Individualized 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 treatments that target distinct expression patterns of tumor antigens, particularly common tumor antigens, and distinct tumor mutations. [Background technology]

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

[0003] Antigen-specific immunotherapy aims to enhance or induce specific immune responses 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 is mediated by T cell receptors (TCRs) expressed on the surface of T cells. The T cell receptors (TCRs) of T cells can bind to major histocompatibility complex (MHC) molecules and interact with immunogenic peptides (epitopes) displayed on the surface of target cells. Specific binding of the TCR initiates a signal cascade within the T cell, leading to proliferation and differentiation into mature effector T cells.

[0004] The identification of an increasing number of pathogen-associated and tumor-associated antigens (TAA) has provided a wide 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 can tend to induce and expand antigen-specific T cells in patients that can specifically recognize and kill pathological cells. Various antigen formats can be used for tumor vaccination, including whole cancer cells, proteins, peptides or immune 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 result 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 100-120 nonsynonymous mutations, many of which are targetable by vaccines. More than 95% of mutations in tumors are unique and patient-specific (Weide et al. 2008: J. Immunother. 31, 180-188). The number of protein-altering somatic mutations that can give rise to tumor-specific T cell epitopes is in the range of 30-400. It has been computationally predicted that there are 40-60 HLA class I-restricted epitopes derived from tumor-specific somatic mutations per patient (Azuma et al. 1993: Nature 366, 76-79). Furthermore, new immunogenic HLA class II-restricted epitopes are likely to result from tumor-associated mutations as well, but the number of these is still unknown.

[0006] In particular, some nonsynonymous mutations are causally involved in malignant transformation and are essential to maintain the oncogenic phenotype (driver mutations) and may be potential "Achilles heel" of cancer cells. Mutations found in primary tumors may also be present in metastases. However, several studies have demonstrated that patients' metastatic tumors often acquire additional clinically meaningful 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). Moreover, the molecular characteristics of many metastases also deviate significantly from those of the primary tumor. [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 Summary of the Invention [Problem to be solved by the invention]

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

[0009] The present invention relates to the concept of personalized therapy, which integrates individual disease genetics and creates customized treatments in oncology. Human cancers express a variety of immunogenic common tumor antigens and carry dozens to hundreds of nonsynonymous mutations, many of which can be targeted by T cells. These mutations are ideal candidates for vaccine development, as they are not subject to central immune tolerance. The present invention uses personalized vaccines, particularly RNA vaccines, that target individual expression patterns of tumor antigens and individual tumor mutations. The concept of the present invention is to utilize genetic tumor changes for the benefit of patients, rather than being hindered by them.

[0010] Instead of searching for a common molecular denominator in many patients for targeting, the present invention utilizes the antigen target repertoire of each individual patient. Rather than accepting the trade-off by providing a treatment designed for the average, the present invention provides the best combination for each individual patient. This is not just a paradigm shift, but opens new possibilities to solve key problems in current cancer drug development, such as wide inter-individual variability and intra-tumor clonal heterogeneity. The present invention allows optimal utilization of the antigen repertoire in the patient.

[0011] Specifically, this application relates to multi-specific targeting of the entire individual tumor antigen repertoire found in each individual cancer patient, including both non-mutated and mutated tumor antigens. To target non-mutated tumor antigens, a tumor antigen targeting portfolio (warehouse) can be used that covers a large proportion of patients. This warehouse is a drug repository for "stock" pre-manufactured vaccines and for combining them with vaccine cocktails for use in individual patients. Combined with vaccines based on mutanomic analysis, this allows optimal utilization of antigen repertoire in patients.

[0012] The present invention involves the identification of patient-specific cancer mutations and targeting the patient's individual cancer mutation "signatures". Identification of nonsynonymous point mutations resulting in amino acid changes that are presented on 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 a set of mutations from cancer cells, such as circulating tumor cells (CTCs), allows the provision of a vaccine that induces an immune response that targets genetically distinct subpopulations as well as potentially the primary tumor, even if it contains tumor metastases. For vaccination, such neoepitopes identified according to the present application are preferably provided to the patient in the form of a polypeptide that contains said neoepitopes, and after appropriate processing and presentation by MHC molecules, the neoepitopes are displayed 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, that contains one or more immunogenic epitopes against which an immune response should be induced. Such an immunogenic gene product may contain the entire tumor antigen against which an immune response should be induced, or may contain a portion thereof, such as a T cell epitope. The strategy of injecting in vitro transcribed RNA (IVT-RNA) directly into a patient by various immunization routes has been tested in various animal models with success. The RNA can be translated in transfected cells, and the expression product can be presented to MHC molecules on the surface of the cell after processing to induce an immune response.

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

[0015] The present invention relates to a method for inducing an efficient and specific immune response in cancer patients by administering a cancer vaccine that targets the individual expression pattern of tumor antigens, such as common tumor antigens, and by administering a cancer vaccine that targets individual tumor mutations.Preferably, the cancer vaccine administered to a patient according to the present invention provides MHC-presented epitopes specific to the patient's tumor, suitable for stimulating, priming and / or expanding T cells against cells expressing antigens (common tumor antigens and antigens with patient-specific mutations) from which the epitopes presented by MHC are derived, specific to the patient's tumor.Therefore, the vaccine described herein can induce or promote a cellular response, preferably cytotoxic T cell activity, against cancer diseases characterized by the presentation of one or more cancer-expressed antigens by class I MHC.The vaccine administered according to the present invention is also specific to the patient's tumor, since it targets cancer-specific mutations.

[0016] In one aspect, the invention provides a method for preventing or treating cancer in a patient, comprising: (i) inducing a first immune response in a patient against one or more tumor antigens; and (ii) inducing in the patient a second immune response against one or more tumor antigens, wherein said second immune response is specific to a cancer-specific somatic mutation present in the patient's cancer cells.

[0017] The steps (i) of inducing a first immune response and (ii) 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 simultaneously with the administration of 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 administration of the vaccine for inducing the second immune response.

[0018] In one embodiment, the primary and / or secondary immune response is induced by administering one or more suitable vaccines, in particular 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 comprises a CD8+ T cell response. In one embodiment, the second immune response comprises 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 against a collection of tumor antigens expressed in the patient's cancer cells. In one embodiment, the first immune response is not specific to a cancer-specific somatic mutation present in the patient's cancer cells. In one embodiment, the first immune response is induced against one or more tumor antigens that are common tumor antigens. In one embodiment, the vaccine for inducing the first immune response induces an immune response specific to a tumor antigen expressed in the majority of cancer cells of a cancer patient, said cancer patient having the same type of cancer, such as the cancer treated according to the present invention, or a different type of cancer. According to the present invention, the first immune response can be induced by providing the patient with a collection of tumor antigens, such as a common tumor antigen or epitopes thereof, said epitopes can be provided in the form of a polyepitope polypeptide (also referred to herein as a multivalent polypeptide) containing these epitopes. Said antigens or epitopes are preferably provided to the patient by administering a nucleic acid, in particular an RNA, encoding said antigens or epitopes. After appropriate processing and presentation by MHC molecules, the epitopes are displayed to the patient's immune system for stimulation of appropriate T cells.

[0021] The epitopes may be present in the polyepitopic polypeptide in the form of vaccine sequences, i.e. in their natural sequence context, for example adjacent to amino acid sequences that flank said epitopes in naturally occurring proteins. Such flanking sequences may each comprise 5 or more, 10 or more, 15 or more, 20 or more, and preferably up to 50, 45, 40, 35 or 30 amino acids, and may flank the epitope sequence at the N-terminus and / or C-terminus. Thus, the vaccine sequence may comprise 20 or more, 25 or more, 30 or more, 35 or more, 40 or more, and preferably up to 50, 45, 40, 35 or 30 amino acids. In one embodiment, the epitopes and / or vaccine sequences are arranged in a head-to-tail direction in the polypeptide. In one embodiment, the epitopes and / or vaccine sequences are separated by a linker. Such linkers are further described below. When a pre-manufactured polyepitope polypeptide is to be used to induce a first immune response, preferably administered without determining the patient's individual tumor antigen expression pattern, the polyepitope polypeptide preferably contains epitopes that are 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 low as possible. To this end, the examples demonstrate that a specific set of only three different tumor antigens is sufficient to cover 88% of the melanoma metastasis patient samples analyzed. In other words, 88% of melanoma metastasis patients express at least one antigen from said specific set of only three different tumor antigens. Therefore, it is expected that the inclusion of at least one epitope from each of said three different tumor antigens in the polyepitope polypeptide will induce a first immune response in 88% of melanoma metastasis patients.It should be understood that these collections of antigens are not necessarily complementary to cover the maximum number of tumor patients while keeping the number of epitopes in the polyepitope polypeptide as small as possible, so that the polyepitope polypeptide does not necessarily contain epitopes from antigens shared in the maximum percentage of tumor patients. Rather, such a set of epitopes is optimized for (i) antigens shared by the maximum percentage of tumor patients and (ii) antigens covering the maximum number of tumor patients while keeping the number of epitopes in the polyepitope polypeptide as small as possible.

[0022] In one embodiment, the vaccine product for inducing a first immune response contains (i) one or more peptides or polypeptides, each peptide or polypeptide comprising one or more tumor antigens, (ii) one or more peptides or polypeptides, each peptide or polypeptide comprising one or more T cell epitopes of one or more tumor antigens, or (iii) a nucleic acid, preferably an RNA, encoding one or more peptides or polypeptides comprised in (i) or (ii). In one embodiment, the polypeptide used for immunization comprises up to 30 epitopes. In one embodiment, the epitopes are present in their natural sequence context 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 sequences are arranged in a head to tail direction. In one embodiment, the epitopes and / or vaccine sequences are separated by a linker.

[0023] In one embodiment, the vaccine product administered to induce the first immune response induces an immune response against a tumor antigen that is common in the cancer to be treated and / or common in various cancers. Preferably, the tumor antigen involved in the induction of the first immune response is a common tumor antigen. In one embodiment, the patient is positive for one or more tumor antigens against which the first immune response is induced. In one embodiment, the patient is positive for all tumor antigens against which the first immune response is induced. According to the present invention, the term "patient is positive for tumor antigen" means that the patient's cancer cells express the tumor antigen.

[0024] In one embodiment, the first immune response is induced by administering one or more vaccine products selected from a set comprising pre-manufactured vaccine products, in particular RNA encoding a peptide or polypeptide comprising a tumor antigen or an immunogenic fragment thereof, such as a T-cell epitope, each pre-manufactured vaccine product inducing an immune response against a tumor antigen. In one embodiment, the set comprises vaccine products inducing immune responses against a variety of tumor antigens, such as at least 3 tumor antigens, at least 5 tumor antigens, at least 8 tumor antigens, at least 10 tumor antigens, at least 15 tumor antigens, at least 20 tumor antigens, at least 25 tumor antigens, at least 30 tumor antigens, or even more.

[0025] In one embodiment, the cancer-specific somatic mutation is present in the exome of the patient's cancer cells. In one embodiment, the cancer-specific somatic mutation is a non-synonymous mutation. In one embodiment, the cancer cells are circulating tumor cells. In one embodiment, the second immune response is induced by administering a vaccine containing a polypeptide comprising a mutation-based neo-epitope, or a nucleic acid, preferably an RNA, encoding said polypeptide. In one embodiment, the polypeptide comprises up to 30 mutation-based neo-epitopes. In one embodiment, the polypeptide further comprises an epitope that does not comprise a cancer-specific somatic mutation expressed by the cancer cells. In one embodiment, the epitopes are present in their natural sequence context to form a vaccine sequence. In one embodiment, the vaccine sequence is about 30 amino acids long. In one embodiment, the neo-epitopes, epitopes and / or vaccine sequences are arranged in a head-to-tail direction. In one embodiment, the neo-epitopes, epitopes and / or vaccine sequences are separated by a linker.

[0026] According to the present invention, a vaccine for inducing a primary immune response comprises: (a) identifying tumor antigens, particularly common tumor antigens, expressed in tumor specimens from cancer patients; and (b) preferably, by selecting a vaccine product from a set comprising pre-manufactured vaccine products, each pre-manufactured vaccine product preferably inducing an immune response against a common tumor antigen, thereby providing a vaccine characterized by the tumor antigen profile, in particular the common tumor antigen profile, obtained in step (a).

[0027] According to the present invention, the term "common tumor antigen" relates to a tumor antigen that is expressed by a majority of cancers, e.g. a majority of cancers of the same type, e.g. a cancer type treated according to the present invention, and / or by a majority of cancers of different types. Thus, the term "common tumor antigen" relates to a tumor antigen that is shared by a majority of different patients with the same cancer type and / or different cancer types. Preferably, such tumor antigens are tumor antigens that carry 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%, in particular 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 are applicable to a majority of cancer patients. According to the present invention, the term "same type of cancer" relates 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 types of cancer" relates to cancers of different medical classifications, such as cancers of different organs or tissues.

[0028] According to the present invention, "inducing an immune response against a tumor antigen" preferably refers to the ability to induce an immune response, preferably a T-cell response, against a tumor antigen, such as a common tumor antigen, or against a cell, such as a cancer cell, expressing and / or presenting a tumor antigen, such as a common tumor antigen, when administered to a patient. Thus, a vaccine for inducing an immune response against a common tumor antigen may contain (i) a peptide or polypeptide comprising a common tumor antigen, or (ii) a peptide or polypeptide comprising one or more T-cell epitopes of a common tumor antigen. In one particularly preferred embodiment, a peptide or polypeptide comprising a common tumor antigen according to the present invention, or (ii) a peptide or polypeptide comprising one or more T-cell epitopes of a common tumor antigen according to the present invention, is administered to a patient in the form of a nucleic acid, preferably an RNA, such as an in vitro transcribed RNA or a synthetic RNA, that can be expressed in a patient's cells, such as an antigen-presenting cell, to generate said peptide or polypeptide.

