation to determine whether amino acid sequence changes are present.

[0299] Mutation Expression: The iCaM pipeline selects somatic mutations found in genes and exons expressed in tumor cells. Expression levels are determined via NGS RNA-Seq of tumor cells (see above). The number of overlapping reads in a gene and exon indicates the expression level. These counts are normalized to RPKM (Reads Per Kilobase of exon model per Million mapped reads) [Mortazavi A. et al. Mapping and quantifying mammalian transcriptomes by RNA-Seq. Nat Methods. 2008 Jul;5(7):621-8. Epub 2008 May 30], and those expressed above 10 RPKM are selected.

[0300] MHC Binding: To determine the likelihood of epitopes containing mutant peptides binding to MHC molecules, the iCAM pipeline runs a modified version of the MHC prediction software from the Immune Epitope Database (http: / / www.iedb.org / ). The local installation includes modifications to optimize data flow through the algorithm. For B16 and Black6 data, predictions were performed using all available Black6 MHC class I alleles and all epitopes for each peptide length. Mutations contained in epitopes ranked at the 95th percentile of the prediction score distribution of the IEDB training data (http: / / mhcbindingpredictions.immuneepitope.org / dataset.html) were selected, and all MHC alleles and all potential epitopes overlapping the mutations were considered.

[0301] Mutation Selection Criteria: Somatic mutations are selected based on the following criteria: a) having unique sequence content, b) being identified by all three programs, c) high mutation confidence, d) non-synonymous protein alteration, e) high transcript expression, and f) good MHC class I binding prediction.

[0302] The output of this process is a list of somatic mutations prioritized based on their likelihood of immunogenicity. There are 2448 somatic mutations in B16 melanoma cells. Of these, 1247 are found in gene transcripts. Of these, 734 cause non-synonymous protein changes. Of these, 149 are present in genes expressed in tumor cells. Of these, 102 of these expressed non-synonymous mutations are predicted to be presented on MHC molecules. These 102 highly likely immunogenic mutations are then proceeded to mutation confirmation (see below).

[0303] Confirmation of mutations Somatic mutations from DNA exome resequencing were identified by two methods: resequencing of the mutated region and RNA-Seq analysis.

[0304] To confirm mutations by resequencing, genomic regions containing mutations were amplified from 50 ng of both tumor DNA and normal control DNA using standard PCR. The size of the amplified products ranged from 150 to 400 nucleotides. Reaction specificity was controlled by loading the PCR products onto a Qiaxel instrument (Qiagen). PCR products were purified using the minElute PCR purification kit (Qiagen). Specific PCR products were sequenced using standard Sanger sequencing (Eurofins) followed by electropherogram analysis.

[0305] Mutation confirmation was also achieved through examination of tumor RNA. Tumor gene and exon expression levels were generated from RNA-Seq (RNA NGS), which maps to transcripts and constructs countable nucleotide sequences. We examined the sequence data itself to identify mutations in tumor samples [Berger MF. et al. Integrative analysis of the melanoma transcriptome. Genome Res. 2010 Apr;20(4):413-27. Epub 2010 Feb 23], providing independent confirmation of identified somatic mutations derived from DNA. Table 1: List of genes containing 50 demonstrated mutations 50 identified and confirmed somatic mutations in genes, as well as annotations for gene symbols, gene names, and predicted localization and function. [Table 1]

[0306] (Example 2) The IVAC selection algorithm enables the detection of immunogenic mutations. To investigate whether a specific T cell response can be induced for identified mutations in B16F10 melanoma cells, naive C57BL / 6 mice (n=5 / peptide) were immunized subcutaneously twice (days 0 and 7) with 100 μg of peptide containing either a mutant or wild-type amino acid sequence (see Table 2) (+ 50 μg of PolyI:C as an adjuvant). All peptides were 27 amino acids long and had a mutant / wild-type amino acid at the center. Mice were sacrificially killed on day 12, and splenocytes were collected. 5 × 10⁶ cells were used as the readout method. 5 Splenocytes / well were used as effectors, and peptide (2 μg / ml) was loaded into 5 × 10⁶ wells. 4 IFNγ ELISpot was performed using bone marrow dendritic cells as target cells. Effector splenocytes were tested against mutant peptides, wild-type peptides, and a control peptide (vesicular stomatitis virus nucleoprotein, VSV-NP).

[0307] Of the 44 sequences tested, we found that six of them induced T-cell immunity only against mutant sequences, but not against wild-type peptides (Figure 3).

[0308] The data demonstrate that identified and prioritized mutations can be used to induce tumor-specific T-cell immunity after being used as peptide vaccines in antigen-naive mice. Table 2: A list of mutant sequences that induced T cell reactivity specific to mutant peptides versus wild-type peptides. Amino acid exchanges are underlined. [Table 2]

[0309] (Example 3) The identified mutations can provide therapeutic antitumor immunity. To investigate whether the identified mutations have the potential to confer antitumor immunity after vaccination of naive mice, we examined this issue using peptides for mutation number 30, which have been shown to induce mutation-selective T cell reactivity. B16F10 cells (7.5 × 10⁶) 4 The mice were subcutaneously inoculated with ) on day 0. Peptide 30 (see Table 1; 100 μg peptide + 50 μg sc PolyI:C) was administered to the mice on days -4, +2, and +9. The control group was given only Poly I:C (50 μg sc). Tumor growth was observed every other day. On day +16, only one of the five mice in the peptide vaccine group developed a tumor, while four of the five mice in the control group showed tumor growth.

[0310] The data demonstrate that peptide sequences incorporating B16F10-specific mutations can confer antitumor immunity capable of efficiently destroying tumor cells (see Figure 4). Since B16F10 is a highly aggressive tumor cell line, the finding that the methods applied to identify and prioritize mutations ultimately led to the selection of mutations that are already potent as vaccines in themselves is a significant proof of concept for the entire process.

[0311] (Example 4) Data supporting polyepitope antigen presentation Demonstrated mutations from patient protein-coding regions constitute a pool from which candidate polyneoepitope vaccine template assemblies can be selected for use as precursors for GMP production of RNA vaccines. Suitable vector cassettes as vaccine scaffolds have already been described (Holtkamp, ​​S. et al., Blood, 108:4009-4017, 2006; Kreiter, S. et al., Cancer Immunol. Immunother., 56:1577-1587, 2007; Kreiter, S. et al., J.Immunol., 180:309-318, 2008). Preferred vector cassettes are modified in the coding and uncoding regions (UTRs) to ensure maximalized translation of encoded proteins over long periods (Holtkamp, ​​S. et al., Blood, 108:4009-4017, 2006; Kuhn, ANet al., Gene Ther., 17:961-971, 2010). Furthermore, the vector scaffold includes an antigen pathway designation module for the simultaneous proliferation of cytotoxic T cells and helper T cells (Kreiter, S. et al., Cancer Immunol. Immunother., 56:1577-1587, 2007; Kreiter, S. et al., J.Immunol., 180:309-318, 2008; Kreiter, S. et al., Cancer Research, 70(22), 9031-9040, 2010) (Figure 5). Importantly, we demonstrated that such RNA vaccines can be used to simultaneously present multiple MHC class I and class II epitopes.

[0312] IVAC polyneoepitope RNA vaccine sequences are constructed from stretches of up to 30 amino acids containing a central mutation. These sequences are ligated head-to-tail with short linkers to form polyneoepitope vaccines encoding up to 30 or more selected mutations and their flanking regions. These patient-specific, individually tailored inserts are codon-optimized and cloned into the RNA backbone described above. Quality control of such constructs includes in vitro transcription and in-cellular expression for functional transcription and translation validation. Translational analysis is performed using antibodies against the C-terminal target domain.

[0313] (Example 5) Scientific proof of concept for RNA polyneoepitope constructs The concept of RNA polyneoepitopes is based on long in vitro transcribed mRNAs consisting of sequentially arranged sequences encoding mutant peptides, linked by linker sequences (see Figure 6). The coding sequences are selected from non-synonymous mutations and are always constructed with codons for mutant amino acids adjacent to a region of 30–75 base pairs from the original sequence. The linker sequences encode amino acids that are not selectively processed by the cell's antigen processing mechanisms.

[0314] The in vitro transcription construct is based on the pST1-A120 vector, which contains a T7 promoter, a tandem β-globin 3'UTR sequence, and a 120 bp poly(A) tail. These have been shown to enhance RNA stability and translation efficiency, thereby increasing the T-cell stimulating ability of the encoded antigen (Holtkamp S. et al., Blood 2006; PMID:16940422). In addition, a transmembrane domain and cytosolic domain containing a termination codon (MHC class I transport signal or MITD) adjacent to a polylinker sequence for cloning an MHC class I signal peptide fragment and epitope were inserted (Kreiter S. et al., J.Immunol., 180:309-318, 2008). The latter increases antigen presentation, thereby enhancing antigen-specific CD8+ It has also been shown to enhance the proliferation of CD4+ T cells and improve effector function.

[0315] For the initial proof of concept, a biepitope vector, i.e., a vector encoding a single polypeptide containing two mutant epitopes, was used. Codon-optimized sequences encoding (i) a 20-50 amino acid mutant epitope, (ii) a glycine / serine-rich linker, (iii) a second 20-50 amino acid mutant epitope, and (iv) an additional glycine / serine-rich linker were designed adjacent to the appropriate recognition site of the restriction endonuclease for cloning into a pST1-based construct, as described above, and were synthesized by a commercial supplier (Geneart, Regensburg, Germany). After sequence validation, these were cloned into a pST1-based vector skeleton to obtain the construct shown in Figure 6.

[0316] The pST1-A120 plasmid described above was linearized with a class II restriction endonuclease. The linearized plasmid DNA was purified by phenol-chloroform extraction and ethanol precipitation. The linearized vector DNA was quantified by spectrophotometric analysis and subjected to in vitro transcription as basically described by Pokrovskaya and Gurevich (1994, Anal. Biochem. 220:420-423). A cap analog was added to the transcription reaction to obtain RNA with a correspondingly modified 5' cap structure. In the reaction mixture, GTP was present at 1.5 mM, and the cap analog at 6.0 mM. All other NTPs were present at 7.5 mM. At the end of the transcription reaction, the linearized vector DNA was digested with 0.1 U / μl TURBO DNase (Ambion, Austin / TX, USA) at 37°C for 15 minutes. RNA was purified from these reaction products using the MEGAclear Kit (Ambion, Austin / TX, USA) according to the manufacturer's protocol. RNA concentration and quality were evaluated by spectrophotometric analysis and analysis using a 2100 Bioanalyzer (Agilent, Santa Clara, CA, USA).