[0029] According to the present invention, the vaccine for inducing the first immune response is preferably selected from a pre-supplied vaccine warehouse, such as a pre-supplied RNA vaccine warehouse. This approach is also referred to herein as "stock". Such a pre-supplied vaccine warehouse relates to a set comprising pre-manufactured vaccine products, each of which induces an immune response against 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 a majority of cancer patients with the same type of cancer and / or a majority of cancer patients with different types of cancer. Thus, the vaccine warehouse used according to the present invention preferably comprises a set of vaccine products applicable to a majority of cancer patients. For example, if a set of common tumor antigens is known for a particular cancer type, it is possible to create such a pre-supplied vaccine warehouse comprising a set of vaccine products, in which the vaccine products in the set induce an immune response against said common tumor antigen. Since such a vaccine repository is selected to be applicable to a large proportion of patients, it is possible to select one or more vaccine products from said pre-supplied vaccine repository, which induce an immune response against one or more tumor antigens expressed in the cancer cells of the particular patient to be treated, and thus target the tumor antigen profile of each patient without the need to provide additional vaccine products specifically designed for the patient to be treated. Such selection can be performed by testing the patient for tumor antigen expression, and then selecting the appropriate vaccine product from the pre-supplied vaccine repository that targets the tumor antigen expressed by the patient's cancer cells. Such selection can be performed based on the transcriptome / peptideome analysis of the patient's tumor. For example, tumor samples from eligible patients can be analyzed for tumor antigen signatures. The common tumor antigen profile can be determined by quantitative multiplex RT-PCR and IHC, and the respective vaccine product can be selected from the repository.Selecting one or more vaccine products from the pre-supplied vaccine archive that induce an immune response against one or more tumor antigens expressed in the cancer cells of the particular patient being treated, and thus target the tumor antigen profile of each patient, can also be performed by randomly selecting a vaccine product from the pre-supplied vaccine archive that is 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, a set of pre-manufactured vaccine products is optimized with respect to the coverage of tumor samples and their tumor antigen expression patterns. In particular, said set comprises 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 small as possible. For this purpose, it is demonstrated in the examples that a specific set of only three different tumor antigens is sufficient to cover 88% of the melanoma metastasis patient samples analyzed. In other words, 88% of melanoma metastasis patients express at least one antigen from said specific set of only three different tumor antigens. It should be understood that such a set does not necessarily include antigens shared in the maximum proportion of tumor patients, since the collection of these antigens does not necessarily have to be complementary to cover the maximum number of tumor patients while keeping the number of vaccine products in the warehouse as small as possible. Rather, such a set of vaccine products is preferably optimized with respect to (i) antigens shared by the maximum proportion of tumor patients and (ii) antigens covering the maximum number of tumor patients while keeping the number of vaccine products in the warehouse as small as possible.

[0031] In one embodiment, the vaccine repository of a tumor antigen, preferably a common tumor antigen, is suitable for targeting at least 60%, more preferably at least 70%, more preferably at least 80%, more preferably at least 90%, especially at least 95% of patients with 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 a patient, or a portion of tumor antigens present in one or more cancer cells of a patient. Preferably, such a tumor antigen profile or collection of tumor antigens includes 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. Thus, the present invention may include the identification of all common tumor antigens present in one or more cancer cells of a patient, or may include the identification of only a portion of common tumor antigens present in one or more cancer cells of a patient. In general, a number of common tumor antigens can be identified in a cancer patient's tumor specimen that provides a sufficient number of common tumor antigens to be targeted by a vaccine.

[0033] The vaccine for inducing a first immune response, when administered to a patient, preferably provides a collection of MHC-presented epitopes from a collection of tumor antigens, such as common tumor antigens, 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 gives rise to T cells that target the epitopes when bound to MHC, and thus target the patient's tumor, preferably the primary tumor as well as tumor metastases, which express the antigen from which the MHC-presented epitope is derived and present the same epitope on the surface of the tumor cells.

[0034] According to the present invention, a vaccine for inducing a secondary immune response comprises: (a) identifying cancer-specific somatic mutations in a tumor specimen of a cancer patient to provide a cancer mutation signature for the patient; and (b) providing a vaccine characterized by the cancer mutation signature obtained in step (a).

[0035] In one embodiment, the method of the invention comprises: i) providing a tumor specimen from a cancer patient and, preferably, a non-tumorigenic specimen derived from a cancer patient; ii) identifying sequence differences between the genome, exome and / or transcriptome of the tumor specimen and the genome, exome and / or transcriptome of the non-tumorigenic specimen; iii) designing a peptide or polypeptide comprising an epitope incorporating the sequence differences determined in step (ii); iv) providing a peptide or polypeptide designed in step (iii) or a nucleic acid, preferably an RNA, encoding said peptide or polypeptide; and v) providing a vaccine containing the peptide or polypeptide or nucleic acid provided in step (iv).

[0036] According to the present invention, tumor specimen refers to any sample, such as a body sample derived from a patient that contains or is expected to contain tumor or cancer cells. The body sample can be any tissue sample, such as blood, tissue sample obtained from primary tumor or from tumor metastasis, or any other sample that contains tumor or cancer cells. Preferably, the body sample is blood, and the cancer-specific somatic mutation or sequence difference is determined in one or more circulating tumor cells (CTCs) contained in the blood. In another embodiment, tumor specimen refers to one or more isolated tumor or cancer cells, such as circulating tumor cells (CTCs), or a sample that contains one or more isolated tumor or cancer cells, such as circulating tumor cells (CTCs).

[0037] A non-tumorigenic specimen relates to any sample, such as a body sample, derived from a patient or another individual, preferably of the same species as the patient, preferably a healthy individual that does not contain or is expected to not contain tumor or cancer cells. The body sample can be any tissue sample, such as a sample from blood or non-tumorigenic tissue.

[0038] According to the present invention, the term "cancer mutation signature" may refer to all cancer mutations present in one or more cancer cells of a patient, or may refer to only a portion of cancer mutations present in one or more cancer cells of a patient. Thus, the present invention may include the identification of all cancer-specific mutations present in one or more cancer cells of a patient, or may include the identification of only a portion of cancer-specific mutations present in one or more cancer cells of a patient. In general, the present invention provides the identification of a number of mutations that provide a sufficient number of neoepitopes for inclusion in a vaccine. "Cancer mutation" refers to sequence differences between nucleic acids contained in cancer cells and nucleic acids contained in normal cells.

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

[0040] In one embodiment, cancer-specific somatic mutation or sequence difference is determined in the genome, preferably the whole genome, of tumor specimen.Therefore, the present invention can comprise identifying the cancer mutation signature of the genome, preferably the whole genome, of one or more cancer cells.In one embodiment, identifying the cancer-specific somatic mutation in the tumor specimen of cancer patient comprises identifying the whole genome cancer mutation profile.

[0041] In one embodiment, cancer-specific somatic mutations or sequence differences are determined in the exome, preferably the whole exome, of tumor specimens. Exomes are the part of an organism's genome that is formed by exons, which are the coding parts of expressed genes. Exomes provide 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 the phenotype of an organism. It is estimated that the exome of the human genome accounts for 1.5% of the whole genome (Ng, PC et al., PLoS Gen., 4(8):1-15, 2008). Thus, the present invention can include identifying the cancer mutation signature of the exome, preferably the whole exome, of one or more cancer cells. In one embodiment, identifying cancer-specific somatic mutations in tumor specimens of cancer patients includes identifying a whole exome cancer mutation profile.

[0042] In one embodiment, cancer-specific somatic mutations or sequence differences are determined in the transcriptome, preferably the whole transcriptome, of tumor specimens.Transcriptome is the set of all RNA molecules, including mRNA, rRNA, tRNA and other non-coding RNA produced in a cell or cell population.In the context of the present invention, transcriptome refers to the set of all RNA molecules produced in a cell, cell population, preferably cancer cell population, or all cells of a given individual at a particular time.Therefore, the present invention can include identifying the cancer mutation signature of the transcriptome, preferably the whole transcriptome, of one or more cancer cells.In one embodiment, identifying cancer-specific somatic mutations in tumor specimens of cancer patients includes identifying the whole transcriptome cancer mutation profile.

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

[0044] In one embodiment, identifying cancer-specific somatic mutations or identifying sequence variances comprises using next generation sequencing (NGS).

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

[0046] To reveal cancer-specific somatic mutations or sequence differences, sequence information obtained from a tumor specimen is preferably compared to a reference, such as sequence information obtained from sequencing nucleic acid, such as DNA or RNA, of 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] The vaccine for inducing the second immune response, when administered to a patient, preferably provides a collection of epitopes presented by MHC, such as 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, incorporating sequence changes based on the identified mutations or sequence differences. Such MHC-presented epitopes incorporating sequence changes based on the 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 results in T cells that target the epitopes when bound to MHC, and thus target the patient's tumor, preferably the primary tumor, as well as tumor metastases, which express the antigen from which the MHC-presented epitopes are derived and present the same epitopes on the surface of the tumor cells.

[0048] To provide a vaccine for inducing a second immune response, the present invention may include the arbitrary incorporation of 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 usefulness of the identified mutations in the epitopes for cancer vaccination. This further step may therefore include one or more of the following: (i) assessing whether the sequence change is located in a known or predicted MHC-presented epitope, (ii) testing in vitro and / or in silico whether the sequence change is located in an MHC-presented epitope, for example testing whether the sequence change is part of a peptide sequence that is processed into and / or presented as an MHC-presented epitope, and (iii) testing in vitro whether the putative mutated epitopes, especially when present in their natural sequence context, for example adjacent to the amino acid sequence adjacent to said epitope in naturally occurring proteins, and when expressed in antigen-presenting cells, are able to stimulate patient T cells with the desired specificity. Such flanking sequences may each comprise 3 or more, 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 up to 30 amino acids, and may flank the epitope sequence at the N-terminus and / or C-terminus.

[0049] The mutation or sequence difference determined according to the present invention can be ranked according to their usefulness as epitopes for cancer vaccination.Therefore, in one aspect, the present invention provides a manual or computer-based analysis process for analyzing and selecting the identified mutations according to their usefulness in each vaccine provided.In a preferred embodiment, said analysis process is a process based on computer algorithm.

[0050] Preferably, the analysis process comprises the following steps: - identifying expressed protein-altering mutations, for example by analyzing the transcript; - Identifying potentially immunogenic mutations, i.e. by comparing the obtained data with available data sets of confirmed immunogenic epitopes, for example those contained in public immune epitope databases, such as the IMMUNE EPITOPE DATABASE AND ANALYSIS RESOURCE at http: / / www.immunoepitope.org; One or more, preferably all, of the above.

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

[0052] In another embodiment, epitopes can be selected and / or ranked by using additional parameters such as protein impact, associated gene expression, sequence uniqueness, predicted presentation potential, and association with cancer genes.

[0053] Analysis of multiple CTCs also allows for the selection and prioritization of mutations, e.g., mutations observed in a larger percentage of CTCs can be prioritized higher than mutations observed in a smaller percentage of CTCs.

[0054] The collection of mutation-based neoepitopes identified and provided by the vaccine for inducing a second immune response is preferably present in the form of a polypeptide (polyepitope polypeptide) containing said neoepitopes or a nucleic acid, in particular an RNA, encoding said polypeptide. Furthermore, the neoepitopes may be present in the polypeptide in the form of a vaccine sequence, i.e. in their natural sequence context, flanked by amino acid sequences flanking said epitopes, for example also in naturally occurring proteins. Such flanking sequences may each comprise 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 up to 30 amino acids, and may flank the epitope sequence at the N-terminus and / or C-terminus. Thus, the vaccine sequence may comprise 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 up to 30 amino acids. In one embodiment, the neoepitopes and / or vaccine sequences are arranged in a head to tail orientation in the polypeptide.

[0055] In one embodiment, the neoepitopes and / or vaccine sequences are separated by a linker, in particular a neutral linker. The term "linker" according to the present invention relates to a peptide added between two peptide domains, such as epitopes or vaccine sequences, to link said peptide domains. There is no particular restriction regarding the linker sequence. However, it is preferred that the linker sequence reduces steric hindrance between the two peptide domains, translates well, and supports or allows the processing of the epitopes. Furthermore, the linker should have no or only few immunogenic sequence elements. The linker should preferably not generate non-endogenous neoepitopes, such as those resulting from junctional stitching between adjacent neoepitopes, which may give rise to undesired immune responses. Therefore, a polyepitope vaccine should preferably include a linker sequence that can reduce the number of undesired MHC-binding junctional 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 proteasomal processing, and thus the use of glycine-rich linker sequences serves to minimize the number of peptides included in the linker that can be processed by the proteasome. Furthermore, glycine was observed to inhibit strong binding at the MHC binding groove position (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 neo-epitopes. The linkers may each comprise 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 in 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 of the linker are glycine and / or serine. In one preferred embodiment, the linker consists essentially of the amino acids glycine and serine. In one embodiment, the linker is the amino acid sequence (GGS). a (GSS) b (GGG) c (SSG) d (GSG) e where a, b, c, d and e are independently numbers selected 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, and is preferably 2 or more, 3 or more, 4 or more, or 5 or more. In one embodiment, the linker comprises a sequence as 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 the vaccine for inducing a secondary immune response is preferably present in the form of a collection of peptides comprising said neoepitopes on different peptides, each of which comprises one or more neoepitopes, which may also overlap, or a collection of nucleic acids, in particular RNA, encoding said peptides.

[0057] The administration of the vaccine to induce the second immune response may provide an MHC class II-presented epitope that can induce a CD4+ helper T cell response against cells expressing the antigen from which the MHC-presented epitope is derived. Alternatively or additionally, the administration of the vaccine to induce the second immune response may provide an MHC class I-presented epitope that can induce a CD8+ T cell response against cells expressing the antigen from which the MHC-presented epitope is derived. Furthermore, the administration of the vaccine to induce the second 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 contain 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, the administration of the vaccine for inducing the second immune response provides a neo-epitope that is an MHC class II-presented epitope and / or can induce a CD4+ helper T cell response against cells expressing the antigen from which the MHC-presented epitope is derived, and a cancer-specific somatic mutation-free epitope that is an MHC class I-presented epitope and / or can induce a CD8+ T cell response against cells expressing the antigen from which the MHC-presented epitope is derived. In one embodiment, the cancer-specific somatic mutation-free epitope is derived from a tumor antigen. In one embodiment, the neo-epitope and the cancer-specific somatic mutation-free epitope have a synergistic effect in the treatment of cancer. Preferably, the vaccine for inducing the second immune response is useful for polyepitope stimulation of cytotoxic 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 invention, is administered to a patient in the form of a nucleic acid, preferably an RNA, such as an in vitro transcribed RNA or a synthetic RNA, that can be expressed in a patient's cells, such as an antigen-presenting cell, to generate the peptide or protein. The invention also envisages administering one or more multiepitope polypeptides, which for the purposes of the invention are encompassed by the term "polyepitope polypeptide", preferably in the form of a nucleic acid, preferably an RNA, such as an in vitro transcribed RNA or a synthetic RNA, that can be expressed in a patient's cells, such as an antigen-presenting cell, to generate one or more polypeptides. When administering two or more multiepitope polypeptides, the neoepitopes provided by the different multiepitope polypeptides may be different or may partially overlap. Once present in a patient's cells, such as an antigen-presenting cell, the peptide or protein is processed to generate immunogenic epitopes, such as neoepitopes.

[0059] Particularly preferred embodiments of the present invention include (i) in vitro transcribed polynucleotide RNA vaccine cocktails containing "off-the-shelf" RNA from a pre-supplied RNA repository targeting common tumor antigen profiles for each patient, and (ii) made-on-demand RNA vaccines encoding neoepitopes derived from patient-specific mutations.

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

[0061] In a further aspect, the present invention provides a vaccine as described herein for use in the methods of treatment as described herein, in particular 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 traditional chemotherapy.

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

[0064] [Figure 1] Figure 1 Top: Process for discovering and prioritizing potentially immunogenic somatic mutations in bulk tumor samples. Figure 1 Bottom: Process applied to the B16 and Black6 lines. [Diagram 2] Figure 2: Example of a documented mutation in Kif18b. A mutation 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, the sequence is a mixture of T / G. [Diagram 3] Figure 3: Immune reactivity against mutant sequences. Mice (n=5) were immunized twice (day 0, day 7) with the mutant peptide sequence (100 μg + PolyI:C 50 μg; sc). On day 12, mice were sacrificed and splenocytes were harvested. IFNγ ELISpot was performed using 5×105 splenocytes / well as effectors and 5×104 bone marrow dendritic cells as target cells, loaded with peptide (2 μg / ml for 2 h at 37°C and 5% CO2). Effector splenocytes were tested against mutant peptide, wild-type peptide and control peptide (vesicular stomatitis virus nucleoprotein, VSV-NP, amino acids 52-59). The average number of spots measured, subtracting background spots against VSV-NP for all mice, is shown (open circles: mice immunized with wild-type peptide; closed squares: mice immunized with mutant peptide). Data are shown for each mouse and are expressed as mean ± SEM. [Figure 4]Figure 4: Survival benefit for mice vaccinated with the newly identified mutant peptide sequences. B16F10 cells (7.5x104) were inoculated subcutaneously on day 0. Mice were vaccinated with peptide 30 (Jerini Peptide Technologies (Berlin); 100μg peptide + PolyI:C 50μg sc (Invivogen)) on days -4, +2 and +9. The control group received Poly I:C (50μg sc) alone. Tumor growth was monitored up to day +16, *, p<0.05 by log-rank (Mantel-Cox) test. [Diagram 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 pathway (see reference Kreiter, Konrad, Sester et al, Cancer Immunol. Immunother. 56:1577-1587, 2007). Figure 5C: Example of preclinical demonstration of antitumor efficacy in the 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. The vaccine vehicle incorporates functional elements for increased expression and optimization of immunogenicity. Up to 30 mutated epitopes separated by linkers can be incorporated in their native sequence context for each molecule. [Figure 6]Figure 6: Design of the construct. Figure 6A: Schematic of the RNA polyepitope construct. Cap; Cap analog; 5'UTR: 5' untranslated region; L: linker; sequence 1: RNA sequence encoding a peptide containing mutated amino acids; 3'UTR: 3' untranslated sequence; polyA: polyA tail. Figure 6B: Sequence of the RNA construct encoding two amino acid sequences containing mutated amino acids from B16F10. The start and stop codons as well as the signal peptide and MITD sequences are not part of the schematic and are represented by the "...." symbol. [Figure 7] Figure 7: Functionality of RNA polyepitopes. Data from IFNγ ELISpot with 5x105 splenocytes / well as effectors and 5x104 BMDCs as target cells. BMDCs were loaded with peptide (2 μg / ml for 2 h at 37 °C and 5% CO2) or transfected with RNA (20 μg) by electroporation. Control RNA was eGFP (left panel) or an RNA construct encoding two unrelated peptides containing mutant amino acids separated by a linker. Data are shown as mean ± SEM. Figure 7A: Data for RNAs encoding mutant peptide 30, wild type peptide 30 and mutations 30 and 31 are shown. Figure 7B: Data for RNAs encoding mutant peptide 12, wild type peptide 12 and mutations 12 and 39 are shown. Figure 7C: Representative ELISpot scans from a single mouse of the readout shown in Figure 7B are shown. [Figure 8] Figure 8: Two embodiments of an RNA polyneoepitope vaccine exhibiting junctional epitopes. RNA vaccines can be constructed with a linker between the peptides encoding the mutations (top) or without a linker (bottom). Good epitopes include those that contain somatic mutations ("*") and bind to MHC molecules. Bad epitopes include epitopes that bind to MHC molecules but contain part of either of the two peptides (bottom) or part of a peptide and a linker sequence (top). [Figure 9A]Figure 9: Discovery and characterization of the "T cell druggable mutanome." Figure 9A: Flow chart outlining the experimental procedure starting from B16F10 and C57BL / 6 samples up to ELISPOT readout. [Figure 9B] Figure 9B: The number of hits for each evaluation step and the steps of selecting mutations for DNA validation and immunogenicity testing are shown. The mutations selected for validation and immunogenicity testing were predicted to be immunogenic and expressed in genes with RPKM>10. [Figure 9C] Figure 9C: T cell druggable mutagenesis was mapped to the B16F10 genome. The rings from outside to inside represent the following subsets: (1) present in all triplicates, (2) with FDR<0.05, (3) located within protein coding regions, (4) causing nonsynonymous changes, (5) localized in expressed genes, and (6) in the validated set. Mouse chromosomes (outer circle), gene density (green), gene expression (green (low) / yellow / red (high)), and somatic mutations (orange). [Figure 10A] Figure 10: Immune responses elicited in vivo by vaccination of mice with long synthetic peptides representing the mutations. Figure 10A,B: IFN-γ ELISPOT analysis of T cell effectors from mice vaccinated with mutant-encoded peptides. Columns represent the mean (± SEM) of 5 mice per group. Asterisks indicate statistically significant differences in reactivity to mutant and wild-type peptides (Student's t-test; p-value < 0.05). Figure 10A: Splenocytes from vaccinated mice were restimulated with BMDCs transfected with peptides encoding the mutations used for vaccination, the corresponding wild-type peptide and an irrelevant control peptide (VSV-NP). [Figure 10B] FIG. 10B: For analysis of T cell reactivity against endogenously processed mutations, splenocytes from vaccinated mice were restimulated with BMDCs transfected with control RNA (eGFP) or RNA encoding the indicated mutations. [Figure 10C]FIG. 10C: Mutation 30 (gene Kif18B, protein Q6PFD6, mutation p.K739N). Sanger sequencing trace and sequence of the mutation (top). Protein domain and location of the mutation (bottom). [Figure 11] Figure 11: Antitumor activity of mutant peptide vaccines in mice bearing aggressively growing B16F10 tumors. Figure 11A: C57BL / 6 mice (n=7) were inoculated subcutaneously in the flank with 7.5x104 B16F10 cells. On days 3 and 10 after tumor inoculation, mice were vaccinated with 100μg MUT30 or MUT44 peptide + 50μg poly(I:C) or adjuvant alone. Figure 11B: C57BL / 6 mice (n=5) were given a single immunization with 100μg MUT30 peptide + 50μg poly(I:C) on day -4. On day 0, mice were inoculated subcutaneously in the flank with 7.5x104 B16F10 cells. Booster immunizations with MUT30 peptide (+poly(I:C)) were given on days 2 and 9. Kaplan-Meier viability blot (left). Tumor growth kinetics (right). [Figure 12] Figure 12: Vaccination with RNA encoding mutations results in 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 RNA encoding mutations. RNAs encoded one (monoepitope, top row), two (biepitopes, middle row) or 16 (polyepitope, bottom row) different mutations. Dots represent the average of three mice per group. Asterisks indicate statistically significant differences in reactivity to mutant and control peptides (VSV-NP) (Student's t-test; p-value <0.05). FACS plots show effectors from animals with the highest IFN-γ secretion for each mutation, indicating the phenotype of the T cell response. [Figure 13]Figure 13: Vaccination with polyepitope RNA encoding mutations results in T cell responses against several mutations. IFN-γ ELISPOT analysis in T cell effectors from mice vaccinated with a polyepitope encoding mutations containing 16 different mutations. Columns represent the mean (±SEM) of three mice per group. Photographs show triplicate wells of cells from one example animal restimulated with the indicated peptide. [Figure 14] Figure 14: Vaccination with five different model epitopes encoded by one RNA results in an immune response against all epitopes encoded. Figure 14A: IFN-γ ELISPOT analysis in T cell effectors from mice vaccinated with a model polyepitope encoding mutations containing five different model epitopes (SIINFEKL, Trp2, VSV-NP, Inf-NP, OVA class II). Splenocytes were restimulated with the indicated peptides. Spots represent the average of triplicate wells from five mice per group. Figure 14B: Pentamer staining of blood lymphocytes from one control mouse and one mouse immunized with the model polyepitope. CD8+ cells stained with Inf-NP pentamer are specific for the Inf-NP peptide. [Figure 15A] Figure 15: Mutation-inducing CD4+ T cells can synergize with CD8+ T cell epitopes to induce potent antitumor effects on B16F10 melanoma. C57BL / 6 mice (n=8) were inoculated with 1x105 B16F10 cells subcutaneously in the flank of the mouse. On days 3, 10 and 17 after tumor inoculation, mice were vaccinated with 100μg MUT30, Trp2 or both peptides + 50μg poly(I:C). Figure 15A: Mean tumor growth kinetics for each group is shown. On day 28, the mean values ​​between single treatment groups and untreated animals and the combination group are statistically different (Mann-Whitney test, p-value<0.05). [Figure 15B] Figure 15B: Kaplan-Meier survival plots of the different groups. The survival curves of 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. Numbers for each step are shown as an example for one B16 sample compared to one black6 sample. "Exon" refers to exon coordinates defined by all protein-coding RefSeq transcripts. [Figure 17] Figure 17: Venn diagram showing the number of somatic mutations in protein-coding exons found by each individually, two or all three software tools. Numbers were calculated after filtering and represent the consensus of all three samples. [Figure 18A] FIG. 18A: Examples of single nucleotide mutations found: a somatic mutation found in all three B16 samples (left), a non-somatic mutation found in all B16 and black6 samples (center) and a mutation found only in one black6 sample (right). [Figure 18B] Figure 18B: Calculated FDR distribution for the dataset with selected validated mutations; the distribution is visualized as the mean estimated ROC curve, and the grey bars show the 95% confidence interval for the mean in both dimensions at uniform sampling positions. The mean was taken from the distribution of the estimated ROC curves of the FDR for all 18 possible combinations (see text). [Figure 19] Figure 19A: Estimated ROC curves for the comparison of three different software tools (duplicates, coverage 38x). Figure 19B: Estimated ROC curves for the comparison of different mean sequencing depths (samtools, no replicates). 38x indicates the coverage obtained by the experiment, the other coverages were downsampled starting from this data. Figure 19C: Estimated ROC curves visualizing the effect of experimental replicates (coverage 38x, samtools). Figure 19D: Estimated ROC curves for different sequencing protocols (samtools, no replicates). The curves were calculated using the results of a 2x100 nucleotide library. [Figure 20]Figure 20A: The 10 validated mutations with the lowest FDR selected with 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: The relative amount of mutations found in the same dataset as in A for a given FDR cutoff value, plotted separately for all mutations in the dataset and for validated mutations. For visual clarity, only values ​​between 0 and 10% FDR are shown. [Figure 21] Figure 21: Antitumor activity of polyepitope RNA vaccines encoding mutations. C57BL / 6 mice (n=10) were inoculated with 1x105 B16F10 cells subcutaneously in the flank of the mice. At days 3, 6, 10, 17 and 21 after tumor inoculation, the mice were vaccinated with liposomal RNA transfection reagent formulated with polytope RNA. The control group received liposomes without RNA. The figure shows Kaplan-Meier survival plots of the various groups. The survival curves are statistically different (log-rank test, p-value=0.0008). [Figure 22] Figure 22: Selection of combinations of tumor antigens as targets for cancer therapy. The combination of only three tumor antigens, DCT, TYR and TPTE, is sufficient to represent 88% of the melanoma metastasis samples analyzed. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[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, which may vary.It should also be understood that the terms used herein are only intended to describe specific embodiments, and are not intended to limit the scope of the present invention, which is limited only by the scope of the appended 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] In the following, the elements of the present invention are described. Although these elements are listed with specific embodiments, it should be understood that they can be combined in any manner and in any number to create additional embodiments. The various described examples and preferred embodiments should not be construed as limiting the invention to only the specifically described embodiments. The description should be understood to support and encompass embodiments that combine the specifically described embodiments with many of the disclosed and / or preferred elements. Furthermore, any order and combination of all elements described in this application should be considered to be disclosed by the description in this application unless otherwise indicated by the context. For example, if in one preferred embodiment the RNA comprises a poly(A) tail of 120 nucleotides, and in another preferred embodiment the RNA molecule comprises a 5' cap analog, in a preferred embodiment the RNA comprises a poly(A) tail of 120 nucleotides and a 5' cap analog.

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

[0068] The practice of the present invention employs, unless otherwise indicated, conventional methods of biochemistry, cell biology, immunology, and recombinant DNA techniques as described in the art (e.g., Molecular Cloning: A Laboratory Manual, 2001). nd Edition, J. Sambrook et al. eds., Cold Spring Harbor Laboratory Press, Cold Spring Harbor 1989).

[0069] Throughout this specification and the claims that follow, unless otherwise required by context, the word "comprises" and variations such as "comprises" refer to the inclusion of a stated member, integer or step, or group of members, integers or steps, and not the exclusion of any other member, integer or step, or group of members, integers or steps, although it is understood that in some embodiments, such other member, integer or step, or group of members, integers or steps may be excluded, i.e., the subject matter resides in the inclusion of a stated member, integer or step, or group of members, integers or steps. The terms "a" and "the" and similar references used in connection with the description of the present invention (particularly in connection with the claims) should be construed to include both the singular and the plural, unless otherwise indicated herein or clearly contradicted by context. The recitation of ranges of values ​​herein is merely intended to serve as a shorthand way of individually referring to each separate value falling within the range. Unless otherwise indicated herein, each separate value is incorporated herein as if individually recited herein.