[0317] To demonstrate that sequences incorporating mutant amino acids, flanked by the linker sequence at 5' and 3', can be processed, presented, and recognized by antigen-specific T cells, we used peptide-vaccinated mouse T cells as effector cells. In IFNγ ELISpot, we tested whether the peptide-vaccinated T cells described above could recognize target cells (bone marrow dendritic cells, BMDCs) transfected with RNA (20 μg generated as described above) by either pulsing with the peptide (2 μg / ml, 37°C, and 5% CO2 for 2 hours) or electroporation. As illustrated in Figure 7 for mutations 12 and 30 (see Table 2), we observed that RNA constructs could produce epitopes recognized by mutation-specific T cells.

[0318] The data provided demonstrates that RNA encoding polyneoepitopes, including glycine / serine-rich linkers, can be translated and processed in antigen-presenting cells, resulting in the presentation of the correct epitope recognized by antigen-specific T cells.

[0319] (Example 6) Designing Polyneoepitope Vaccines - Linker Relevance A polyneoepitope RNA construct comprises a skeletal construct on which multiple somatic mutation-coding peptides linked by a linker peptide sequence are arranged. In addition to codon optimization and increased RNA stability and translation efficiency through the skeletal structure, one embodiment of an RNA polyneoepitope vaccine includes a linker designed to increase the presentation of MHC class I and II antigen peptides and reduce the presentation of harmful epitopes.

[0320] Linker: The linker sequence was designed to link multiple mutant-containing peptides. The linker should enable the generation and presentation of mutant epitopes and prevent the generation of harmful epitopes, such as those generated at junctional sutures between adjacent peptides or between the linker sequence and endogenous peptides. These "conjugated" epitopes can not only reduce the effectiveness of the vaccine by competing with the intended epitopes to be presented on the cell surface, but can also cause undesirable autoimmune responses. Therefore, we designed the linker sequence to a) avoid generating "conjugated" peptides that bind to MHC molecules, b) avoid proteasome processing that generates "conjugated" peptides, and c) be efficiently translated and processed by the proteasome.

[0321] To avoid the generation of "conjugation" peptides that bind to MHC molecules, we compared various linker sequences. For example, glycine inhibits strong binding at the MHC binding groove site [Abastado JP. et al., J Immunol. 1993 Oct 1;151(7):3569-75]. We examined multiple linker sequences and linker lengths and calculated the number of "conjugation" peptides that bind to MHC molecules. Using software tools from the Immune Epitope Database (IEDB, http: / / www.immuneepitope.org / ), we calculated the likelihood that a given peptide sequence contains a ligand that binds to an MHC class I molecule.

[0322] In the B16 model, we identified 102 expressed non-synonymous cell mutations predicted to be presented on MHC class I molecules. Using 50 confirmed mutations, we computer-designed various vaccine constructs, including those using no linker or different linker sequences, and computer-calculated the number of harmful "conjugated" peptides using the IEDB algorithm (Figure 8).

[0323] Table 5 shows the results for several different linkers, different linker lengths, and the use of no linker and five linkers. The number of conjugation peptides that bind to MHC ranges from 2 to 91 for 9-amino acid and 10-amino acid epitope predictions (top and middle). Linker size affects the number of conjugation peptides (bottom). For this sequence, the fewest 9-amino acid epitopes are predicted for the 7-amino acid linker sequence GGSGGGG.

[0324] Linkers 1 and 2 (see below) used in the experimentally tested RNA polyneoepitope vaccine constructs also showed a good number of predicted conjugated neoepitopes. This also applies to the prediction of decamers and decamers.

[0325] This reveals that the linker sequence is critical to the generation of undesirable MHC-binding epitopes. Furthermore, the length of the linker sequence affects the number of undesirable MHC-binding epitopes. We observe that G-rich sequences inhibit the generation of MHC-binding ligands. Table 3: Influence of linkers (10 amino acid epitopes). Predicted number of undesirable epitopes, defined as MHC class I binding epitopes containing conjugation sequences, for each peptide linker. Here, 10 amino acid epitopes are considered. Glycine-rich linkers have the fewest conjugation epitopes. [Table 3] Table 4: Influence of the linker moiety (9 amino acid epitopes). The predicted number of undesirable epitopes, defined as MHC class I binding epitopes containing the conjugation sequence, for each peptide linker. Here, 9 amino acid epitopes are considered. The glycine-rich linker has the fewest conjugation epitopes. [Table 4] Table 5: Influence of the linker portion. Predicted number of undesirable epitopes, defined as MHC class I binding epitopes containing the conjugation sequence, for each peptide linker. Here, 9-amino acid epitopes are considered. Top: Number of 9-amino acid conjugation epitopes for no linker and five distinct linkers. Middle: Number of 10-amino acid conjugation epitopes for no linker and five distinct linkers. Bottom: Number of 99-amino acid conjugation epitopes for similar linkers of different lengths. Glycine-rich linkers have the fewest conjugation epitopes. [Table 5]

[0326] To avoid proteasome processing that could generate "conjugation" peptides, we considered using different amino acids in the linker. Glycine-rich sequences reduce proteasome processing [Hoyt MA et al. (2006). EMBO J 25(8):1720-9; Zhang M. and Coffino P. (2004) J Biol Chem 279(10):8635-41]. Therefore, glycine-rich linker sequences work to minimize the number of linker-containing peptides that can be processed by the proteasome.

[0327] The linker should allow the mutation-containing peptide to be efficiently translated and processed by the proteasome. The amino acids glycine and serine are flexible [Schlessinger A and Rost B., Proteins. 2005 Oct 1;61(1):115-26]; including them in the linker results in a more flexible protein. We have incorporated glycine and serine into the linker to increase the flexibility of the protein, which should enable more efficient translation and processing by the proteasome, and subsequently better access to the antigenic peptide it encodes.

[0328] Therefore, the linker should be glycine-rich to prevent the formation of undesirable epitopes that bind to MHC, to hinder the proteasome's ability to process the linker peptide (which can be achieved by including glycine), and to be flexible to increase access to the mutation-containing peptide (which can be achieved by a combination of glycine and serine amino acids). For this reason, in one embodiment of the vaccine construct of the present invention, the sequences GGSGGGGSGG and GGSGGGSGGS are preferably included as linker sequences.

[0329] (Example 7) RNA Polyneoepitope Vaccine The RNA polyneoepitope vaccine construct is based on the pST1-A120 vector, which contains a T7 promoter, a tandem β-globin 3'UTR sequence, and a 120 bp poly(A) tail. These have been shown to enhance RNA stability and translation efficiency, thereby increasing the T-cell stimulating ability of the encoded antigen (Holtkamp S. et al., Blood 2006; PMID:16940422). In addition, a transmembrane domain and cytosolic domain containing a termination codon (MHC class I transport signal or MITD) adjacent to the polylinker sequence for cloning the MHC class I signal peptide fragment and epitope were inserted (Kreiter S. et al., J.Immunol., 180:309-318, 2008). The latter increases antigen presentation, thereby enhancing antigen-specific CD8 + It has also been shown to enhance the proliferation of CD4+ T cells and improve effector function.

[0330] Three RNA constructs were constructed to provide RNA polyneoepitope constructs for 50 identified and validated mutations in B16F10. Each construct consists of (i) a 25-amino acid mutant epitope, (ii) a glycine / serine-rich linker, and (iii) a codon-optimized sequence encoding the mutant epitope sequence followed by a repeat of the glycine / serine-rich linker. The mutant epitope-containing sequence and linker strands are adjacent to appropriate recognition sites of restriction endonucleases for cloning into the pST1-based constructs described above. Vaccine constructs were designed and synthesized by GENEART. After sequence validation, these were cloned into pST1-based vector backbones to obtain RNA polyneoepitope vaccine constructs.

[0331] Explanation of the clinical approach Clinical application involves the following steps: • Eligible patients must consent to DNA analysis using next-generation sequencing. • Obtain tumor specimens (paraffin-embedded and formalin-fixed tissue) and peripheral blood cells from standard diagnostic procedures and use them for the mutation analysis described above. • Confirm the discovered mutation. Based on prioritization, design the vaccine. For RNA vaccines, create master plasmid templates by gene synthesis and cloning. Plasmids are used for the production of clinical-grade RNA, and for quality control and release of RNA vaccines. • Send the vaccine drug product to each clinical trial site for clinical application. RNA vaccines can be used as naked vaccines in formulation buffers or encapsulated in nanoparticles or liposomes for direct injection into lymph nodes, subcutaneous, intravenous, or intramuscular injection. Alternatively, RNA vaccines can be used for in vitro transfection of dendritic cells, for example, for adoptive transfer.

[0332] The entire clinical process takes less than six weeks. The "delay" between patient informed consent and drug availability is carefully addressed by the clinical trial protocol, which includes allowing standard treatment regimens to continue until the investigational drug becomes available.

[0333] (Example 8) Its use for identifying tumor metastases and for tumor vaccination We applied NGS exome resequencing to discover mutations in the B16F10 mouse melanoma cell line, identifying 962 non-synonymous somatic point mutations and 563 in expressed genes. Potential driver mutations occur in classical tumor suppressor genes (Pten, Trp53, Tp63, Pml) as well as genes involved in proto-oncogene signaling pathways that regulate cell proliferation (e.g., Mdm1, Pdgfra), cell adhesion and migration (e.g., Fdz7, Fat1), or apoptosis (Casp9). Furthermore, B16F10 also harbors mutations in Aim1 and Trap, which have been previously described as frequently altered in human melanoma.

[0334] The immunogenicity and specificity of 50 demonstrated mutations were tested using C57BL / 6 mice immunized with long peptides encoding the mutant epitopes. One-third of these (16 / 50) were shown to be immunogenic. Of these, 60% selectively induced an immune response to the mutant sequences compared to the wild-type sequences.

[0335] We tested our hypothesis in a tumor transplantation model. Peptide-mediated immunization provided in vivo tumor suppression in both defensive and therapeutic settings, and we considered mutant epitopes, including single amino acid substitutions, to be effective vaccines.

[0336] animal C57BL / 6 mice (Jackson Laboratories) were bred at the University of Mainz in accordance with federal and state government policies regarding animal research.