[0070] All methods described herein can be performed in any suitable order unless otherwise indicated herein or clearly contradicted by context. The use of any and all examples or exemplary language (e.g., "etc.") provided herein is intended merely to better illustrate the invention and does not pose limitations on the scope of the invention or the claims. No language in this specification should be construed as indicating any non-claimed element as essential to the practice of the invention.

[0071] Several documents are cited throughout the text of this specification. Each of the documents cited herein, whether above or below, is incorporated herein by reference in its entirety (including all patents, patent applications, scientific publications, manufacturer's specifications, instructions, etc.). Nothing herein should be construed as an admission that the invention is not entitled to antedate such disclosure by virtue of prior invention.

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

[0073] The term "recombinant" in the context of the present invention means "created through genetic engineering". Preferably, a "recombinant entity" such as a recombinant polypeptide in the context of the present invention is the result of a combination of entities such as amino acid or nucleic acid sequences that do not exist in nature and preferably are not combined in nature. For example, a recombinant polypeptide in the context of the present invention may contain several amino acid sequences, such as neoepitopes or vaccine sequences derived from different proteins or different parts of the same protein, fused together, for example by peptide bonds or suitable linkers.

[0074] As used herein, the term "naturally occurring" refers to the fact that an entity can be found in nature. For example, a peptide or nucleic acid that is present in an organism (including viruses), can be isolated from a natural source, and has not been intentionally modified by the hand of man in a laboratory is naturally occurring.

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

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

[0077] "Inducing an immune response" may mean that there was no immune response against a particular antigen before induction, but it may also mean that there is a certain level of immune response against a particular antigen before induction, and that the immune response is enhanced after induction. Thus, "inducing an immune response" also encompasses "enhancing an immune response". Preferably, after inducing an immune response in a subject, the subject is protected from developing a disease, such as a cancer disease, or the disease state is improved by inducing an immune response. For example, an immune response against a tumor-expressed antigen may be induced in a patient with a cancer disease or a subject at risk of developing a cancer disease. Inducing an immune response in this case may mean that the disease state of the subject is improved, that the subject does not develop metastasis, or that a subject at risk of developing a cancer disease does not develop a cancer disease.

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

[0079] "Cellular immune response", "cellular response", "cellular response to antigen" or similar terms are intended to include cellular responses to cells characterized by presentation of antigen by MHC class I or class II. Cellular responses involve cells called T cells or T lymphocytes that act as either "helpers" or "killers". Helper T cells (CD4 + Anti-tumor CTL responses against tumor cells expressing one or more tumor-expressed antigens, preferably presenting such tumor-expressed antigens in conjunction with class I MHC.

[0080] An "antigen" according to the present invention includes any substance that induces an immune response. In particular, an "antigen" relates to any substance, preferably a peptide or protein, that reacts specifically with antibodies or T lymphocytes (T cells). According to the present invention, the term "antigen" includes any molecule that comprises at least one epitope. Preferably, in the context of the present invention, an antigen is a molecule that, optionally 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 cell-mediated immune response. In the context of an embodiment of the present invention, the antigen is preferably presented in the context of an MHC molecule by cells, preferably by antigen-presenting cells, including abnormal cells, in particular cancer cells, which results in an immune response against the antigen. The antigen is preferably a product that corresponds to a naturally occurring antigen or is derived from a naturally occurring antigen. Such naturally occurring antigens include tumor antigens.

[0081] In a preferred embodiment, 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, particularly one that is present mainly intracellularly or as a surface antigen of the tumor cell. For example, tumor antigens include carcinoembryonic antigen, α1-fetoprotein, isoferritin and fetal sulfoglycoprotein, α2-H-iron protein and γ-fetoprotein. According to the present invention, tumor antigens preferably include any antigen that is expressed in a tumor or cancer and in a tumor or cancer cell, and is specific to the tumor or cancer and to the tumor or cancer cell, optionally in terms of type and / or expression level. In one embodiment, the term "tumor antigen" or "tumor-associated antigen" refers to a protein that is specifically expressed in a limited number of tissues and / or organs or at a particular developmental stage under normal conditions, for example, a tumor antigen may be specifically expressed in gastric tissue, preferably in gastric mucosa, in reproductive organs, such as in testis, in trophoblast tissue, such as in placenta, or in germ lineage cells under normal conditions, and is expressed or aberrantly expressed in one or more tumor or cancer tissues. In this context, "limited number" preferably means 3 or less, more preferably 2 or less. Tumor antigens in the context of the present invention include, for example, differentiation antigens, preferably cell type-specific differentiation antigens, i.e. proteins that are specifically expressed in a specific cell type at a specific developmental stage under normal conditions, cancer / testis antigens, i.e. proteins that are specifically expressed in the testis 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 the context of the present invention, tumor antigens expressed by cancer cells in a subject, for example a patient suffering from a cancer disease, are preferably self-proteins in said subject. In a preferred embodiment, tumor antigens in the context of the present invention are specifically expressed in tissues or organs that are non-essential under normal conditions, i.e. tissues or organs that do not lead to the death of the subject when damaged by the immune system, or in organs or structures of the body that are not or hardly accessed by the immune system.

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

[0083] The term "immunogenicity" relates to the relative effectiveness of an antigen to induce an immune response.

[0084] An "antigenic peptide" according to the present invention preferably relates to a part or fragment of an antigen capable of stimulating an immune response, preferably a cellular response, against the antigen or against cells characterized by the expression of the antigen, preferably characterized by the presentation of the antigen, such as abnormal cells, in particular cancer cells. Preferably, the antigenic peptide is capable of stimulating a cellular response against cells characterized by the presentation of the antigen by class I MHC, preferably stimulating antigen-responsive cytotoxic T lymphocytes (CTLs). Preferably, the antigenic peptide according to the present invention is an MHC class I and / or class II presented peptide or can be processed to generate an MHC class I and / or class II presented peptide. Preferably, the antigenic peptide comprises an amino acid sequence that substantially corresponds to the amino acid sequence of a fragment of the antigen. Preferably, said fragment of the antigen is an MHC class I and / or class II presented peptide. Preferably, the antigenic peptide according to the present invention comprises an amino acid sequence that substantially corresponds to the amino acid sequence of such a fragment and is processed to generate such a fragment, i.e. an MHC class I and / or class II presented peptide derived from the antigen.

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

[0086] If the peptide is part of a larger entity, such as a vaccine sequence or polypeptide, which comprises additional sequences, and is presented after processing, in particular after cleavage, the peptide generated by processing has a length suitable for binding to an MHC molecule, in particular a class I MHC molecule, preferably 7-20 amino acids long, more preferably 7-12 amino acids long, more preferably 8-11 amino acids long, in particular 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. the sequence corresponds substantially, preferably completely identical, to a fragment of the antigen. Thus, the antigen peptide or vaccine sequence according to the invention comprises in one embodiment a sequence of 7-20 amino acids long, more preferably 7-12 amino acids long, more preferably 8-11 amino acids long, in particular 9 or 10 amino acids long, which corresponds substantially, preferably completely identical, to a fragment of the antigen, which, after processing of the antigen peptide or vaccine sequence, forms the peptide to be presented. According to the present invention, such peptides generated by processing comprise the identified sequence changes.

[0087] According to the present invention, an antigenic peptide or epitope may be present in the 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 peptides or epitopes are generated after appropriate processing.

[0088] Peptides with amino acid sequences that substantially correspond to the sequence of peptides presented by class I MHC may differ in one or more residues that are not essential for TCR recognition of peptides presented by class I MHC or for peptide binding to MHC. Such substantially corresponding peptides can also stimulate antigen-responsive CTL and can be considered immunologically equivalent. Peptides with amino acid sequences that differ from the presented peptides in residues that do not affect TCR recognition but improve the stability of binding to MHC can improve the immunogenicity of the antigenic peptide and can be referred to herein as "optimized peptides". Using existing knowledge of which of these residues are more likely to affect binding to either MHC or TCR, a rational approach to the design of substantially corresponding peptides can be used. The resulting functional peptides are contemplated as antigenic peptides.

[0089] The antigenic peptide should be recognizable by a T cell receptor when presented by MHC. Preferably, when recognized by a T cell receptor, the antigenic peptide can induce, in the presence of an appropriate costimulatory signal, the clonal expansion of T cells bearing a T cell receptor that specifically recognizes the antigenic peptide. Preferably, the antigenic peptide, especially when presented in the context of an MHC molecule, can stimulate an immune response, preferably a cellular response, against the antigen from which they are derived, or against cells characterized by the expression of the antigen, preferably characterized by the presentation of the antigen. Preferably, the antigenic peptide can stimulate a cellular response against cells characterized by the presentation of the antigen by class I MHC, preferably stimulating an antigen-responsive CTL. Such cells are preferably target cells.

[0090] "Antigen processing" or "processing" refers to the degradation of a peptide or protein, such as a polypeptide or antigen, into processing products that are fragments of said peptide or protein (e.g., degradation of a polypeptide into peptides), and the association (e.g., by binding) of one or more of these fragments with an MHC molecule for presentation to a specific T cell by a cell, preferably an antigen-presenting cell.

[0091] An "antigen-presenting cell" (APC) is a cell that presents peptide fragments of protein antigens in association with MHC molecules on its cell surface. Some APCs can activate antigen-specific T cells.

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

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

[0094] Dendritic cells (DCs) are a population of leukocytes 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 immune responses and that activation of these cells is an essential step for the induction of antitumor immunity.

[0095] Dendritic cells are conveniently classified as "immature" and "mature" cells and can be used as a simple method to distinguish between two well-characterized phenotypes, however, this nomenclature should not be construed to exclude 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, which correlates with high expression of Fcγ receptors and mannose receptors. The mature phenotype is typically characterized by lower expression of these markers, but also by high expression of cell surface molecules responsible for T cell activation, such as MHC class I and class II, adhesion molecules (e.g., CD54 and CD11), and costimulatory molecules (e.g., CD40, CD80, CD86, and 4-1BB).

[0097] Maturation of dendritic cells is referred to as a state of dendritic cell activation in which such antigen-presenting dendritic cells result in the priming of T cells, whereas presentation by immature dendritic cells results in tolerance. Dendritic cell maturation is primarily triggered by biomolecules with microbial characteristics (bacterial DNA, viral RNA, endotoxins, etc.) detected by innate receptors, proinflammatory cytokines (TNF, IL-1, IFN), ligation of CD40 on the dendritic cell surface by CD40L, and substances released from cells undergoing 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 alpha.

[0098] Non-professional antigen-presenting cells do not constitutively express the MHC class II proteins required for interaction with naive T cells; these are expressed only after stimulation of the non-professional antigen-presenting cells with specific cytokines, such as IFNγ.

[0099] An "antigen-presenting cell" can be loaded with an MHC class I-presented peptide by transducing the cell with a nucleic acid, preferably RNA, encoding a peptide or a polypeptide comprising the peptide to be presented, e.g., a nucleic acid encoding an antigen.

[0100] In some embodiments, the pharmaceutical composition containing gene delivery vehicle that targets dendritic cell or other antigen-presenting cell can be administered to patient, causing transfection to occur in vivo.The in vivo transfection of dendritic cell can generally be carried out by any method known in the art, such as the one described in WO97 / 24447 or the gene gun approach described by Mahvi et al., Immunology and cell Biology 75:456-460,1997.

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

[0102] "Target cell" is intended to mean a cell that is the target of an immune response, such as a cellular immune response. Target cells include cells that present an antigen or an antigen epitope, i.e., a peptide fragment derived from an antigen, and include any undesirable cell, such as a cancer cell. In a preferred embodiment, the target cell is a cell that expresses an antigen as described herein, and preferably presents said antigen with class I MHC.

[0103] The term "epitope" refers to an antigenic determinant in a molecule such as an antigen, i.e. a part in a molecule or a fragment of a molecule that is recognised by the immune system, e.g. by T cells, particularly when presented in the context of an MHC molecule. An epitope of a protein, such as a tumor antigen, preferably comprises a continuous or discontinuous portion of said protein and is preferably 5-100, preferably 5-50, more preferably 8-30, most preferably 10-25 amino acids in length, e.g. 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 in length. It is particularly preferred that the epitope in the context of the present invention is 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 thus can be an "MHC binding peptide" or an "antigenic peptide". The term "MHC binding peptide" relates to a peptide that binds to an MHC class I and / or MHC class II molecule. For class I MHC / peptide complexes, the binding peptide is typically 8-10 amino acids long, although longer or shorter peptides can be effective. For class II MHC / peptide complexes, the binding peptide is typically 10-25 amino acids long, particularly 13-18 amino acids long, although longer and shorter peptides can be effective.

[0105] The terms "epitope", "antigenic peptide", "antigenic 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 against the antigen or against a cell that expresses or contains, preferably presents, the antigen. Preferably, the term relates to an immunogenic portion of an antigen. Preferably, this is the portion of the antigen that is recognized (i.e. specifically bound) by a T cell receptor, especially when presented in the context of an MHC molecule. Preferred such immunogenic portions bind to MHC class I or class II molecules. 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 a reference, such as a normal non-cancerous cell or a germline cell, but is found in a cancer cell. This particularly includes situations where a corresponding epitope is found in a normal non-cancerous cell or a germline cell, but due to one or more mutations in the cancer cell, the sequence of the epitope has been altered to create a neoepitope.

[0107] The term "portion" refers to a fraction. With respect to a particular structure, such as an amino acid sequence or a protein, the term "portion" may refer to a continuous or discontinuous fraction of said structure. Preferably, a portion of an amino acid sequence comprises 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 said amino acid sequence. Preferably, when the portion is a discontinuous fraction, said discontinuous fraction is composed of 2, 3, 4, 5, 6, 7, 8 or more portions of the structure, each portion being a continuous element of the structure. For example, a non-contiguous fraction of an amino acid sequence may consist of 2, 3, 4, 5, 6, 7, 8 or more, preferably 4 or less, portions of said amino acid sequence, where each portion preferably comprises at least 5 consecutive amino acids, at least 10 consecutive amino acids, preferably at least 20 consecutive amino acids, preferably at least 30 consecutive amino acids of the amino acid sequence.