[0337] cell The B16F10 melanoma cell line was purchased from the American Type Culture Collection in 2010 (product: ATCC CRL-6475, lot number: 58078645). Early (3rd and 4th generation) passaged cells were used in tumor experiments. Cells were typically tested for Mycoplasma species. No further cell verification was performed after receipt.

[0338] Next-generation sequencing Nucleic acid extraction and sample preparation: DNA and RNA from bulk B16F10 cells, as well as DNA from C57BL / 6 tail tissue, were extracted in triplicate sets using the Qiagen DNeasy Blood and Tissue kit (for DNA) and the Qiagen RNeasy Micro kit (for RNA).

[0339] DNA exome sequencing: Exome capture for DNA resequencing was performed in triplicate using a capture assay based on Agilent Sure-Select mouse solution (Gnirke A et al., Nat Biotechnol 2009;27:182-9), designed to capture all mouse protein-coding regions. 3 μg of purified genomic DNA (gDNA) was fragmented into 150-200 bp fragments using a Covaris S2 sonicator. The fragments were repaired at the ends according to the manufacturer's instructions, 5' phosphorylated, and 3' adenylated. Illumina paired-end adapters were ligated to the gDNA fragments using a 10:1 molar ratio of adapter to gDNA. The fragments were enriched before capture, and flow cell-specific sequences were added using Illumina PE PCR primers 1.0 and 2.0 for four PCR cycles. Adapter-linked gDNA fragments were hybridized to Agilent's SureSelect biotinylated mouse whole exome RNA library bait with 500 ng of PCR-enriched gDNA fragments at 65°C for 24 hours. The hybridized gDNA / RNA bait complexes were extracted using streptavidin-coated magnetic beads, washed, and the RNA bait was cleaved during elution in SureSelect elution buffer. These eluted gDNA fragments were captured and amplified by 10 cycles of PCR. The exome-enriched gDNA library was clustered using cBot with the Truseq SR cluster kit v2.5 at 7 pM, and 50 bp were sequenced using the Truseq SBS kit-HS 50 bp on an Illumina HiSeq2000.

[0340] RNA gene expression, "transcriptome" profiling (RNA-Seq): Barcoded mRNA-seq cDNA libraries were prepared in triplicates from 5 μg of total RNA (modified Illumina mRNA-seq protocol). mRNA was isolated using Seramag Oligo(dT) magnetic beads (Thermo Scientific) and fragmented using divalent cations and heat. The resulting fragments (160–220 bp) were converted to cDNA using random primers and SuperScript II (Invitrogen), and the second strand was then synthesized using DNA polymerase I and RNase H. The cDNA was repaired at the ends according to the manufacturer's instructions, 5' phosphorylated, and 3' adenylated. A single 3' T overhang Illumina multiplex-specific adapter was ligated with T4 DNA ligase using an adapter-to-cDNA insert ratio of 10:1 molars. The cDNA libraries were purified and size-selected at 200–220 bp (E-Gel 2% SizeSelect gel, Invitrogen). Concentration, addition of Illumina 6-nucleotide index sequences and flow cell-specific sequences were performed by PCR using Phusion DNA polymerase (Finnzymes). All purification up to this step was performed using 1.8x volume AgincourtAMPure XP magnetic beads. All quality control was performed using Invitrogen's Qubit HS assay, and fragment sizes were determined using Agilent's 2100 Bioanalyzer HS DNA assay. The barcoded RNA-Seq libraries were clustered and sequenced as described above.

[0341] NGS data analysis, gene expression: Sequence reads output from RNA samples were preprocessed according to the Illumina standard protocol, which included filtering of low-quality reads. Sequence reads were aligned to the mm9 reference genome sequence (Waterston RH et al., Nature 2002;420:520-62) using bowtie (version 0.12.5) (Langmead B et al., Genome Biol 2009;10:R25). For genome alignment, two mismatches were tolerated, and only the optimal alignment ("-v2-best") was recorded; default parameters were used for transcriptome alignment. Reads that could not be aligned to the genome sequence were aligned to a database of all possible exon-exon junction sequences of RefSeq transcripts (Pruitt KD et al., Nucleic Acids Res 2007;35:D61-D65). Expression levels were determined by intersecting the read coordinates with those of the RefSeq transcript. Overlapping exons and junction reads were counted and normalized to RPKM expression units (reads which map per kilobase of exon model per million mapped reads) (Mortazavi A et al., Nat Methods 2008;5:621-8).

[0342] NGS data analysis and somatic mutation detection: Somatic mutations were identified as described in Example 9. 50-nucleotide single-ended reads were aligned to the mm9 reference mouse genome using bwa (default option, version 0.5.8c) (Li H and Durbin R, Bioinformatics 2009;25:1754-60). Ambiguous reads mapping to multiple locations in the genome were removed. Mutations were identified using three software programs: samtools (version 0.1.8) (Li H, Bioinformatics 2011;27:1157-8), GATK (version 1.0.4418) (McKenna A et al, Genome Res 2010;20:1297-303), and SomaticSniper (http: / / genome.wustl.edu / software / somaticsniper) (Ding L et al., Hum Mol Genet 2010;19:R188-R196). A confidence value of the "false detection rate" (FDR) was assigned to the potential mutations identified in all B16F10 triplets (see Example 9).

[0343] Mutation selection, demonstration, and function Selection: Mutations had to meet the following criteria to be selected: (i) present in all B16F10 triplets but absent in all C57BL / 6 triplets, (ii) FDR ≤ 0.05, (iii) uniform in C57BL / 6, (iv) occurring in RefSeq transcripts, and (v) causing non-synonymous changes scored as true mutations. Selection for demonstration and immunogenicity testing required that the mutation be expressed in the gene (mean RPKM > 10 across the entire replicate).

[0344] Verification: DNA-derived mutations were classified as verified if they were confirmed by either Sanger sequencing or B16F10 RNA-Seq reads. All selected mutants were amplified from 50 ng of DNA derived from B16F10 cells and C57BL / 6 tail tissue using adjacent primers, and the purified product was visualized (QIAxcel system, Qiagen) and purified (QIAquick PCR Purification Kit, Qiagen). Amplicons of expected size were excised from the gel, purified (QIAquick Gel Extraction Kit, Qiagen), and subjected to Sanger sequencing using the positive primers used for PCR amplification (Eurofins MWG Operon, Ebersberg, Germany).

[0345] Functional impact: The effects of selected mutations were evaluated using SIFT (Kumar P et al., Nat Protoc 2009;4:1073-81) and POLYPHEN-2 (Adzhubei IA et al., Nat Methods 2010;7:248-9), programs that predict the functional importance of amino acids to protein function based on protein domain location and interspecies sequence conservation. Gene function was inferred using the Ingenuity IPA tool.

[0346] Synthetic peptides and adjuvants Ovalbumin Class I (OVA) 258-265 ), Class II (OVA Class II) 330-338 ), influenza nucleoprotein (Inf-NP) 366-374 ), vesicular stomatitis virus nucleoprotein (VSV-NP 52-59 ) and tyrosinase-related protein 2 (Trp2 180-188All peptides, including those listed above, were purchased from Jerini Peptide Technologies (Berlin, Germany). The synthetic peptides were 27 amino acids long and had either a mutant (MUT) or wild-type (WT) amino acid at position 14. Polyinosinic acid: Polycytidylic acid (poly(I:C), InvivoGen) was used as an adjuvant for subcutaneous injection. Inf-NP 366-374 We purchased peptide-specific MHC pentamers from ProImmune Ltd.

[0347] Mouse immunity Age-matched female C57BL / 6 mice were subcutaneously injected into the flank with 100 μg of peptide and 50 μg of poly(I:C) (total volume 200 μl) formulated in PBS (5 mice per group). All groups were immunized with two different mutant-coding peptides on days 0 and 7, with one peptide per flank. The mice were sacrificially killed 12 days after the initial injection, and splenocytes were isolated for immunological testing.

[0348] Alternatively, age-matched female C57BL / 6 mice were intravenously injected with 20 μg of in vitro transcription RNA formulated with 20 μl of Lipofectamine® RNAiMAX (Invitrogen) in PBS in a total injection volume of 200 μl (3 mice per group). All groups were immunized on days 0, 3, 7, 14, and 18. Mice were sacrificially killed 23 days after the initial injection, and splenocytes were isolated for immunological testing. DNA sequences exhibiting one (monoepitope), two (biepitope), or sixteen (polyepitope) mutations were constructed using 50 amino acids with a mutation at position 25 (biepitope) or 27 amino acids with a mutation at position 14 (monoepitope and polyepitope), separated by a 9-amino acid glycine / serine linker, and cloned into the pST1-2BgUTR-A120 backbone (Holtkamp et al., Blood 2006;108:4009-17). In vitro transcription and purification from this template have been previously described (Kreiter et al., Cancer Immunol Immunother 2007;56:1577-87).

[0349] Enzyme-linked immunoassay spot assay The enzyme-linked immunospot (ELISPOT) assay (Kreiter S et al., Cancer Res 2010;70:9031-40) and the generation of syngeneic myeloid dendritic cells (BMDCs) as stimulants have been previously described (Lutz MB et al., J Immunol Methods 1999;223:77-92). BMDCs were pulsed with peptides (2 μg / ml) or transfected with in vitro transcription (IVT) RNA encoding a designated mutant or control RNA (eGFP-RNA). Two mutant sequences, each containing 50 amino acids with a mutation at position 25 and separated by a 9-amino acid glycine / serine linker, were cloned into the pST1-2BgUTR-A120 backbone (Holtkamp S et al., Blood 2006;108:4009-17). In vitro transcription and purification from this template have been previously described (Kreiter S et al., Cancer Immunol Immunother 2007;56:1577-87). For the assay, 5 × 10 4 Peptide or RNA-modified BMDCs were placed in a microtiter plate coated with anti-IFN-γ antibody (10 μg / mL, clone AN18; Mabtech) in a 5 × 10⁶ size. 5 Fresh isolated splenocytes were co-incubated. After 18 hours at 37°C, cytokine secretion was detected with anti-IFN-γ antibody (clone R4-6A2; Mabtech). Spot counts were counted and analyzed using ImmunoSpot® S5 Versa ELISPOT Analyzer, ImmunoCapture® Image Acquisition software, and ImmunoSpot® Analysis software version 5. Statistical analysis was performed by Student's t-test and Mann-Whitney U test (non-parametric tests). A p-value < 0.05 was given in the test, or the mean spot count was > 30 spots / 5 × 10⁻⁶ 5 If the cells were effector cells, the response was considered significant. Reactivity was assessed by the average number of spots (-: <30; +: >30; ++: >50; +++: >200 spots / well).