[0108] The terms "part" and "fragment" are used interchangeably herein and refer to a continuous element. A part of a structure, e.g. an amino acid sequence or a protein, refers to a continuous element of said structure. A part, portion or fragment of a structure preferably comprises one or more functional properties of said structure. For example, a part, portion or fragment of an epitope, peptide or protein is preferably immunologically equivalent to the epitope, peptide or protein from which it is derived. In the context of the present invention, a "part" of a structure, e.g. an amino acid sequence, preferably comprises, 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%, at least 99% of the entire structure or entire amino acid sequence.

[0109] The term "immunoreactive cells" in the context of the present invention relates to cells that exert effector functions during an immune response. "Immune reactive cells" are preferably capable of binding to cells characterized by the presentation of antigens or antigenic peptides derived from antigens and are capable of mediating an immune response. For example, such cells secrete cytokines and / or chemokines, secrete antibodies, recognize cancerous cells and optionally eliminate such cells. For example, immune reactive 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, "immunoreactive cells" are T cells, preferably CD4 + and / or CD8 + T cells.

[0110] Preferably, an "immunoreactive cell" recognises an antigen or an antigenic peptide derived from an antigen with some degree of specificity, particularly when presented in association with an MHC molecule, such as on the surface of an antigen-presenting cell or an abnormal cell such as a cancer cell. Preferably, said recognition enables the cell which recognises the antigen or an antigenic peptide derived from said antigen to become responsive or reactive. The cell may be a helper T cell (CD4 T cell) bearing a receptor which recognises the antigen or an antigenic peptide derived from the antigen in association with an MHC class II molecule. + In the case of CD4+ T cells, such responsiveness or reactivity may be determined by the release of cytokines and / or CD8 +The activation of lymphocytes (CTL) and / or B cells may include activation. If the cell is a CTL, such responsiveness or reactivity may include elimination, for example by apoptosis or perforin-mediated cytolysis, of cells presented in the context of MHC class I molecules, i.e., cells characterized by presentation of antigen by class I MHC. According to the present invention, CTL responsiveness may include sustained calcium flux, cell division, production of cytokines such as IFN-γ and TNF-α, upregulation of activation markers such as CD44 and CD69, and specific cytolytic killing of target cells expressing the antigen. CTL responsiveness may also be determined using artificial reporters that pinpoint CTL responsiveness. CTLs that recognize and are responsive or reactive to an antigen or antigenic peptides derived from an antigen are also referred to herein as "antigen-responsive CTLs." If the cell is a B cell, such responsiveness may include release of immunoglobulins.

[0111] The terms "T cells" and "T lymphocytes" 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 a 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 special 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 different function.

[0113] T helper cells assist other white blood cells in immunological processes, including, among other functions, 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 become activated when presented with peptide antigens by MHC class II molecules expressed on the surface of antigen-presenting cells (APCs). Once activated, they divide rapidly and secrete small proteins called cytokines that regulate or aid in active immune responses.

[0114] Cytotoxic T cells destroy virus-infected and tumor cells, and are also involved in transplant 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] The majority of T cells have a T cell receptor (TCR) that 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. γδ T cells (gamma delta T cells) are a small subset of T cells that have a different T cell receptor (TCR) on their surface. However, in γδ T cells, the TCR is composed of one γ chain and one δ chain. This group of T cells is much rarer than αβ T cells (2% of all T cells).

[0116] The first signal in T cell activation is given by the T cell receptor binding to a short peptide presented by the major histocompatibility complex (MHC) on another cell. This ensures that only T cells with a TCR specific for that peptide are activated. The partner cell is usually a professional antigen-presenting cell (APC), usually a dendritic cell in the case of a naive response, although B cells and macrophages can 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, since the end of the binding groove of MHC class II molecules is open.

[0117] According to the present invention, a T cell receptor is capable of binding to a predefined target if it has significant affinity for the predefined target in a standard assay and binds to the predefined target. "Affinity" or "binding affinity" is often measured using the equilibrium dissociation constant (K D ) A T cell receptor is (substantially) incapable of binding to a target if it has no significant affinity for the target in a standard assay and does not significantly bind to said target.

[0118] A T cell receptor is preferably capable of specifically binding to a predefined target. A T cell receptor is specific for said predefined target if it is capable of binding to said predefined target but is (substantially) incapable of binding to other targets, i.e. it has no significant affinity for other targets in standard assays and does not significantly bind to other targets.

[0119] Cytotoxic T lymphocytes can be generated in vivo by incorporating antigen or antigen peptide into antigen-presenting cells in vivo. Antigen or antigen peptide can be present as protein, DNA (e.g., in a vector) or RNA. Antigen can be processed to generate peptide partners of MHC molecules, while fragments thereof can be presented without further processing. The latter is especially true when they can bind to MHC molecules. In general, administration to patients is possible by intradermal injection. However, injection can also be performed by intranodal injection into lymph nodes (Maloy et al. (2001), Proc Natl Acad Sci USA 98:3299-303). The resulting cells present the complex of interest and are recognized by autologous cytotoxic 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 occurrence of cytolytic activity. For CD4+ T cells, the preferred method for detecting specific T cell activation is detecting T cell proliferation. For CD8+ T cells, the preferred method for detecting specific T cell activation is detecting occurrence of cytolytic activity.

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

[0122] MHC regions are divided into three subgroups: class I, class II, and class III. MHC class I proteins contain α chains and β2 microglobulin (not part of the MHC encoded by chromosome 15). These present antigen fragments to cytotoxic T cells. On most immune system cells, especially antigen-presenting cells, MHC class II proteins contain α and β chains and present antigen fragments to T helper cells. MHC class III regions code for other immune components such as complement components and some that code for cytokines.

[0123] In humans, the genes in the MHC region that code for antigen-presenting proteins on the cell surface are called human leukocyte antigen (HLA) genes. However, the abbreviation MHC is often used to refer to the HLA gene products. The 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 invention, the MHC molecule is an HLA molecule.

[0125] By "cells characterized by antigen presentation" or "cells presenting antigen" or similar expressions is meant cells, such as abnormal cells, e.g. cancer cells, or antigen-presenting cells, that present the antigen they express, or a fragment derived from said antigen, in the context of an MHC molecule, in particular an MHC class I molecule, e.g. by processing the antigen. Similarly, the term "diseases characterized by antigen presentation" refers to diseases involving cells characterized by antigen presentation, in particular by class I MHC. Presentation of antigen by cells can be performed by transfecting the cells with a nucleic acid, such as an RNA, that codes for the antigen.

[0126] By "fragment of a presented antigen" or similar expressions is meant that the fragment, e.g., when added directly to an antigen-presenting cell, can be presented by MHC class I or class II, preferably MHC class I. 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, e.g. immunologically equivalent amino acid sequences, exhibit the same or essentially the same immunological properties and / or exert the same or essentially the same immunological effect, e.g. in terms of the type of immunological effect, such as the induction of a humoral and / or cellular immune response, the strength and / or duration of the immune response induced, or the specificity of the immune response induced. In the context of the present invention, the term "immunologically equivalent" is preferably used in reference to the immunological effect or properties of a peptide used for immunization. For example, an amino acid sequence is immunologically equivalent to a reference amino acid sequence if, when exposed to the immune system of a subject, said amino acid sequence induces an immune response with specificity that reacts with the reference amino acid sequence.

[0128] The term "immune effector function" in the context of the present invention encompasses any function mediated by components of the immune system that results in the inhibition of tumor growth and / or tumor development, including, for example, the killing of tumor cells, or the inhibition of tumor dissemination and metastasis. Preferably, the immune effector function in the context of the present invention is a T cell-mediated effector function. Such functions include the activation of helper T cells (CD4 + In the case of T cells, recognition of antigens or antigenic peptides derived from antigens in association with MHC class II molecules by T cell receptors, release of cytokines and / or CD8 +This includes activation of lymphocytes (CTLs) and / or B cells, and in the case of CTLs, recognition of antigens or antigenic peptides derived from antigens in association with MHC class I molecules by the T cell receptor, elimination by, for example, apoptosis or perforin-mediated cytolysis of cells presented in association with MHC class I molecules, i.e., cells characterized by presentation of antigen by class I MHC, production of cytokines such as IFN-γ and TNF-α, and specific cytolytic killing 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] A "nucleic acid" according to the present invention is preferably deoxyribonucleic acid (DNA) or ribonucleic acid (RNA), more preferably RNA, most preferably in vitro transcribed RNA (IVT RNA) or synthetic RNA. A nucleic acid according to the present invention includes genomic DNA, cDNA, mRNA, recombinantly produced molecules and chemically synthesized molecules. According to the present invention, a nucleic acid may be present as a single-stranded or double-stranded linear or covalently closed circular molecule. A nucleic acid may be isolated according to the present invention. The term "isolated nucleic acid" means according to the present invention that a 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. A nucleic acid may be used in the form of RNA, which may be prepared by in vitro transcription, in particular from a DNA template, for introduction into a cell, i.e. for transfection of a cell. Said RNA may be further modified by sequence stabilization, capping and polyadenylation before application.

[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 (nucleotide substitution, addition or deletion) in a nucleic acid sequence compared to a reference. "Somatic mutations" can occur in any cell of the body except germ cells (sperm and eggs) and therefore are not transmitted to offspring. These changes can (but do not always) cause cancer or other diseases. Preferably, the mutation is a nonsynonymous mutation. The term "nonsynonymous 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, chromosomycin and RNA editing.

[0134] According to the present invention, the term "indel" refers to a special class of mutations defined as mutations that result in co-localized insertions and deletions and net gains or losses of nucleotides. In the coding region of a genome, unless the length of the indel is a multiple of three, it will result 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 of the nucleotides.

[0135] Fusions can generate hybrid genes formed from two previously separate genes. This can occur as a result of translocations, interstitial deletions, or chromosomal inversions. Often the fusion gene is an oncogene. Oncogenic fusion genes can result in gene products with new or different functions from the two fusion partners. Alternatively, a proto-oncogene may be fused to a strong promoter, whereby the oncogenic function is initiated by the upregulation of the upstream fusion partner caused 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 "chromosipsis" refers to a genetic phenomenon in which specific regions of the genome are disrupted by a single disruptive event and then sutured back together.

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

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

[0139] According to the present invention, a "reference" can be used to correlate and compare the results obtained by the method of the present invention from tumor specimens.Typically, the "reference" can be obtained based on one or more normal specimens, particularly specimens not affected by cancer disease, obtained from the patient or one or more different individuals, preferably healthy individuals, particularly individuals of the same species.The "reference" can be empirically determined by testing a sufficiently large number of normal specimens.

[0140] Any suitable sequencing method can be used according to the present invention, with next-generation sequencing (NGS) technology being preferred. Third-generation sequencing methods may replace NGS technology in the future to speed up the sequencing step of the method. For clarity, the term "next-generation sequencing" or "NGS" in the context of the present invention refers to all novel high-throughput sequencing technologies that randomly read a nucleic acid template in parallel along the entire genome by dividing the entire genome into small pieces, as opposed to the "traditional" sequencing method 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 transcribed sequences of the genome) or methylome (all methylated sequences of the genome) in a very short time, for example within 1-2 weeks, preferably within 1-7 days, or most preferably within less than 24 hours, and in principle enable a single-cell sequencing approach. Several NGS platforms are commercially available or mentioned in the literature, 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 the context of the present invention. Non-limiting examples of such NGS technologies / platforms are as follows: 1) The "sequencing by synthesis" technique known as pyrosequencing, implemented in the GS-FLX 454 Genome Sequencer® from 454 Life Sciences (Branford, Connecticut), a Roche affiliate, first described in, for example, 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 by vigorous vortexing into aqueous micelles containing the PCR reactants surrounded by oil for emulsion PCR amplification. During the pyrosequencing process, as the polymerase synthesizes the DNA strands, light emitted from the phosphate molecules during nucleotide incorporation is recorded. 2) The "sequencing by synthesis" approach developed by Solexa (now part of Illumina Inc., San Diego, California), implemented for example in the Illumina / Solexa Genome Analyzer® and Illumina HiSeq 2000 Genome Analyzer®, based on reversible dye-terminators. In this technique, all four nucleotides are added simultaneously to oligo-primed cluster fragments in a flow cell channel along with DNA polymerase. Bridge amplification extends the cluster strands with all four fluorescently labeled nucleotides for sequencing. 3) The "sequencing by ligation" approach, implemented for example in 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 position. The oligonucleotides are annealed and ligated; selective ligation by DNA ligase to match sequences results in an informative signal of the nucleotide at that position. Prior to sequencing, the DNA is amplified by emulsion PCR. The resulting beads, each containing only copies 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 uses a "sequencing by ligation" approach by using emulsion PCR based on randomly positioned beads to amplify DNA fragments for parallel sequencing. 4) Single molecule sequencing technology, such as implemented in the PacBio RS system from Pacific Biosciences (Menlo Park, California) or the HeliScope platform from Helicos Biosciences (Cambridge, Massachusetts). The defining feature of this technology is its ability to sequence single DNA or RNA molecules 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 from Visigen Biotechnology (Houston, Texas). Other fluorescence-based single molecule technologies are from USGenomics (GeneEngine) and Genovoxx (AnyGene). 5) Nanotechnology for single molecule sequencing, using various nanostructures arranged on a chip to observe the movement of polymerase molecules on a single strand during replication, for example. Non-limiting examples of nanotechnology-based approaches are the GridON® platform from Oxford Nanopore Technologies (Oxford, UK), the Hybridization-Assisted Nanopore Sequencing (HANS®) platform developed by Nabsys (Providence, Rhode Island), and a patent-protected 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. Ion Torrent Systems (San Francisco, California), for example, uses a high-density array of micromachined wells to perform this biochemical process in a massively parallel manner. Each well holds a different DNA template. Beneath the wells is an ion-sensitive layer, and beneath that is a proprietary ion sensor.