[0350] Intracellular cytokine assay Aliquots of splenocytes prepared for the ELISPOT assay were subjected to intracellular flow cytometry analysis of cytokine production. For this purpose, 2 × 10⁶ samples were used for each sample. 6 Splenocytes were spread in 96-well plates on medium (RPMI + 10% FCS) supplemented with the Golgi inhibitor prefeldin A (10 μg / mL). Cells from each animal were divided into 2 × 10⁶ 5 Cells were re-stimulated with peptide-pulsed BMDC at 37°C for 5 hours. After incubation, cells were washed with PBS, resuspended in 50 μl of PBS, and extracellularly stained with the following anti-mouse antibodies: anti-CD4 FITC, anti-CD8 APC-Cy7 (BD Pharmingen) at 4°C for 20 minutes. After incubation, cells were washed with PBS and then resuspended in 100 μl of Cytofix / Cytoperm (BD Bioscience) solution at 4°C for 20 minutes for permeabilization of the outer membrane. After permeabilization, cells were washed with Perm / Wash-Buffer (BD Bioscience), resuspended in 50 μL / sample of Perm / Wash-Buffer, and intracellularly stained with the following anti-mouse antibodies: anti-IFN-γ PE, anti-TNF-α PE-Cy7, anti-IL2 APC (BD Pharmingen) at 4°C for 30 minutes. After washing with Perm / Wash-Buffer, cells were resuspended in PBS containing 1% paraformaldehyde for flow cytometry analysis. Samples were analyzed using the BD FACSCanto® II hemocytometer and FlowJo (version 7.6.3).

[0351] B16 Melanoma Tumor Model For tumor vaccination experiments, 7.5 × 10 4B16F10 melanoma cells were subcutaneously inoculated into the flanks of C57BL / 6 mice. In a prophylactic setting, immunization with a mutation-specific peptide was performed 4 days before tumor inoculation and on 2 and 9 days after tumor inoculation. For therapeutic experiments, the peptide vaccine was administered on 3 and 10 days after tumor injection. Tumor size was measured every 3 days, and mice were euthanized when the tumor diameter reached 15 mm.

[0352] Alternatively, for tumor vaccination experiments, 1 × 10 5 B16F10 melanoma cells were subcutaneously inoculated into the flanks of age-matched female C57BL / 6 mice. Peptide vaccination was performed by subcutaneous injection into the flanks of 100 μg of peptide and 50 μg of poly(I:C) (total volume 200 μl) formulated in PBS at 3, 10, and 17 days after tumor inoculation. RNA immunization was performed using 20 μg of in vitro transcription RNA encoding a mutation, formulated in PBS with 20 μl of Lipofectamine® RNAiMAX (Invitrogen) at a total injection volume of 200 μl. As a control, one group of animals was injected with RNAiMAX (Invitrogen) in PBS. Animals were immunized at 3, 6, 10, 17, and 21 days after tumor inoculation. Tumor size was measured every 3 days using a calipas, and mice were euthanized when the tumor diameter reached 15 mm.

[0353] Identification of non-synonymous mutations in B16F10 mouse melanoma Our objective was to identify potentially immunogenic somatic point mutations in B16F10 mouse melanoma using NGS, test them for in vivo immunogenicity by peptide vaccination of mice, and measure the induced T cell response by the ELISPOT assay (Figure 9A). We extracted, captured, and sequenced triplets of the C57BL / 6 wild-type background genome and the exomes of B16F10 cells. For each sample, more than 100 million single-ended 50-nucleotide reads were generated. Of these, 80% aligned specifically to the mouse mm9 genome and 49% aligned to the target, demonstrating successful target enrichment and resulting in more than 20-fold coverage for 70% of the target nucleotides in each triplet sample. RNA-Seq of B16F10 cells, also profiled in triplets, generated an average of 30 million single-ended 50-nucleotide reads, of which 80% aligned to the mouse transcriptome.

[0354] DNA reads (exome capture) from B16F10 and C57BL / 6 were analyzed to identify somatic mutations. Copy number difference analysis (Sathirapongsasuti JF et al., Bioinformatics 2011;27:2648-54) revealed deletions in B16F10, including DNA amplification and homozygous deletions of the tumor suppressor Cdkn2a (cyclin-dependent kinase inhibitor 2A, p16Ink4A). Focusing on point mutations to identify potential immunogenic mutations, we identified 3570 somatic point mutations with an FDR ≤ 0.05 (Figure 9B). The most frequent class of mutations was C>T / G>A transitions, typically resulting from ultraviolet light (Pfeifer GP et al., Mutat Res 2005;571:19-31). Of these somatic mutations, 1392 occurred in transcripts, and 126 mutations occurred in uncoding regions. Of the 1266 mutations in coding regions, 962 caused non-synonymous protein changes, and of these, 563 occurred in expressed genes (Figure 9B).

[0355] Assignment and demonstration of identified mutations to carrier genes. Notably, many of the mutant genes (962 genes, including non-synonymous somatic point mutations) have been associated with cancer phenotypes. Mutations were found in established tumor suppressor genes, including Pten, Trp53 (also known as p53), and Tp63. In Trp53, the most widely established tumor suppressor (Zilfou JT et al., Cold Spring Harb Perspect Biol 2009;1:a001883), a mutation from asparagine to aspartic acid at protein position 127 (p.N127D) is predicted to localize within the DNA-binding domain and alter function via SIFT. Pten contains two mutations (p.A39V, p.T131P), both of which are predicted to have detrimental effects on protein function. The p.T131P mutation is adjacent to a mutation (p.R130M) that has been shown to reduce phosphatase activity (Dey N et al., Cancer Res 2008;68:1862-71). Furthermore, mutations were found in genes related to DNA repair pathways, such as Brca2 (breast cancer 2, juvenile), Atm (mutation in ataxia diastolecularis), Ddb1 (damage-specific DNA-binding protein 1), and Rad9b (RAD9 homolog B). Furthermore, mutations also occurred in other tumor-related genes, including Aim1 (tumor suppressor "absent-in-melanoma 1"), Flt1 (oncogene Vegr1, fms-related tyrosine kinase 1), Pml (tumor suppressor "promyelocytic leukemia"), Fat1 ("FAT tumor suppressor homolog 1"), Mdm1 (TP53-binding nucleoprotein), Mta3 (metastasis-related family 1, member 3), and Alk (anaplastic lymphoma receptor tyrosine kinase). We found a mutation at p.S144F in Pdgfra (platelet-derived growth factor receptor α polypeptide) (Verhaak RG et al., Cancer Cell 2010;17:98-110), a cell membrane-bound receptor tyrosine kinase of the MAPK / ERK pathway that had been previously identified in tumors. The mutation occurs at p.L222V in Casp9 (caspase 9, apoptosis-related cysteine ​​peptidase).CASP9 cleaves poly(ADP-ribose) polymerase (PARP) through proteolysis, regulates apoptosis, and is associated with several cancers (Hajra KM et al., Apoptosis 2004;9:691-704). The mutations we identified potentially affect PARP and apoptosis signaling. Most interestingly, no mutations were found in Braf, c-Kit, Kras, or Nras. However, mutations were identified in Rassf7 (RAS-related protein) (p.S90R), Ksr1 (ras 1 kinase suppressor) (p.L301V), and Atm (PI3K pathway) (p.K91T), all of which are predicted to have a significant impact on protein function. Trrap (a transformation / transcription domain-associated protein) was identified this year in human melanoma specimens as a novel potential melanoma target (Wei X et al., Nat Genet 2011;43:442-6). In B16F10, the Trrap mutation is predicted to occur at p.K2783R, impairing the overlapping phosphatidylinositol kinase (PIK)-associated kinase FAT domain.

[0356] From 962 non-synonymous mutations identified using NGS, we selected 50 mutations, including 41 mutations with an FDR < 0.05, for PCR-based validation and immunogenicity testing. Selection criteria were the location in the expressed gene (RPKM > 10) and predicted immunogenicity. Notably, we were able to validate all 50 mutations (Table 6, Figure 9B). Table 6: Mutations selected for demonstration. From left: Assigned ID, gene symbol, amino acid substitution and position, gene name, predicted intracellular localization and type (Ingenuity). [Table 6]

[0357] Figure 9C shows the location, gene density, gene expression, mutations, and filtered mutations (inner ring) of chromosome B16F10.

[0358] In vivo immunogenicity testing of long peptides exhibiting mutations To provide antigens for immunogenicity testing of these mutations, we used long peptides, which offer several advantages over other peptides for immunotherapy (Melief CJ and van der Burg SH, Nat Rev Cancer 2008;8:351-60). The long peptides are antigen-specific CD8 +Furthermore, it can induce CD4+ T cells (Zwaveling S et al., Cancer Res 2002;62:6187-93; Bijker MS et al., J Immunol 2007;179:5033-40). In addition, long peptides require processing to be presented to MHC molecules. Such uptake is most efficiently carried out by dendritic cells, which are optimal for priming a potent T cell response. Suitable peptides, on the other hand, do not require trimming and are exogenously loaded onto all cells expressing MHC molecules, including inactivated B and T cells, leading to immunological tolerance and fructoliside induction (Toes RE et al., J Immunol 1996;156:3911-8; Su MW et al., J Immunol 1993;151:658-67). For each of the 50 demonstrated mutations, we designed a 27-amino acid peptide containing either the centrally located mutation or the wild-type amino acid. Therefore, any potential MHC class I and class II epitopes of 8–14 amino acid lengths carrying the mutation could be processed from this precursor peptide. As an adjuvant for peptide vaccination, we used poly(I:C), known to promote cross-presentation and enhance vaccine efficacy (Datta SK et al., J Immunol 2003;170:4102-10; Schulz O et al., Nature 2005;433:887-92). The 50 mutations were tested in vivo in mice for T cell induction. Impressively, 16 of the peptides encoding the 50 mutations were found to induce an immune response in immunized mice. The induced T cells exhibited diverse reactivity patterns (Table 7). Table 7: Summary of T-cell reactivity measured following vaccination with mutation-encoding peptides. Statistical analysis was performed by Student's t-test and Mann-Whitney test (non-parametric test). A p-value < 0.05 was given in the test, or the mean number of spots > 30 spots / 5 × 10⁻⁶ 5If the cells were effector cells, the response was considered significant. Reactivity was assessed by the average number of spots (-: <30; +: >30; ++: >50; +++: >200 spots / well). [Table 7]

[0359] Eleven peptides induced an immune response that selectively recognized mutant epitopes. This is exemplified in mice immunized with mutants 30 (MUT30, Kif18b) and 36 (MUT36, Plod2) (Figure 10A). The ELISPOT test revealed a potent mutant-specific immune response without cross-reactivity to wild-type peptides or an unrelated control peptide (VSV-NP). Five peptides, including mutants 05 (MUT05, Eef2) and 25 (MUT25, Plod2) (Figure 10A), resulted in immune responses that equally recognized both mutant and wild-type peptides. The majority of mutant peptides failed to induce a significant T-cell response, as exemplified by mutants 01 (MUT01, Fzd7), 02 (MUT02, Xpot), and 07 (MUT07, Trp53). The immune response induced by some of the discovered mutations is sufficiently positive as a positive control for mouse melanoma tumor antigen tyrosinase-related protein 2 (Trp2). 180-188 Immunogenicity (500 spots / 5×10) is produced by immunizing mice with MHC class I epitopes as described in Figure 10A. 5The response was within the range of cells (Bloom MB et al., Exp Med 1997;185:453-9; Schreurs MW et al. Cancer Res 2000;60:6995-7001). For selected peptides that induce a potent mutation-specific T cell response, we confirmed immunorecognition by an independent approach. Instead of the long peptides, we used in vitro transcription RNAs (IVT RNAs) encoding the mutant peptide fragments MUT17, MUT30, and MUT44 for immunological readout. In the ELISPOT assay, BMDCs transfected with the mutant-encoding RNA or unrelated RNA were used as antigen-presenting cells (APCs), and immunized mouse splenocytes were used as the effector cell population. BMDCs transfected with mRNA encoding MUT17, MUT30, and MUT44 were specifically and potently recognized by mouse splenocytes immunized with the respective long peptides (Figure 10B). Significantly lower reactivity was recorded for transfected BMDCs with control RNA, which is likely due to nonspecific activation of BMDCs by single-stranded RNA (Student's t-test; MUT17: p=0.0024, MUT30: p=0.0122, MUT44: p=0.0075). These data confirm that the induced mutant-specific T cells do indeed recognize the endogenously processed epitope. Two mutations that induce favorable recognition of the mutant epitope are located in the genes Actn4 and Kif18b. The somatic mutation in ACTN4 (actinin, α4) is located at p.F835V within the calcium-binding "EF hand" protein domain. Both SIFT and POLYPHEN predict a significant impact of this mutation on protein function, although this gene is not an established oncogene. However, mutation-specific T cells against ACTN4 have recently been associated with favorable patient outcomes (Echchakir H et al., Cancer Res 2001;61:4078-83).KIF18B (kinesin family member 18B) is a kinesin with ATP and nucleotide binding involved in the regulation of microtubule motility and cell division (Lee YM et al., Gene 2010;466:16-25) (Figure 10C). The DNA sequence at the position encoding p.K739 is homogeneous in reference C57BL / 6, but B16F10 DNA reads reveal a heterozygous somatic mutation. Both nucleotides were detected in B16F10 RNA-Seq reads and confirmed by Sanger sequencing. KIF18B has not previously been associated with cancer phenotypes. The mutant p.K739N does not localize to any known functional or conserved protein domain (Figure 10C, bottom), and is therefore most likely a passenger mutation rather than a driver mutation. These examples suggest that there is no correlation between the ability to induce an immune response recognizing the mutation and its functional or immunological relevance.

[0360] In vivo evaluation of the antitumor activity of vaccine candidates. To evaluate whether an in vivo-induced immune response leads to antitumor activity in tumor-bearing mice, we selected MUT30 (a mutation in Kif18b) and MUT44 as examples. These mutations had been shown to selectively induce a potent immune response to the mutant peptide and to be processed endogenously (Figure 10A, B). The therapeutic potential of vaccination with the mutant peptide was estimated at 7.5 × 10⁻⁶. 5 We investigated the effects of immunizing mice with either MUT30 or MUT44 as an adjuvant 3 and 10 days after B16F10 transplantation. Tumor growth was inhibited by vaccination with both peptides compared to the control group (Figure 11A). Since B16F10 is a highly aggressive tumor, we also tested a protective immune response. Mice were immunized with the MUT30 peptide and showed 7.5 × 10⁶ tumor growth at 4 days. 5B16F10 cells were subcutaneously inoculated, and booster immunization with MUT30 was performed 2 and 9 days after tumor attack induction. MUT30-treated mice showed complete tumor protection and 40% survival, while all mice in the control group died within 44 days (Figure 11B, left). Mice that developed tumors despite immunization with MUT30 showed slower tumor growth and a 6-day extension in average survival time compared to the control group (Figure 11B, right). These data suggest that vaccination against a single mutation can already confer antitumor activity.

[0361] Immunity with RNA encoding mutations Various RNA vaccines were constructed using 50 demonstrated mutations from the B16F10 melanoma cell line. DNA sequences exhibiting one (monoepitope), two (biepitope), or 16 (polyepitope) different mutations were constructed using 50 amino acids with a mutation at position 25 (biepitope) or 27 amino acids with a mutation at position 14 (monoepitope and polyepitope), and isolated by a 9-amino acid glycine / serine linker. These constructs were cloned into the pST1-2BgUTR-A120 backbone for in vitro transcription of mRNA (Holtkamp et al., Blood 2006;108:4009-17).

[0362] To test the in vivo ability to induce T cell responses to various RNA vaccines, groups of three C57BL / 6 mice were immunized with RNA and RNAiMAX lipofectamine formulations followed by intravenous injection. After five immunizations, the mice were sacrificially killed and restimulated with the corresponding mutant-coding peptide or control peptide (VSV-NP). Splenocytes were then analyzed for mutation-specific T cell responses using intracellular cytokine staining and IFN-γ ELISPOT analysis.

[0363] Figure 12 shows an example for each vaccine design. In the top column, mice were vaccinated with monoepitope RNA encoding MUT30 (mutation in Kif18b), which induces MUT30-specific CD4+ T cells (see illustrative FACS plot). In the middle column, the graph and FACS plot show the induction of MUT08 (mutation in Ddx23)-specific CD4+ T cells after immunization with biepitopes encoding MUT33 and MUT08. In the bottom column, mice were immunized with polyepitopes encoding 16 different mutations, including MUT08, MUT33, and MUT27 (see Table 8). The graph and FACS plot illustrate that MUT27-responsive T cells exhibit a CD8 phenotype. Table 8: Outline of mutations and gene names encoded by monoepitope, biepitope, and polyepitope RNA vaccines [Table 8]

[0364] Using the same polyepitope, we generated the data shown in Figure 13. The graphs show ELISPOT data after restimulation of splenocytes with control (VSV-NP), MUT08, MUT27, and MUT33 peptides, demonstrating that polyepitope vaccines can induce specific T cell responses to several different mutations.

[0365] Considering these factors together, the data suggest the possibility of inducing mutation-specific T cells using RNA encoding monoepitope, biepitope, and polyepitope. Furthermore, the data suggest that CD4 from a single construct can be used. + This also shows induction of CD8+ T cells and induction of several different specificities.

[0366] Immunity using model epitopes To further characterize the polyepitope RNA vaccine design, we constructed DNA sequences containing five different known model epitopes, including one MHC class II epitope (ovalbumin class I (SIINFEKL), class II (OVA class II), influenza nucleoprotein (Inf-NP), vesicular stomatitis virus nucleoprotein (VSV-NP), and tyrosinase-related protein 2 (Trp2)). We isolated the epitopes with the same 9-amino acid glycine / serine linker used for the mutant polyepitope. We cloned this construct into the pST1-2BgUTR-A120 backbone for in vitro transcription of mRNA.

[0367] Five C57BL / 6 mice were vaccinated using in vitro transcribed RNA by intranodal immunization (four immunizations with 20 μg of RNA into inguinal lymph nodes). Five days after the last immunization, blood samples and splenocytes were collected from the mice for analysis. Figure 14A shows IFN-γ ELISPOT analysis of splenocytes restimulated with indicated peptides. It is clearly observed that all three MHC class I epitopes (SIINFEKL, Trp2, and VSV-NP) induce a very high number of antigen-specific CD8+ T cells. The MHC class II epitope OVA class II also induces a potent CD4+ T cell response. A fourth MHC class I epitope was analyzed by staining Inf-NP-specific CD8+ T cells with a fluorescently labeled pentameric MHC peptide complex (pentamer) (Figure 14B).

[0368] These data demonstrate that polyepitope design using glycine / serine linkers to isolate different immunogenic MHC class I and class II epitopes can induce specific T cells against all encoded epitopes, regardless of their immunodominance.

[0369] Antitumor response after treatment with a polyepitope RNA vaccine encoding mutations Using the same polyepitope analyzed for immunogenicity in Figure 13, we investigated the antitumor activity of mutant encoding RNA against B16F10 tumor cells. Specifically, in a group of C57BL / 6 mice (n=10), 1 × 10⁶ 5 B16F10 melanoma cells were subcutaneously inoculated into the flank. Mice were immunized with polyepitope RNA using liposome transfection reagents on days 3, 6, 10, 17, and 21. The control group received only liposome injections.

[0370] Figure 21 shows the survival curves for both groups, revealing a significantly improved average survival time of 27 days compared to the 18.5-day average survival time in the control group, with one out of ten mice surviving without tumors.

[0371] Antitumor response after treatment with a combination of mutant and normal peptides The demonstrated antitumor activity of the mutation was evaluated by therapeutic in vivo tumor experiments using MUT30 as a peptide vaccine. Specifically, a group of C57BL / 6 mice (n=8) were subjected to 1 × 10⁶ experiments. 5 B16F10 melanoma cells were subcutaneously inoculated into the flank. On days 3, 10, and 17, polyI:C was used as an adjuvant to inoculate MUT30 and tyrosinase-related protein 2 (Trp2). 180-188 Mice were immunized with either Trp2 or a combination of both peptides. Trp2 is a known CD8 expressed by B16F10 melanoma cells. + It is an epitope.