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

[0142] Several targeted NGS methods for exome sequencing have been described in the literature (for a review, see, for example, Teer and Mullikin 2010: Human Mol Genet 19(2), R145-51), all of which can be used with the present invention. Many of these methods use hybridization techniques (e.g., described as genome capture, genome partitioning, genome enrichment, etc.), including 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 the sequence of a germline sample, it is preferable to determine the sequence in the replicates of one or both of these sample types in order 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 a reference sample, such as the sequence of a germline sample, two or three or more times. Alternatively or additionally, the sequence of a tumor sample is determined two or three or more times. It may also be possible to determine the sequence of a reference sample, such as the sequence of a germline sample, and / or the sequence of a tumor sample two or more times by determining the sequence in genomic DNA at least once and determining the sequence in RNA of said reference sample and / or said tumor sample at least once. By determining the mutations between the replicates of a reference sample, such as the germline sample, for example, the expected false positive rate (FDR) of somatic mutations as a statistical quantity can be estimated. Technical repetitions of one sample should produce identical results, and any mutation detected in this "same vs. same comparison" is a false positive. In particular, to determine the false discovery rate for somatic mutation detection in tumor samples compared to reference samples, the technical repeats of the reference samples can be used as a reference to estimate the number of false positives.In addition, various quality-related metrics (e.g., coverage or SNP quality) can be combined into a single quality score using machine learning approaches.For a given somatic mutation, all other mutations with a higher quality score can be counted, which allows ranking of all mutations in a data set.

[0144] According to the present invention, high throughput whole genome single cell genotyping methods 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. Tumor tissue / cells and healthy tissue are harvested from a given patient. 2. Genetic material is extracted from cancerous and healthy cells and then the exome (DNA) is sequenced using standard next generation sequencing (NGS) protocols. NGS coverage is such that heterozygous alleles with a frequency of at least 5% can be detected. The transcriptome (RNA) is also extracted from the cancer cells, converted to cDNA, and sequenced to determine which genes are expressed by the cancer cells. 3. Identify expressed nonsynonymous single nucleotide variations (SNVs) as described herein. Filter out sites that are SNPs in healthy tissues. 4. N=96 mutations from (3) are selected over a range of frequencies. SNP genotyping assays based on fluorescence detection are designed and synthesized for these mutations (examples of such assays include TaqMan-based SNP assays by Life Technologies or SNPtype assays by Fluidigm). The assays include specific target amplification (STA) primers to amplify amplicons containing a given SNV (this is standard in TaqMan and SNPtype assays). 5. Individual cells are isolated from tumor and healthy tissues by either laser microdissection (LMD) or separation into single cell suspensions 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 enriched for cancerous cells. Enrichment methods include specific staining, sorting by cell size, histological examination during LMD, etc. 6. Individual cells are isolated in PCR tubes containing master mix and STA primers and amplicons containing SNVs are amplified. Alternatively, the genome of a single cell is amplified by whole genome amplification (WGA) as previously described (Frumkin D. et al. (2008) Cancer Research 68:5924). Cell lysis is achieved by either a 95°C heating step or a dedicated lysis buffer. 7. STA amplified samples are diluted and loaded onto Fluidigm genotyping arrays. 8. Use samples from healthy tissue as positive controls to determine homozygous allele clusters (no mutations). Because NGS data show that homozygous mutations are extremely rare, typically only two clusters are expected: XX and XY, where X=healthy. 9. There is no limit to the number of arrays that can be performed, and in practice it is possible to assay up to about 1000 single cells (about 10 arrays). When performed on 384 plates, sample preparation can be shortened to a few days. 10.The SNV for each cell is then determined.

[0146] In another embodiment of high throughput whole genome single cell genotyping, an NGS platform may be used. Such an approach may include the following steps: 1. Steps 1 to 6 above are the same, except that N (the number of SNVs to be tested) can be much larger than 96. In the case of WGA, this is followed by several cycles of STA. 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. Thus, each cell contains a unique barcode. 3. The amplicons from all cells are mixed and sequenced by NGS. The practical limit to the number of cells that can be multiplexed is the number of plates that can be prepared. Since samples can be prepared in 384 plates, the practical limit is about 5000 cells. 4. Detect SNVs (or other structural abnormalities) in individual cells based on sequence data.

[0147] Tumor phylogenetic reconstruction based on single cell genotyping ("phylogenetic antigen prioritization") may be used in accordance with the present invention to prioritize antigens. In addition to prioritizing antigens based on criteria such as expression, type of mutation (nonsynonymous vs. other mutations), MHC binding properties, and the like, further dimensions of prioritization designed to address intra- and inter-tumor heterogeneity and biopsy biases can be used, for example as described below.

[0148] 1. Identification of the Most Abundant Antigen Based on the single-cell assays described above in connection with 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 suggests that homozygous mutations (hits on both alleles) are rare events. Therefore, a phylogenetic tree of tumor somatic mutations can be created from single-cell SNV datasets without the need for haplotyping. Germline sequences are used to root the phylogenetic tree. An algorithm to reconstruct ancestral sequences is used to reconstruct sequences at nodes close to 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). Mutations in ancestral sequences are predicted to be fixed in the tumor because the probability that two mutations occur in the same allele at the same genomic position is low.

[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: if, for example, two SNVs are likely to be present in all cells derived 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 where the biopsy was taken. In addition, basal antigens are likely to be present in metastatic tumors derived from the primary tumor. Therefore, prioritizing basal antigens for IVAC may greatly increase the chances that IVAC will be able to eradicate the entire tumor and not just a portion of it.

[0151] If secondary tumors are present and are also sampled, an evolutionary tree of all tumors can be inferred, which can improve the robustness of the phylogenetic tree and allow detection of mutations underlying all tumors.

[0152] 3. Identification of antigens with maximum tumor coverage Another approach to obtain antigens with maximal 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, a phylogenetic analysis based on single-cell mutations from all biopsies can be performed.

[0153] In the case of metastases, biopsies from all tumors can be obtained and mutations identified by NGS that are common to all tumors can be selected.

[0154] 4. Using CTCs to Prioritize Antigens that Inhibit Metastasis Metastatic tumors are believed to originate from a single cell. Therefore, by genotyping circulating tumor cells (CTCs) in a patient, in conjunction with genotyping individual cells extracted from different tumors of a given patient, it is possible to reconstruct the evolutionary history of a cancer. It is expected that we will observe metastatic tumors evolving from the original tumor through clades of CTCs derived from the primary tumor.

[0155] In the following (An unbiased method to identify, count and gene probe CTCs), we describe 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, we can then reconstruct a phylogenetic tree of the primary tumor, the CTCs and secondary tumors arising from metastasis (if present). Based on this phylogenetic tree, mutations (passenger or driver) that occurred at or shortly after the time the CTCs first detached from the primary tumor can be identified. The genomes of CTCs arising from the primary tumor are expected to be evolutionarily more similar to the primary tumor genome than the secondary tumor genome. Furthermore, the genomes of CTCs arising from the primary tumor are expected to contain unique mutations that are fixed in the secondary tumor or are likely to be fixed if a secondary tumor is formed in the future. These unique mutations can be prioritized for IVAC to target (or prevent) metastasis.

[0156] The advantage of prioritizing CTC mutations versus primary basal mutations is that antigens derived from CTCs can recruit T cells to specifically target metastases, and therefore represent an independent weapon (using different antigens) from T cells targeting the primary tumor. In addition, when few (or no) secondary tumors are present, the likelihood of immune escape from CTC-derived antigens is expected to be lower, since the probability of tumor escape should scale with the number of cancer cells carrying a given antigen.

[0157] 5. Identification of antigens that co-occur on the same cells ("cocktail" IVAC) Tumors are thought to evolve to suppress mutations resulting from the selective pressures of the immune system and therapeutic approaches. Cancer vaccines that target multiple antigens that co-occur on the same cell and are also frequent in tumors have a greater chance of overriding tumor escape mechanisms, thus reducing the chance of relapse. Such "cocktail vaccines" are analogous to combination antiretroviral therapy for HIV-positive patients. Co-occurring mutations can be identified by phylogenetic analysis or by examining SNV alignments of all cells.

[0158] Further, according to the present invention, an unbiased method for identifying, enumerating and gene probing CTCs can be used. Such an approach can include the following steps: 1. Obtain a tumor biopsy and determine the somatic mutation map. 2. Option 1: Select N > 96 mutations for further consideration based on a previously established prioritization scheme. Option 2: Single-cell assays (see High-Throughput Whole-Genome Single-Cell Genotyping Methods above), then perform a phylogenetic analysis, selecting the primary base mutations and possibly the most recent mutations with N > 96 to maximize diversity. The former mutations are useful for identifying CTCs (see below) and the latter for performing the phylogenetic analysis (see chapter "Identification of Co-Occurring Antigens on the Same Cell ('Cocktail' IVAC)"). 3. Obtain whole blood from cancer patients. 4. Lyse red blood cells. 5.CD45 + The cells are depleted (eg, by sorting, magnetic beads coupled to anti-CD45 antibodies, etc.) to remove leukocytes and enrich for CTCs. 6. Remove free DNA by DNAase digestion. The source of free DNA can be DNA present in blood or DNA from dead cells. 7. Sort remaining cells into PCR tubes and perform STA (based on selected mutations) and screen with Fluidigm (high-throughput whole genome single cell genotyping method described above). CTCs should generally be positive for multiple SNVs. 8. Cells identified as cancerous (=CTCs) based on the panel of screened SNVs can then be further analyzed phylogenetically (see chapter "Identification of co-occurring antigens on the same cells ('cocktail' IVAC)").

[0159] This method can also 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 verified / profiled by Fluidigm / NGS to derive their mutations.

[0160] This method can be used to count CTCs, and since it is based on the mutational profile of cancer somatic mutations unique to a patient, rather than relying on one particular marker that may or may not be expressed by cancer cells, it is an unbiased method for detecting and counting CTCs.

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

[0162] In one embodiment of this approach, a pan-tumor phylogenetic analysis to recover driver mutations is performed.

[0163] For example, one can detect driver mutations from n=1 tumors.

[0164] In the chapter "Identification of Primary Basal Antigens Based on Rooted Tree Analysis" above, we describe a method to recover ancestral sequences and / or identify cells with sequences close to the root of the tree. Since by definition these are sequences close to the root of the tree, the number of mutations in these sequences is expected to be significantly lower than the number of mutations in the bulk cancer sample. Therefore, by selecting sequences close to the root of the tree, many passenger mutations are expected to be "phylogenetically filtered out". This procedure has the potential to greatly enrich for driver mutations. Driver mutations can then be used to identify / select patient treatments or can be used as leads for novel therapies.

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

[0166] By reconstructing the primary basal mutations from many tumors of a particular type, the chances of detecting driver mutations can be greatly increased. Basal sequences close to the tree root are expected to filter out many passenger mutations, greatly increasing the signal-to-noise ratio in detecting driver mutations. Therefore, this method has the potential to detect (1) less frequent driver mutations and (2) frequent driver mutations from fewer samples.

[0167] In another embodiment of the approach involving tumor phylogenetic reconstruction based on single-cell genotyping to enrich for driver mutations ("phylogenetic filtering"), a phylogenetic analysis is performed to recover metastases causing driver mutations.

[0168] In the chapter above, "Using CTCs to prioritize antigens that inhibit metastasis," we describe a method to detect CTC-associated mutations. This method can also be used to enrich for driver mutations that lead to metastasis. For example, by mapping the combined phylogeny of primer tumors, secondary tumors, and CTCs, CTCs derived from primary tumors should link between clades of primary and secondary tumors. Such phylogenetic analysis can help pinpoint unique mutations at this transition between primer tumors and secondary tumors. One fraction of these mutations may be driver mutations. Furthermore, by comparing unique CTC mutations from different cases of the same cancer (i.e., tumors with n>1), we can further enrich for unique driver mutations that lead to metastasis.

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

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

[0171] In the context of the present invention, the term "RNA" refers to a molecule that comprises ribonucleotide residues, preferably consisting entirely or substantially of ribonucleotide residues. "Ribonucleotide" refers to a nucleotide that has a hydroxyl group at the 2' position of a β-D-ribofuranosyl group. The term "RNA" includes recombinantly produced RNA, such as double-stranded RNA, single-stranded RNA, isolated RNA, such as partially or completely purified RNA, essentially pure RNA, synthetic RNA, and 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 the addition of non-nucleotide material, for example at one or more nucleotides of the RNA, for example at one or both ends or within the RNA. Nucleotides in an RNA molecule may also include non-standard nucleotides, such as non-naturally occurring 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" includes and preferably relates to "mRNA". The term "mRNA" means "messenger RNA" and relates to a "transcription product" that is produced by using a DNA template and codes for a peptide or protein. Typically, mRNA includes a 5'-UTR, a protein coding region and a 3'-UTR. mRNA has a limited half-life in cells and in vitro. In the context of the present invention, mRNA can be produced by in vitro transcription from a DNA template. 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 necessary.For example, RNA can be stabilized and its translation efficiency can be increased by one or more modifications that have the effect of stabilizing 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 the RNA used according to the present invention, within the coding region, i.e. within the sequence that codes for the peptide or protein to be expressed, the RNA can be modified to increase GC content to increase the stability of mRNA and to carry out codon optimization, thus enhancing translation in cells, preferably without changing the sequence of the peptide or protein to be expressed.

[0174] As used herein, the term "modification" in relation to RNA includes any modification of RNA that is not naturally occurring in said RNA.

[0175] In one embodiment of the invention, the RNA used according to the 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 modified ribonucleotides to increase its stability and / or reduce its cytotoxicity.For example, in one embodiment, in the RNA used according to the present invention, cytidine is partially or completely replaced with 5-methylcytidine, preferably completely replaced.Alternatively or additionally, in one embodiment, in the RNA used according to the present invention, uridine is partially or completely replaced with pseudouridine, preferably completely replaced.

[0177] In one embodiment, the term "modification" relates to providing an RNA with a 5'-cap or a 5'-cap analog. The term "5'-cap" refers to the cap structure found at the 5' end of an mRNA molecule, and generally consists of a guanosine nucleotide linked to the mRNA by a unique 5'-5' triphosphate bond. In one embodiment, the guanosine is methylated at position 7. The term "conventional 5'-cap" refers to the naturally occurring RNA 5'-cap, preferably a 7-methylguanosine cap (m 7 In the context of the present invention, the term "5'-cap" includes 5'-cap analogs that resemble an RNA cap structure and that are preferably modified to have the ability to stabilize the RNA when bound to the RNA and / or enhance translation of the RNA in vivo and / or in a cell.