[0372] Figure 15A shows the mean tumor growth in each group. In groups immunized with a combination of a known CD8+ T cell epitope and MUT30, which induces CD4+ T cells, tumor growth was clearly inhibited almost completely until day 28. While the known Trp2 epitope alone was not sufficient to provide a good antitumor effect in this setting, both single-treatment groups (MUT30 and Trp2) still provided inhibition of tumor growth compared to the untreated group from the start of the experiment to day 25. These data are supported by the survival curves shown in Figure 15B. Clearly, the mean survival rate was increased in mice injected with a single peptide, with 1 / 8 of the mice vaccinated with Trp2 surviving. In addition, the group treated with both peptides showed an even better survival rate, with 2 / 8 of the mice surviving.

[0373] When considered together, the two epitopes work synergistically to provide a potent antitumor effect.

[0374] (Example 9) A framework for confidence-based somatic mutation detection and its application to B16F10 melanoma cells. NGS is unbiased in that it enables high-throughput detection of mutations throughout the genome or within targeted regions such as protein-coding exons.

[0375] However, despite its groundbreaking nature, NGS platforms are still prone to errors that lead to false mutation calls. Furthermore, the quality of results depends on the parameters and analysis methods of the experimental design. Mutation calls typically include scores designed to distinguish true mutations from errors, but the usefulness of these scores is not fully understood, nor is their interpretation in terms of optimizing the experiment. This is especially true when comparing tissue states, i.e., comparing tumor tissue with normal tissue in terms of somatic mutations. As a result, researchers are forced to rely on personal experience in determining experimental parameters and arbitrary filtering thresholds for selecting mutations.

[0376] Our study aims to a) establish a framework for comparing parameters and methods for identifying somatic mutations, and b) assign confidence values ​​to identified mutations. We sequence triplicate samples from C57BL / 6 mice and the B16F10 melanoma cell line. Using these data, we formulate a false detection rate of detected somatic mutations, which will then be used to evaluate existing mutation detection software and experimental protocols.

[0377] Various experimental and algorithmic factors contribute to the false positive rate for mutations identified by NGS [Nothnagel, M. et al., Hum. Genet. 2011 Feb 23 [Epub ahead of print]]. Error sources include PCR artifacts, priming bias [Hansen, KD, et al., Nucleic. Acids. Res. 38, e131 (2010); Taub, MA et al., Genome Med. 2, 87 (2010)], bias in target enrichment [Bainbridge, MNet al., Genome Biol. 11, R62 (2010)], sequence influence [Nakamura, K. et al., Acids Res. (2011) first published online May 16, 2011 doi:10.1093 / nar / gkr344], base calling causing sequence errors [Kircher, M. et al., Genome Biol. 10, R83 (2009). Epub 2009 Aug 14], and further downstream analyses, such as mutation calling around indels [Li, H., Bioinformatics]. This includes read alignments that cause coverage variations and sequencing errors that affect 27,1157-1158 (2011) [Lassmann, T. et al., Bioinformatics 27,130-131 (2011)].

[0378] No general statistical model has been described to explain the influence of various error sources on somatic mutation calling; only individual aspects are covered without removing all biases. Recent computer methods for measuring the expected amount of false-positive mutation calls include the use of transition / transversion ratios of mutation sets [Zhang, Z., Gerstein, M., Nucleic Acids Res 31, 5338-5348 (2003); DePristo, MA et al., Nature Genetics 43, 491-498 (2011)], machine learning [DePristo, MA et al., Nature Genetics 43, 491-498 (2011)], and inheritance errors when working with family genomes [Ewen, K Ret al., Am. J. Hum. Genet. 67, 727-736 (2000)] or pooled samples [Druley, TE et al., Nature Methods 6, 263-265 (2009); Bansal, V., Bioinformatics 26, 318-324 (2010)]. For optimization, Druley et al. [Druley, TE et al., Nature Methods 6, 263-265 (2009)] used short plasmid sequence fragments, but these may not be representative of the sample. While it is possible to compare sets of single nucleotide variants (SNVs) and selected experiments with SNVs identified by other techniques [Van Tassell, CP et al., Nature Methods 5, 247-252 (2008)], it is difficult to evaluate novel somatic mutations.

[0379] As an example, using an exome sequencing project, we propose calculating the false detection rate (FDR) based solely on NGS data. This method can be applied to the selection and prioritization of diagnostic and therapeutic targets, and also supports the development of algorithms and methods by enabling us to define confidence-driven recommendations for similar experiments.

[0380] To discover mutations, DNA from the tail tissue of three C57BL / 6 (black6) mice (littermates) and DNA from B16F10 (B16) melanoma cells were enriched separately in triplicate form for protein-coding exons (Agilent Sure Select Whole Mouse Exome) to obtain six samples. RNA was extracted from B16 cells in triplicate form. Single-ended 50-nucleotide (1 × 50 nt) reads and paired-ended 100-nucleotide (2 × 100 nt) reads were generated using Illumina HiSeq 2000. Each sample was loaded into a separate lane, yielding an average of 104 million reads per lane. DNA reads were aligned to the mouse reference genome using bwa [Li, H. Durbin, R., Bioinformatics 25, 1754-1760 (2009)], and RNA reads were aligned using bowtie [Langmead, B. et al., Genome Biol. 10, R25 (2009)]. An average coverage of 38 times 97% of the target region was achieved for a 1 × 50 nt library, and the 2 × 100 nt experiment resulted in an average coverage of 165 times 98% of the target region.

[0381] Somatic mutations were independently identified using the software packages SAMtools [Li, H. et al., Bioinformatics 25, 2078-2079 (2009)], GATK [DePristo, MA et al., Nature Genetics 43, 491-498 (2011)], and SomaticSNiPer [Ding, L. et al., Hum.Mol.Genet (2010) first published online September 15, 2010] (Figure 16) by comparing single nucleotide mutations found in B16 samples with corresponding loci in black6 samples (B16 cells were originally derived from black6 mice). Latent mutations were filtered, respectively, according to the recommendations of the respective software creators (SAMtools and GATK) or by selecting appropriate lower thresholds for the somatic cell score in SomaticSNiPer.

[0382] To create a false detection rate (FDR) for mutation detection, we first intersected mutation sites and obtained 1,355 high-quality somatic mutations as a consensus across all three programs (Figure 17). However, the differences observed in the results of the applied software tools were substantial. To avoid erroneous conclusions, we developed a method to assign an FDR to each mutation using replication. Technical replicates of the sample should produce identical results, and any mutations detected in this “vs. identical comparison” are false positives. Therefore, to determine the false detection rate for somatic mutation detection in tumor samples compared to normal samples ("tumor comparison"), we can use technical replicates of normal samples as a reference to estimate the number of false positives.

[0383] Figure 18A shows examples of mutations found in the black6 / B16 data, including somatic mutations (left), non-somatic mutations (center), and potential false positives (right) relative to the reference. Each somatic mutation can be associated with a quality score Q. The number of false positives in tumor comparisons represents the number of false positives in identical comparisons. Therefore, for a given mutation with a quality score Q detected in tumor comparisons, we estimate the false detection rate by computer-calculating the ratio of identical mutations with a Q or better score to the total number of mutations found in tumor comparisons with a Q or better score.

[0384] Most mutation detection frameworks calculate multiple quality scores, which presents a problem in defining Q. Here, we apply a random forest classifier [Breiman, L., Statist. Sci. 16, 199-231 (2001)] to combine multiple scores into a single quality score Q. See the Methods chapter for details on quality score and FDR calculation.

[0385] The potential bias in the comparative method is differential coverage; therefore, we normalize the false detection rate with respect to coverage:

number

[0386] We calculate common coverage by counting all bases of the reference genome covered by both tumor and normal samples, or by both "same vs. same" samples.

[0387] By estimating the number of false positives and positives for each FDR (see "Methods"), we construct receiver operational characteristic (ROC) curves and calculate the area under the curve (AUC) for each mutation detection method, thus enabling a comparison of strategies for mutation detection (Figure 18B).

[0388] Furthermore, the selection of reference data can affect the calculation of FDR. Using the available black6 / B16 data, it is possible to generate 18 triplets (combinations of black6 vs. black6 and black6 vs. b16). When comparing the resulting FDR distributions for sets of somatic mutations, the results are in agreement (Figure 18B).

[0389] Using this definition of false detection rate, we established a general framework for evaluating the impact of numerous experimental and algorithmic parameters on the set of somatic mutations that occur. Next, we apply this framework to examine the impact of software tools, coverage, paired-end sequencing, and the number of technical replicates on somatic mutation identification.

[0390] Firstly, the choice of software tool has a clear impact on the somatic mutations identified (Figure 19A). With respect to the data tested, SAMtools yielded the highest enrichment of true positives in the set of somatic mutations ranked by FDR. However, we note that all tools provide numerous parameters and quality scores for individual mutations. Here we used the default settings specified by the algorithm developers; we anticipate that the parameters can be optimized, and we emphasize that the FDR framework defined here is designed to perform and evaluate such optimizations.

[0391] For the aforementioned B16 sequencing experiment, we sequenced each sample in individual flow cell lanes, achieving an average base coverage of 38x for the target region for each sample. However, this coverage is not considered necessary to obtain an equally good set of somatic mutations and would likely reduce costs. The impact of coverage depth on whole-genome SNV detection has also been recently discussed [Ajay, SSet al., Genome Res. 21, 1498-1505 (2011)]. To examine the impact of coverage on exon capture data, we downsampled the number of aligned sequence reads for all 1×50nt libraries to generate approximate coverages of 5x, 10x, and 20x, respectively, and then reapplied the mutation calling algorithm. As expected, higher coverage resulted in a better (i.e., fewer false positives) set of somatic mutations, but the improvement from 20x coverage to the maximum was small (Figure 19B).

[0392] The proper approach is to simulate and rank various experimental setups using available data and frameworks. Comparing the duplex to the tripplex, the tripplex does not offer any advantage over the duplex (Figure 19C), but the duplex provides a clear improvement over the no-replicate trial. With respect to the ratio of somatic mutations in a given set, we observe enrichment at 5% FDR from 24.2% to 71.2% for the duplex compared to the no-replicate run, and 85.8% for the tripplex. Despite the enrichment, using the tripplex intersection removes more mutations with lower FDR than those with higher FDR, as indicated by the lower ROC AUC and the leftward shift of the curve (Figure 19C): specificity is slightly increased at the cost of lower sensitivity.

[0393] Using an additionally sequenced 2×100nt library, five simulated libraries were created by in silico removal of a second read and / or the 3' and 5' ends of the reads, resulting in 1×100nt, two 2×50nt, and two 1×50nt libraries, respectively. These libraries were compared using the calculated FDR of predicted mutations (Figure 19D). Despite a much higher average coverage (≥77 vs. 38), somatic mutations found using the 2×50nt 5' and 1×100nt libraries had lower ROC AUCs and therefore a poorer FDR distribution than the 1×50nt library. This phenomenon stems from the accumulation of high-FDR mutations in low-coverage regions because the sets of low-FDR mutations found are highly similar. As a result, the optimal sequencing length must be either small enough so that the sequenced bases are concentrated near the capture probe sequence (potentially losing information about somatic mutation status in uncovered regions) or close to the fragment length to effectively fill the coverage gap (in our case, 2 × 100 nt = 200 nt total length for a fragment of about 250 nt). This is further supported by the fact that the ROC AUC of a 2 × 50 nt 3' library (simulated by using only the 3' end of a 2 × 100 nt library) is higher than that of a 2 × 50 nt 5' library (simulated by using only the 5' end of a 2 × 100 nt library), despite the lower base quality of the 3' read ends.

[0394] These observations allow us to define the best practice procedures for somatic mutation detection. Near-optimal results are achieved in these relatively homogeneous samples, while also considering cost, by using 20x coverage and technical duplication across all evaluation parameters in both samples. A 1×50nt library, producing approximately 100 million reads, appears to be the most practical choice for achieving this coverage. This holds true across all possible dataset pairs. We retrospectively applied these parameter settings and calculated the FDR for 50 selected mutations from all three method intersections shown in Figure 17, without using additional filtering of live mutation calls. All mutations were confirmed by a combination of Sanger resequencing and B16 RNA-Seq sequence reads. Of these mutations, 44 should have been found using a 5% FDR cutoff value (Figure 20). As negative controls, we rearranged the loci of 44 predicted mutations with high FDR (>50%) and examined their respective sequences in the RNA-Seq data. We found that 37 of these mutations were not demonstrated, and the remaining 7 loci of potential mutations were not covered by RNA-Seq reads, resulting in no sequencing response.

[0395] We demonstrate the application of the framework to four specific problems, but it is by no means limited to these parameters and can be applied to examine the influence of all experimental or algorithmic parameters, such as the influence of alignment software, the choice of mutation metrics, or the vendor's choice for exome selection.

[0396] We performed all experiments on a set of B16 melanoma cell experiments; however, this method is not limited to these data. The only requirement is that a “vs. identical” reference dataset is available, meaning that at least one technical replicate of non-tumor samples should be performed for each new protocol. Our experiments demonstrate that this method is robust in terms of the selection of technical replicates to a certain extent, but therefore replicates are not necessarily required in every single experiment. However, this method requires that various quality assessment measures be equivalent between the reference dataset and the remaining datasets.

[0397] Within the scope of this contribution, we developed a statistical framework for false-detection rate-driven detection of somatic mutations. This framework can be applied not only to diagnostic or therapeutic target selection but also to general comparisons of experimental and computer protocol processes regarding the generated false-true data. Here, we applied this idea to determine protocols with respect to software tools, coverage, replication, and paired-end sequencing.

[0398] method Library capture and sequencing Next-generation sequencing, DNA sequencing: In this case, exome capture for DNA resequencing was performed using a capture assay based on Agilent Sure-Select solution [Gnirke, A., et al., Nat. Biotechnol. 27, 182-189 (2009)], which was designed to capture all known mouse exons.

[0399] 3 μg of purified genomic DNA was fragmented into 150-200 nt fragments using a Covaris S2 sonicator. The gDNA fragments were repaired at the ends using T4 DNA polymerase and Krenow DNA polymerase, and 5' phosphorylated using T4 polynucleotide kinase. The blunt-end gDNA fragments were 3' adenylated (3'-5' exo-minus) using Krenow fragments. A single 3' T overhang Illumina paired-end adapter was ligated to the gDNA fragment using T4 DNA ligase in a 10:1 molar ratio of adapter to genomic DNA insert. The adapter-ligated gDNA fragments were enriched before capture, and flow cell-specific sequences were added using Illumina PE PCR primers 1.0 and 2.0 and Herculase II polymerase (Agilent) in four PCR cycles.

[0400] 500 ng of adapter-linked, PCR-enriched gDNA fragments were hybridized to Agilent's SureSelect biotinylated mouse whole exome RNA library bait at 65°C for 24 hours. The hybridized gDNA / RNA bait complex was extracted using streptavidin-coated magnetic beads. The gDNA / RNA bait complex was washed, and the RNA bait was cleaved during elution in SureSelect elution buffer to retain the captured adapter-linked, PCR-enriched gDNA fragments. The captured gDNA fragments were PCR-amplified using Herculase II DNA polymerase (Agilent) and SureSelect GA PCR primers for 10 cycles.

[0401] Purification was performed using 1.8x volume AMPure XP magnetic beads (Agencourt). For quality control, we used Invitrogen's Qubit HS assay, and fragment sizes were determined using Agilent's 2100 Bioanalyzer HS DNA assay.

[0402] Exome-enriched gDNA libraries were clustered using the Truseq SR cluster kit v2.5 with cBot at 7 pM, and then sequenced using the Truseq SBS kit on Illumina HiSeq2000.

[0403] Exome data analysis Sequence reads were aligned to the reference mouse genome assembly mm9 [Mouse Genome Sequencing Consortium, Nature 420, 520-562 (2002)] using bwa (version 0.5.8c) [Li, H. Durbin, R., Bioinformatics 25, 1754-1760 (2009)] with default options. Ambiguous reads—reads that map to multiple locations in the genome, such as those provided by the bwa output—were removed. The remaining alignments were screened, indexed, converted to Binary Compression Format (BAM), and read quality scores were converted from Illumina standard phred+64 to standard Sanger quality scores using a shell script.

[0404] For each sequencing lane, mutations were identified using three software programs: SAMtools (version 0.1.8) [Li, H. et al., Bioinformatics 25, 2078-2079 (2009)], GATK (version 1.0.4418) [DePristo, MA et al., Nature Genetics 43, 491-498 (2011)], and SomaticSniper [Ding, L. et al., Hum.Mol.Genet (2010), first published online September 15, 2010]. For SAMtools, the author's recommended options and filtering criteria, including a first filtering and a maximum coverage of 200, were used (http: / / sourceforge.net / apps / mediawiki / SAMtools / index.php?title=SAM_FAQ; accessed September 2011). For the second filtering in SAMtools, the minimum indel quality score was 50, and the minimum point mutation quality was 30. For GATK mutation calling, we followed the best practice guidelines designed by the creators as presented in the GATK user manual (http: / / www.broadinstitute.org / gsa / wiki / index.php / The_Genome_Analysis_Toolkit; accessed October 2010). For each sample, we performed local realignment near the indel region, followed by base quality recalibration. The UnifiedGenotyper module was applied to the resulting alignment data files. Where necessary, known polymorphisms of dbSNP [Sherry, ST et al., Nucleic Acids Res. 29, 308-311 (2009)] (version 128 for mm9) were supplied to individual steps. The variant score recalibration step was omitted and replaced with the hard filtering option. For SomaticSniper mutation calling, we used the default option and further examined only predicted mutations with a “somatic cell score” of 30 or higher.In addition, for each potential mutant locus, non-zero coverage in normal tissue was required, and all mutations located within repeat sequences defined by the RepeatMasker track of the UCSC Genome Browser for mouse genome assembly mm9 were removed [Fujita, PA et al., Nucleic Acids Res. 39, 876-882 (2011)].

[0405] RNA-Seq Barcoded mRNA-seq cDNA libraries were prepared from 5 μg of total RNA using a modified version of the Illumina mRNA-seq protocol. mRNA was isolated using SeramagOligo(dT) magnetic beads (Thermo Scientific). The isolated mRNA was fragmented using divalent cations and heat to produce fragments ranging from 160 to 200 bp. The fragmented mRNA was converted to cDNA using random primers and SuperScript II (Invitrogen), and the second strand was then synthesized using DNA polymerase I and RNase H. The cDNA was repaired at the ends using T4 DNA polymerase and Krenow DNA polymerase, and 5' phosphorylated using T4 polynucleotide kinase. The blunt-end cDNA fragments were 3' adenylated (3'-5' exo-minus) using Krenow fragments. A single 3' T overhang Illumina multiplex-specific adapter was ligated to the cDNA fragments using T4 DNA ligase. The cDNA library was purified and size-selected at 300 bp using an E-Gel 2% SizeSelect gel (Invitrogen). Concentration, Illumina 6-nucleotide index sequences, and flow cell-specific sequences were added by PCR using Phusion DNA polymerase (Finnzymes). All cleansing was performed using 1.8x volume Agencourt AMPure XP magnetic beads.

[0406] Barcoded RNA-Seq libraries were clustered using cBot with the Truseq SR cluster kit v2.5 at 7 pM, and then sequenced using the Truseq SBS kit on an Illumina HiSeq2000.

[0407] Raw output data from HiSeq was processed according to the Illumina standard protocol, which included removal of low-quality reads and demultiplexing. Sequence reads were then aligned to a reference genome sequence using bowtie [Langmead, B. et al., Genome Biol. 10, R25 (2009)] [Mouse Genome Sequencing Consortium, Nature 420, 520-562 (2002)]. Alignment coordinates were compared to the exon coordinates of RefSeq transcripts [Pruitt, KDet al., Nucleic Acids Res. 33, 501-504 (2005)], and the overlap alignment count was recorded for each transcript. Sequence reads that were not aligned to the genome sequence were aligned to a database of all possible exon-exon junction sequences of RefSeq transcripts [Pruitt, KDet al., Nucleic Acids Res. 33, 501-504 (2005)]. Alignment coordinates were compared with RefSeq exon coordinates and junction coordinates, reads were counted, and normalized for each transcript to RPKM (number of reads which map per nucleotide kilobase of transcript per million mapped reads) [Mortazavi, A. et al., Nat. Methods 5, 621-628 (2008)].