[0178] Preferably, the 5' end of the RNA has the following general formula: [ka]

[0023] wherein R1 and R2 are independently hydroxy or methoxy, and W - , X - and Y - are independently oxygen, sulfur, selenium, or BH3. In a preferred embodiment, R1 and R2 are hydroxy and W - , X - and Y - is oxygen. In a further preferred embodiment, one of R1 and R2, preferably R1, is hydroxy and the other is methoxy, and W - , X - and Y - is oxygen. In a further preferred embodiment, R1 and R2 are hydroxy and W - , X - and Y - One of the following, preferably X -is sulfur, selenium or BH3, preferably sulfur, and the other is oxygen. In a further preferred embodiment, one of R1 and R2, preferably R2, is hydroxyl, the other is methoxy, and W - , X - and Y - One of the following, preferably X - is sulfur, selenium or BH3, preferably sulfur, and the others are oxygen.

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

[0180] W - , X - and Y - Cap structures having at least one of the substituents S is sulfur, i.e., phosphorothioate moieties, exist in various diastereoisomeric forms, all of which are encompassed herein. Additionally, the present invention encompasses all tautomers and stereoisomers of the above formula.

[0181] For example, R1 is methoxy, R2 is hydroxy, and X - is sulfur, and W - and Y - The cap structure having the above structure, where m is oxygen, exists in two diastereoisomeric forms (Rp and Sp). These can be separated by reversed-phase HPLC and are designated D1 and D2 according to their order of elution from a 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 an RNA with a 5'-cap or 5'-cap analog can be achieved by in vitro transcription of a DNA template in the presence of the 5'-cap or 5'-cap analog and co-transcriptionally incorporating the 5'-cap into the generated RNA strand, or the RNA can be generated, for example, by in vitro transcription, and the 5'-cap can be post-transcriptionally attached to the RNA using a capping enzyme, for example, vaccinia virus capping enzyme.

[0183] RNA may contain further modifications.For example, further modifications of RNA used in the present invention may be changes in 5'UTR or 3'UTR, such as the extension or truncation of naturally occurring poly(A) tail, or the introduction of untranslated regions (UTRs) not related to the coding region of said RNA, such as the replacement of existing 3'UTR with one or more, preferably two copies of 3'UTR from 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, preferably two copies of 3'UTR from said globin gene.

[0184] RNA with unmasked polyA sequences is translated more efficiently than RNA with masked polyA sequences. The term "poly(A) tail" or "polyA sequence" refers to a sequence of adenyl (A) residues typically located at the 3' end of an RNA molecule, and "unmasked polyA sequence" means that the polyA sequence at the 3' end of an RNA molecule ends with the A of the polyA sequence and is not followed by any nucleotides other than A located at the 3' end, i.e. downstream, of the polyA sequence. Furthermore, a long polyA sequence of about 120 base pairs provides optimal transcript stability and translation efficiency of RNA.

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

[0186] In addition, the incorporation of a 3' untranslated region into the 3' untranslated region (UTR) of an RNA molecule can result in increased translation efficiency. Synergistic effects can be achieved by incorporating two or more such 3' untranslated regions. The 3' untranslated regions can be autologous 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 combination of the above mentioned modifications, ie incorporation of a polyA sequence, unmasking of the polyA sequence and incorporation of one or more 3' untranslated regions, has a synergistic effect on increasing RNA stability and translation efficiency.

[0188] The term "stability" of an RNA relates to the "half-life" of the RNA. "Half-life" relates to the period required to remove half of the activity, amount or number of a molecule. In the context of the present invention, the half-life of an RNA is an indication of the stability of said RNA. The half-life of an RNA may affect the "duration of expression" of the RNA. An RNA with a long half-life can be expected to be expressed for a long period of time.

[0189] Needless to say, according to the present invention, if it is desired to reduce the stability and / or translation efficiency of the RNA, it is possible to modify the RNA so as to interfere with the function of the above-mentioned elements that increase the stability and / or translation efficiency of the RNA.

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

[0191] According to the present invention, the term expression also includes "ectopic expression" or "abnormal expression". "Ectopic expression" or "abnormal expression" means, according to the present invention, that expression is altered, preferably increased, compared to a reference, for example, the state of a subject that does not have a disease associated with ectopic or abnormal expression of a particular protein, for example, a tumor antigen. Increased expression refers to an increase of at least 10%, in particular at least 20%, at least 50% or at least 100% or more. In one embodiment, expression is only found in diseased tissues, while expression in healthy tissues is suppressed.

[0192] The term "specifically expressed" means that a protein is essentially expressed only in a particular tissue or organ. For example, a tumor antigen specifically expressed in the gastric mucosa means that said protein is primarily expressed in the gastric mucosa and not expressed in other tissues or to a significant extent in other tissues or organ types. Thus, a protein that is exclusively expressed in cells of the gastric mucosa and to a significantly lower extent in other tissues, such as the testis, is specifically expressed in cells of the gastric mucosa. In some embodiments, a tumor antigen may also be specifically expressed in more than one tissue type or organ, for example, two or three tissue types or organs, but preferably not more than three different tissues 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 preferably expressed to about the same extent in the lung and the stomach under normal conditions, said tumor antigen is specifically expressed in the lung and the stomach.

[0193] In the context of the present invention, the term "transcription" refers to the process in 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 the term "in vitro transcription" refers to a process in which RNA, in particular 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 called transcription vectors and are included in the term "vector" according to the present invention. According to the present invention, the RNA used in the present invention is preferably in vitro transcribed RNA (IVT RNA) and can be obtained by in vitro transcription of a suitable DNA template. The promoter for controlling the transcription can be any promoter for any RNA polymerase. Particular examples of RNA polymerases are T7, T3 and SP6 RNA polymerases. Preferably, the in vitro transcription according to the present invention is controlled by the T7 or SP6 promoter. The DNA template for in vitro transcription can be obtained by cloning a nucleic acid, in particular a 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" according to the present invention relates to the process in a cell's ribosomes in which a chain of messenger RNA directs the assembly of a sequence of amino acids to make a peptide or polypeptide.

[0195] According to the present invention, expression control sequences or regulatory sequences that can be functionally linked to a nucleic acid can be homologous or heterologous with respect to the nucleic acid. A coding sequence and a regulatory sequence are "functionally" linked together when they are covalently linked 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 a coding sequence and a regulatory sequence allows the coding sequence to be translated into a functional protein, induction of the regulatory sequence results in the transcription of the coding sequence without causing a reading frame shift of the coding sequence or without rendering the coding sequence unable to be translated into the desired protein or peptide.

[0196] The term "expression control sequence" or "regulatory sequence" includes, according to the present invention, promoters, ribosome binding sequences and other control elements that control the transcription of a nucleic acid or the translation of an induced RNA. In certain embodiments of the present invention, the regulatory sequence can be controlled. The exact structure of the regulatory sequence may vary depending on the species or cell type, but generally includes 5' non-transcribed sequences involved in the initiation of transcription or translation as well as 5' and 3' non-translated sequences, such as TATA boxes, capping sequences, CAAT sequences, etc. In particular, the 5' non-transcribed regulatory sequence includes a promoter region that includes a promoter sequence for transcriptional control of an operably linked gene. The regulatory sequence may also include an enhancer sequence or an upstream activating sequence.

[0197] Preferably, according to the invention, the RNA to be expressed in the cell is introduced into said cell. In one embodiment of the method according to the 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 "encoding RNA" are used interchangeably herein and mean, with respect to a particular peptide or polypeptide, that the RNA, when present in an appropriate environment, preferably within a cell, is capable of being expressed to produce said peptide or polypeptide. Preferably, the RNA according to the present invention is capable of interacting with the translation machinery of the cell to provide the peptide or polypeptide that the RNA is capable of expressing.

[0199] Terms such as "transfer", "introduce" or "transfect" are used interchangeably herein and relate to the introduction of nucleic acid, particularly exogenous or heterologous nucleic acid, particularly RNA, into cells. According to the present invention, the cells may form part of an organ, tissue and / or organism. According to the present invention, administration of the nucleic acid is accomplished as naked nucleic acid or in combination with an administration reagent. Preferably, administration of the nucleic acid is in the form of naked nucleic acid. Preferably, the RNA is administered in combination with a stabilizing agent, such as an RNase inhibitor. The present invention also envisages repeated introduction of the nucleic acid into cells to allow long-term sustained expression.

[0200] Any carrier can be used to transfect cells with which RNA can be bound, for example by forming a complex with RNA or by forming a vesicle in which RNA is enclosed or encapsulated, resulting in increased stability of RNA compared to naked RNA.Carriers useful 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 a complex with negatively charged nucleic acid.Any cationic lipid can be used according to the present invention.

[0201] Preferably, introduction of RNA encoding a peptide or polypeptide into a cell, particularly a cell present in vivo, results in expression of said peptide or polypeptide in the cell. In certain embodiments, targeting of nucleic acid to a specific cell is preferred. In such embodiments, the carrier (e.g., retrovirus or liposome) applied for administration of nucleic acid to a cell exhibits a targeting molecule. For example, a molecule such as an antibody specific for a surface membrane protein on the target cell or a ligand for a receptor on the target cell can be incorporated into or bound to the nucleic acid carrier. When the nucleic acid is administered by liposome, a protein that binds to a surface membrane protein associated with endocytosis can be incorporated into the liposomal formulation to enable targeting and / or uptake. Such proteins include capsid proteins or fragments thereof specific for a particular cell type, antibodies against proteins that are internalized, proteins that target intracellular locations, etc.

[0202] According to the present invention, the term "peptide" refers to a substance comprising 2 or more, preferably 3 or more, preferably 4 or more, preferably 6 or more, preferably 8 or more, preferably 10 or more, preferably 13 or more, preferably 16 or more, preferably 21 or more, and preferably 8, 10, 20, 30, 40 or 50, in particular up to 100 amino acids covalently linked by peptide bonds. The term "polypeptide" or "protein" refers to large peptides, preferably peptides having more than 100 amino acid residues, although in general the terms "peptide", "polypeptide" and "protein" are synonymous and are used interchangeably herein.

[0203] According to the present invention, the term "sequence changes" in relation to a peptide or protein relates to amino acid insertion variants, amino acid addition variants, amino acid deletion variants and amino acid substitution variants, preferably amino acid substitution variants. All these sequence changes according to the present invention can potentially generate new epitopes.

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

[0205] Amino acid addition variants include amino- and / or carboxy-terminal fusions of one or more amino acids, for example 1, 2, 3, 4 or 5 or more amino acids.

[0206] Amino acid deletion variants are characterized by the removal of one or more amino acids from the sequence, for example the removal of 1, 2, 3, 4 or 5 or more amino acids.

[0207] Amino acid substitution variants are characterized by at least one residue in the sequence being removed and another residue being inserted in its place.

[0208] The term "derived from" means according to the present invention that a particular entity, in particular a particular sequence, is present in the subject, in particular an organism or molecule, from which it originates. In the case of amino acid sequences, in particular a particular sequence region, "derived from" means in particular that the related amino acid sequence is derived from the amino acid sequence in which it is present.

[0209] The term "cell" or "host cell" preferably refers 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 carrying out its normal metabolic functions. Preferably, said term relates to any cell that can be transformed or transfected with an exogenous nucleic acid according to the present invention. The term "cell" according to the present invention encompasses prokaryotic cells (e.g. 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). The exogenous nucleic acid can be found inside the cell (i) freely distributed by itself, (ii) incorporated into a recombinant vector, or (iii) integrated into the host cell genome or mitochondrial DNA. Mammalian cells, such as cells from humans, mice, hamsters, pigs, goats and primates, are particularly preferred. The cells may be derived from many tissue types, including primary cells and cell lines. Particular examples include keratinocytes, peripheral blood leukocytes, bone marrow stem cells and embryonic stem cells. In further embodiments, the cells are antigen-presenting cells, in particular dendritic cells, monocytes or macrophages.

[0210] Cells containing a nucleic acid molecule preferably express the peptide or polypeptide encoded by said nucleic acid.

[0211] The term "clonal expansion" refers to the process by which a specific entity multiplies. In the context of the present invention, the term is preferably used in connection with an immune response in which lymphocytes are stimulated by an antigen and proliferate, amplifying the specific lymphocytes that recognize said antigen. Preferably, clonal expansion leads to differentiation of lymphocytes.

[0212] Terms such as "reduce" or "inhibit" relate to the ability to cause an overall decrease in the level, preferably by 5% or more, 10% or more, 20% or more, more preferably 50% or more, and most preferably 75% or more. "Inhibit" or similar phrases encompass complete or essentially complete inhibition, i.e., a reduction to zero or essentially zero.

[0213] Terms such as "increase", "enhance", "promote" or "prolong" preferably relate to an increase, enhancement, promotion or prolongation of approximately at least 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%, in particular at least 300%. These terms may also relate to an increase, enhancement, promotion or prolongation from zero or an unmeasurable or undetectable level to a level above zero or a measurable or detectable level.

[0214] The agent, composition and method described herein can be used to treat subjects with diseases, such as diseases characterized by the presence of abnormal cells that express antigens and present antigenic peptides.A particularly preferred disease is cancer disease.The agent, composition and method described herein can also be used for immunization or vaccination to prevent diseases described herein.

[0215] According to the present invention, the term "disease" refers to any pathological condition, including cancer diseases, particularly the forms of cancer diseases mentioned herein.

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

[0217] "Disease involving cells expressing an antigen" means, according to the present invention, that the expression of the 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 condition in a healthy tissue or organ. Increase refers to an increase of at least 10%, in particular at least 20%, at least 50%, at least 100%, at least 200%, at least 500%, at least 1000%, at least 10000% or even more. In one embodiment, the expression is only found in the diseased tissue, while the expression in healthy tissue is suppressed. According to the present invention, diseases involving or associated with cells expressing an antigen include cancer diseases.

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

[0219] Malignant tumor is essentially synonymous with cancer. Malignant disease, malignant neoplasm and malignant tumor are essentially synonymous with cancer.

[0220] According to the present invention, the term "tumor" or "tumor disease" refers to an abnormal proliferation of cells (called neoplastic, tumorigenic or tumor cells), preferably forming swellings or lesions. By "tumor cells" is meant abnormal cells that grow by rapid and uncontrolled cell proliferation and continue to grow after the stimuli that initiated the new growth have ceased. Tumors show a partial or complete lack of structural organization and functional coordination with normal tissues and usually form a distinct tissue mass that can be either benign, premalignant or malignant.