[0408] SNV Verification We selected SNVs for Sanger sequencing and RNA validation. We identified SNVs that were predicted by all three programs, non-synonymous, and found in transcripts with a minimum of 10 RPKM. Of these, we selected 50 with the highest SNP quality scores provided by the programs. As negative controls, we selected 44 SNVs with an FDR of ≥50%, present in only one cell line sample, and predicted by only one mutation calling program. Using DNA, we validated the selected mutations by PCR amplification of the region using 50 ng of DNA, followed by Sanger sequencing (Eurofins MWG Operon, Ebersberg, Germany). The reaction was successful for 50 loci and 32 loci in the positive and negative controls, respectively. Validation was also performed by examining tumor RNA-Seq reads.

[0409] FDR calculation and machine learning Random Forest Quality Score Calculation: Commonly used mutation calling algorithms (DePristo, MA et al., Nature Genetics 43, 491-498 (2011), Li, H. et al., Bioinformatics 25, 2078-2079 (2009), Ding, L. et al., Hum.Mol.Genet (2010) first published online September 15, 2010) output multiple scores, all of which potentially influence the quality of mutation calling. These include, but are not limited to, the base quality of interest assigned by the instrument, the quality alignment for this location, the number of reads covering this location, or the score for the difference between two genomes compared at this location. Ranking of mutations is necessary for calculating the false discovery rate, but this cannot be done directly for all mutations because conflicting information can be obtained from the various quality scores.

[0410] We employ the following strategy to achieve complete ranking. In the first step, we apply a very strict definition of significance by assuming that a mutation has better quality than another mutation only if and only if it outperforms another mutation in all categories. Thus, the quality trait S=(s1,...,s n The set of ) is s for all i=1,...,n i >t i If and only if, then T=(t1,...,t n ) is preferable, and is expressed as S > T. We define the intermediate FDR (IFDR) as follows:

number

[0411] However, in many closely related cases, comparisons are not feasible and therefore do not yield benefits from the vast amount of available data, so we simply consider IFDR as an intermediate step. Therefore, we leverage the good generalization properties of random forest regression [Breiman, L., Statist. Sci. 16, 199-231 (2001)] and train a random forest implemented in R (R Development Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria, 2010, Liaw, A., Wiener, M., R News 2, 18-22 (2002)).

[0412] For m mutations, each having n quality characteristics, the value range for each characteristic was determined, and up to p values ​​were sampled from this range at uniform intervals; if the set of values ​​for a quality characteristic was less than p, this set was used instead of the sample set. Next, in the n-dimensional quality space, "p nCreate each possible combination of sampled or selected quality values ​​that yields the maximum value of the data points. A random sample of 1% of these points and their corresponding IFDR values ​​were used as predictors and responses, respectively, for random forest training.

[0413] The resulting regression score is our generalized quality score Q: which can be viewed as a locally weighted combination of individual quality scores. This allows for a direct single-value comparison of any two mutations and the calculation of the actual false-find rate:

number

[0414] To train the random forest model used to produce the results of this study, we calculate the sample IFDR for somatic mutations in all samples before selecting a random 1% subset. This ensures that the entire available quality space is mapped to FDR values. We used the following quality characteristics, respectively: "SNP Quality", "Coverage Depth", "Consensus Quality", and "RMS Mapping Quality" (SAMtools, p=20); "SNP Quality", "Coverage Depth", "Variant Confidence / Unfiltered Depth", and "RMS Mapping Quality" (GATK, p=20); or "SNP Quality", "Coverage Depth", "Consensus Quality", "RMS Mapping Quality", and "Somatic Score" (SomaticSNiPer, p=12). The varying values ​​of p ensure a comparable set size.

[0415] Common Coverage Calculation: The number of possible mutation calls can introduce significant bias into the definition of the false detection rate. The number of mutations called is comparable and can be used as a basis for false detection rate calculations only if the number of possible locations where mutations occur is the same for our tumor comparison and peer-to-peer comparison. To correct for this potential bias, we use a common coverage ratio. As common coverage, we define a base number relating to at least one coverage in both samples used for mutation calling. Common coverage is calculated separately for tumor comparison and peer-to-peer comparison.

[0416] ROC estimation Receiver Operational Characteristic (ROC) curves and their corresponding Area Under the Curve (AUC) are useful for constructing classifiers and visualizing their performance [Fawcett, T., Pattern Recogn. Lett. 27, 861-874 (2006)]. We extend this concept to evaluate the performance of experimental and computational procedures. However, plotting ROC graphs requires knowledge of all true positive and false positive (TP and FP) cases in a dataset, information that is usually not given and is difficult to establish for high-throughput data (such as NGS data). Therefore, we plot the ROC graph and calculate the AUC using the FDR calculated to estimate the respective TP and FP rates. The central concept is that the FDR of a single mutation in the dataset gives the proportion of how much this mutation contributes to the sum of TP / FP mutations, respectively. Also, for a list of random assignments to TP and FP, the resulting ROC AUC is equal to 0.5 in our method, indicating a completely random prediction.

[0417] We have two conditions:

number

number

[0418] To obtain the estimated ROC curve, mutations in the dataset are selected using FDR, and for each mutation, a point is plotted on the cumulative TPR and FPR values ​​up to that mutation, calculated by dividing each TPR by the sum of all TPR values. The AUC is calculated by summing the areas of all consecutive trapezoids between the curve and the x-axis.

[0419] (Example 10) Selection of tumor antigen combinations as targets for cancer treatment In this example, we evaluated whether it is possible to establish a set of tumor antigens that can be shared at least partially by a large proportion of tumor patients and provide a set of vaccine products applicable to a broad spectrum of cancer patients.

[0420] For this purpose, RNA was extracted from melanoma metastasis samples using the RNeasy Lipid Tissue Mini Kit (Qiagen). cDNA synthesis was performed using the SuperScript II Reverse Transcriptase Kit (Invitrogen) and oligo dT. Expression was analyzed using the BioMark® HD System (Fluidigm), and relative expression was calculated using HPRT as a housekeeping gene.

[0421] In this way, we were able to detect the relative expression of several genes, including DCT (=TRP2), TYR, and TPTE, in melanoma samples. Furthermore, we were able to confirm that the combination of just three tumor antigens, namely DCT (isoform 1), TYR, and TPTE, was sufficient to represent 88% of the patient samples analyzed (Figure 22).

Claims

1. Personalized vaccines for use in methods for preventing or treating cancer in patients, This personalized vaccine is The RNA comprises an RNA encoding a polyepitope polypeptide containing a neoepitope based on the mutation of the patient, The above method involves the following steps: (i) A step of administering a first vaccine to induce a first immune response, The first vaccine for inducing the aforementioned first immune response is (a) Peptides or polypeptides containing a combination of common tumor antigens, (b) Peptides or polypeptides containing a T cell epitope of a common tumor antigen combination, (c) a nucleic acid encoding the peptide or polypeptide of (a) or (b), The common tumor antigen is expressed in the cancer cells of the patient, and the common tumor antigen is a tumor antigen shared by at least 60% of various patients having the same and / or different cancer types. The first immune response is induced in the patient against the tumor antigen, and (ii) A step of administering the personalized vaccine to induce a secondary immune response, wherein the secondary immune response is induced against a neoepitope based on the mutation in the patient, and the secondary immune response is specific to the somatic mutation present in the patient's cancer cells, Personalized vaccines.

2. A personalized vaccine for use according to claim 1, The first and / or second immune response is a cellular response, and / or The aforementioned first immune response includes a CD8+ T cell response, and / or The administration of the personalized vaccine may provide MHC class II presenting neoepitopes. The second immune response involves a personalized vaccine, including a CD4+ T cell response.

3. A personalized vaccine for use according to claim 1 or 2, The aforementioned first immune response is not specific to cancer-specific somatic mutations present in the patient's cancer cells, and / or The aforementioned tumor antigen is common in the cancer being treated, and / or The aforementioned tumor antigen is a personalized vaccine, common in various cancers.

4. A personalized vaccine for use according to any one of claims 1 to 3, The patient is positive for one or more tumor antigens, and is receiving a personalized vaccine.

5. A personalized vaccine for use according to claim 3, The aforementioned cancer-specific somatic mutations are present in the exomes of the patient's cancer cells, and / or The aforementioned cancer-specific somatic mutations are non-synonymous mutations, resulting in a personalized vaccine.

6. A personalized vaccine for use according to any one of claims 1 to 5, Each of the polyepitope polypeptides containing mutation-based neoepitopes contains up to 30 mutation-based neoepitopes. Each of the polyepitope polypeptides further comprises an epitope that does not contain cancer-specific somatic mutations expressed by the cancer cells. Personalized vaccine.

7. A personalized vaccine for use according to claim 6, A personalized vaccine in which the neoepitope based on the mutation and the epitope exist in their natural sequence context to form a vaccine sequence, the vaccine sequence being 30 amino acids long.

8. A personalized vaccine for use according to claim 6 or 7, The neoepitope based on the mutation, the epitope, and / or vaccine sequence are aligned from head to tail and / or separated by a linker. The linker is an amino acid sequence (GGS) ａ (GSS) ｂ (GGG) ｃ (SSG) ｄ (GSG) ｅ Personalized vaccines, including a, b, c, d, and e, where a, b, c, d, and e are independently selected numbers from 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20, and a + b + c + d + e is not 0.

9. A personalized vaccine for use according to any one of claims 1 to 8, The aforementioned first immune response is a personalized vaccine induced by the administration of an RNA vaccine.

10. A personalized vaccine for use according to any one of claims 1 to 9, Personalized vaccines in which cytidine is partially or completely substituted with 5-methylcytidine, or uridine is partially or completely substituted with pseudouridine.

11. A personalized vaccine for use according to any one of claims 1 to 10, The 5' end of the RNA vaccine includes a cap structure having the following general formula: 【Chemistry 1】 In the formula, R １ and R ２ are independently hydroxy or methoxy, and W － , X － and Y － are independently oxygen, sulfur, selenium or BH ３ A personalized vaccine that is

12. A personalized vaccine for use according to any one of claims 1 to 11, The RNA is mRNA containing 5'-UTR and 3'-UTR, in this personalized vaccine.

13. A personalized vaccine for use according to any one of claims 1 to 12, The RNA is a personalized vaccine comprising a poly(A) tail having a length of 100 to 150 adenosine residues.

14. A personalized vaccine for use according to claim 12 or 13, The aforementioned 3'-UTR is a personalized vaccine containing two copies of 3'-UTR derived from the globin gene.

15. A personalized vaccine for use according to any one of claims 1 to 14, A personalized vaccine in which the RNA is conjugated to a carrier containing a lipid-containing carrier, a cationic lipid, a liposome, a micelle, or nanoparticles.

Images