[0221] A benign tumor is one that lacks all three of the malignant characteristics of cancer. Thus, 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 mass of tissue that is the result of new formation. Neoplasia (from the Greek for new growth) is an abnormal proliferation of cells. The growth of the cells exceeds the growth of the normal tissues around them and is not coordinated with those tissues. The growth continues in the same excessive manner after the stimulus has stopped. This usually causes a lump or tumor. Neoplasms can be benign, premalignant or malignant.

[0223] "Tumor growth" or "tumor growth" 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 the present invention, the terms "cancer" and "cancer disease" are used interchangeably with the terms "tumor" and "tumor disease".

[0225] Cancers are classified according to the type of cells that the tumor resembles, and therefore the presumed tissue of origin: 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, blood 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 metastases thereof. Examples are lung carcinoma, breast carcinoma, prostate carcinoma, colon carcinoma, renal cell carcinoma, cervical carcinoma or metastases of the above mentioned cancer types or tumors. The term "cancer" according to the present invention also includes cancer metastasis and cancer recurrence.

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

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

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

[0230] "Bronchiole carcinoma" is a carcinoma of the lung thought to originate from the epithelium of the terminal bronchioles, where the neoplastic tissue extends along the alveolar walls and grows as nodules within the alveoli. Mucin may be evident in some of the cells and in the intra-alveolar material, including exfoliated cells.

[0231] "Adenocarcinoma" is a cancer that arises from glandular tissue. This tissue is also part of a larger tissue category known as epithelial tissue. Epithelial tissue includes skin, glands, and various other tissues that line the body cavities and organs of the body. Epithelium is derived embryologically from ectoderm, endoderm, and mesoderm. To be classified as an 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 originate, while those that are not well-differentiated may not. By staining the cells from the biopsy, the pathologist determines whether the 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 in the body. Although not every gland secretes the same substances, as long as there is an exocrine function to the cells, they are considered glands, and therefore the malignant form is named adenocarcinoma. Malignant adenocarcinomas invade other tissues and often metastasize if given enough time to do so. Ovarian adenocarcinoma is the most common type of ovarian cancer. It includes serous and mucinous adenocarcinoma, clear cell adenocarcinoma and endometrioid adenocarcinoma.

[0232] Renal cell carcinoma, also known as renal cell carcinoma or renal cell adenocarcinoma, is a kidney cancer that begins in the lining of the proximal tubules, the very small tubes in the kidney that filter blood and remove waste products. Renal cell carcinoma is by far the most common type of kidney cancer in adults and is the most deadly of all genitourinary 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 little finger-like projections (called papillae) in some, if not most, of the tumors.

[0233] Lymphomas and leukemias are malignant diseases that originate in hematopoietic (blood-forming) cells.

[0234] A blastic tumor or blastoma is a tumor (usually malignant) that resembles immature or embryonic tissue. 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, which depends on the detachment of malignant cells from the primary tumor, invasion of the extracellular matrix, penetration of the endothelial basement membrane to enter the body cavities and vessels, and then invasion of the target organ after being carried by blood. Finally, the growth of new tumors at the target site, i.e. secondary or metastatic tumors, depends on angiogenesis. Tumor metastasis often occurs even after removal of the primary tumor, because tumor cells or tumor components remain and can develop metastatic potential. In one embodiment, the term "metastasis" according to the present invention relates to "distant metastasis", which refers to metastasis far away 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 spreads to the liver, the secondary tumor will be made up of abnormal ovarian cells, not abnormal liver cells. The tumor in the liver will then be called metastatic ovarian cancer, not liver cancer.

[0237] In ovarian cancer, metastasis can occur: by direct contact or spread. Metastases can invade adjacent tissues or organs located near or around the ovaries, such as the fallopian tubes, uterus, bladder, rectum, etc.; by seeding or shedding into the peritoneal cavity, which is the most common way ovarian cancer spreads; by cancer cells breaking through the surface of the ovarian mass and "dropping" to other structures in the abdomen, such as the liver, stomach, colon, or diaphragm; by breaking off from the ovarian mass, invading the lymphatics, and then migrating to other areas of the body or distant organs, such as the lungs or liver; by breaking off from the ovarian mass, invading the blood system, and migrating to other areas of the body or distant organs.

[0238] According to the present invention, metastatic ovarian cancer includes cancer of the fallopian tubes, cancer in abdominal organs such as intestine cancer, uterine cancer, bladder cancer, rectum cancer, liver cancer, stomach cancer, colon cancer, diaphragm cancer, lung cancer, cancer in the 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 sites and / or several 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 that have separated from a primary tumor or tumor metastasis and circulate in the bloodstream. CTCs may 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 1-10 CTCs per mL of whole blood. Test methods have been developed to isolate CTCs. Several test methods for isolating CTCs have been described in the art, for example a technique that exploits 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 specimen with an antibody against EpCAM coupled to magnetic particles, followed by separation of the labeled cells in a magnetic field. The isolated cells are then stained with antibodies against another epithelial marker, cytokeratin, as well as the common leukocyte marker CD45, to distinguish rare CTCs from contaminating 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 that are made functional by coating with an antibody against EpCAM. CTCs are then stained with secondary antibodies against cytokeratin or tissue-specific markers, e.g., PSA in prostate cancer or HER2 in breast cancer, and visualized by automated scanning of the microposts in multiple planes along three-dimensional coordinates. The CTC chip can identify cytokeratin-positive circulating tumor cells in patients 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 blood tubes.The CellSearch® system is a US Food and Drug Administration (FDA) approved method for enumeration of CTCs in whole blood, and is based on a combination of immunomagnetic labeling and automated digital microscopy. There are other methods for isolating CTCs described in the literature, all of which can be used with the present invention.

[0240] Recurrence or regression occurs when a person re-injures a condition that he or she previously suffered from. For example, if a patient has had a tumor disease, has been successfully treated for said disease, but develops said disease again, the newly developed disease may be considered as recurrence or regression. However, according to the present invention, recurrence or regression of a tumor disease may also occur at the site of the original tumor disease, but this is not necessarily the case. Thus, for example, if a patient has had an ovarian tumor and has been successfully treated, recurrence or regression may be the development of an ovarian tumor or the development of a tumor at a site other than the ovary. Recurrence or regression of a tumor includes the situation where a tumor occurs at a site other than the site of the original tumor as well as the situation where a tumor occurs at the site of the original tumor. Preferably, the original tumor that the patient has been treated for is a primary tumor, and the tumor at a site other than the site of the original tumor is a secondary or metastatic tumor.

[0241] "Treat" refers to administering to a subject a compound or composition as described herein to prevent or eliminate disease, including reducing tumor size or number in the subject; to halt or slow the progression of disease in the subject; to inhibit or slow the onset of new disease in the subject; to reduce the frequency or severity of symptoms and / or recurrences in a subject who currently has or has previously had the disease; and / or to prolong or increase the survival of the subject. In particular, the term "treat disease" includes curing, shortening the duration, ameliorating, preventing, slowing or inhibiting the progression or worsening, or preventing or delaying the onset of the disease or its symptoms.

[0242] "At risk" refers to a subject, i.e., a patient, who is identified as having a higher than normal probability of developing disease, particularly cancer, compared with the general population.In addition, subjects who have had or currently have disease, particularly cancer, are subjects who are at high risk of developing disease, since such subjects may continue to develop disease.Subjects who currently have or have had cancer are also at high risk of cancer metastasis.

[0243] The term "immunotherapy" refers to a treatment involving the activation of a specific immune response. In the context of the present invention, terms such as "protect", "prevent", "prophylactic", "preventive" or "protective" refer to the prevention or treatment or both of the onset and / or spread of a disease in a subject, in particular to minimizing the possibility that the subject will develop a disease or to delay the onset of the disease. For example, as mentioned above, a person at risk of a tumor is a candidate for a treatment to prevent the tumor.

[0244] Prophylactic administration of immunotherapy, such as the compositions described herein, preferably protects the recipient from developing the disease. Therapeutic administration of immunotherapy, such as the compositions described herein, may result in the inhibition of disease progression / growth. This includes slowing down disease progression / growth, particularly stopping disease progression, preferably resulting in the elimination of the disease.

[0245] Immunotherapy may be performed using any of a variety of techniques in which the agents 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 specific to an antigen or cells expressing the antigen.

[0246] Within certain aspects, immunotherapy can be active immunotherapy, in which case the treatment is based on the in vivo stimulation of the endogenous host immune system to react against abnormal cells by administration of immune response modifiers (such as the polypeptides and nucleic acids provided herein).

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

[0248] The term "immunization" or "vaccination" refers to the process of treating a subject with the goal of inducing an immune response for therapeutic or prophylactic reasons.

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

[0250] The terms "subject", "individual", "organism" or "patient" are used interchangeably and relate to vertebrate animals, preferably mammals. For example, in the context of the present invention, mammals are 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., as well as captive animals, such as zoo animals. The term "animal" as used herein also includes humans. The term "subject" may also include patients, i.e. animals, preferably humans with a disease, preferably humans with a disease as described herein.

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

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

[0253] As part of a composition for immunization or vaccination, one or more agents described herein are preferably administered with one or more adjuvants to induce or increase an immune response. The term "adjuvant" refers to a compound that prolongs or enhances or promotes an immune response. The composition of the present invention preferably exerts its action without the addition of an adjuvant. Nevertheless, the composition of the present application may contain any known adjuvant. Adjuvants include a heterogeneous group of compounds such as oil emulsions (e.g. Freund's adjuvant), inorganic compounds (such as alum), bacterial products (such as Bordetella pertussis toxin), liposomes and immune stimulating complexes. Examples of adjuvants are monophosphoryl lipid A (MPL SmithKline Beecham), saponins such as QS21 (SmithKline Beecham), DQS21 (SmithKline Beecham; WO 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 may 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 immune responses and can therefore be used in vaccination, including costimulatory molecules provided in protein or nucleic acid form, such as B7-1 and B7-2 (CD80 and CD86, respectively).

[0256] According to the present invention, a "tumor specimen" is a body sample that contains tumor cells, such as circulating tumor cells (CTCs), or cancer cells, in particular tissue samples and / or cell samples that contain body fluids. According to the present invention, a "non-tumorigenic specimen" is a body sample that does not contain tumor cells, such as circulating tumor cells (CTCs), or cancer cells, in particular tissue samples and / or cell samples that contain body fluids. Such body samples may be obtained in conventional ways, for example by tissue biopsy, including punch biopsy, and by taking blood, bronchial aspirate, sputum, urine, feces or other body fluids. According to the present invention, the term "sample" also encompasses processed samples, such as fractions or isolates of biological samples, for example nucleic acid or cell isolates.

[0257] The therapeutically active substances, vaccines and compositions described herein may be administered by any conventional route, including injection or infusion. Administration may be performed, for example, orally, intravenously, intraperitoneally, intramuscularly, subcutaneously or transdermally. In one embodiment, administration is performed intranodally, such as by injection into a lymph node. Another form of administration envisages in vitro transfection of antigen-presenting cells, such as dendritic cells, with the nucleic acid described herein, followed by administration of the antigen-presenting cells.

[0258] The agent described herein is administered in an effective amount. "Effective amount" refers to an amount that achieves a desired reaction or a desired effect, either alone or together with further administration. In the case of treating a particular disease or a particular condition, the desired reaction is preferably related to the inhibition of the progression of the disease. This includes slowing down the progression of the disease, in particular preventing or reversing the progression of the disease. In the treatment of a disease or condition, the desired reaction can also be slowing down or preventing the onset of said disease or said condition.

[0259] The effective amount of the agent described herein depends on the condition to be treated, the severity of the disease, the individual parameters of the patient, including age, physiological state, size and weight, duration of treatment, type of concomitant treatment (if any), specific route of administration and similar factors.Therefore, the dose of the agent described herein to be administered may depend on such various parameters.If the initial dose is insufficient in the patient, 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 the therapeutically active material to produce a desired reaction or a desired effect.

[0261] The pharmaceutical compositions described herein are generally administered in pharma- ceutical compatible amounts and in pharma-ceutical compatible preparations. The term "pharma-ceutical compatible" refers to non-toxic substances that do not interact with the action of the active ingredients of the pharmaceutical composition. Such preparations may usually contain supplementary immune enhancing substances such as salts, buffer substances, preservatives, carriers, adjuvants, e.g., CpG oligonucleotides, cytokines, chemokines, saponins, GM-CSF and / or RNA, and, where appropriate, other therapeutically active compounds. When used in medicines, the salts should be pharma-ceutical compatible. However, salts that are not pharma-ceutical compatible may be used to prepare pharma-ceutical compatible salts and are encompassed by the present invention. Such pharmacologically and pharma-ceutical compatible 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. Pharmaceutically compatible salts can also be prepared as alkali metal or alkaline earth metal salts, such as sodium salts, potassium salts or calcium salts.

[0262] The pharmaceutical compositions described herein may contain a pharma- ceutical compatible carrier. The term "carrier" refers to a natural or synthetic organic or inorganic component with which active ingredients are combined to facilitate application. According to the present invention, the term "pharma-ceutical compatible carrier" includes one or more compatible solid or liquid fillers, diluents or encapsulating substances that are suitable for administration to a patient. The components of the pharmaceutical compositions described herein are usually those that do not interact with each other in a manner that would substantially impair the desired pharmaceutical effect.

[0263] The pharmaceutical compositions described herein may contain suitable buffer substances, such as, for example, acetic acid in a salt, citric acid in a salt, boric acid in a salt and phosphoric acid in a salt.

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

[0265] The pharmaceutical compositions are usually presented in unit dosage form and may be prepared in a manner known per se. The pharmaceutical compositions described herein may, for example, be in the form of capsules, tablets, lozenges, solutions, suspensions, syrups, elixirs or emulsions.

[0266] The composition suitable for parenteral administration usually comprises a sterile aqueous or non-aqueous preparation of the active compound, which is preferably isotonic with the blood of the recipient.Examples of suitable carriers and solvents are Ringer's solution and isotonic sodium chloride solution.In addition, sterile fixed oils are usually used as a solution or suspension medium.

[0267] The present invention will be described in detail by the following figures and examples, which are used for illustration only and are not intended to be limiting. The descriptions and examples make further embodiments, which are also encompassed by the present invention, accessible to those skilled in the art. EXAMPLES

[0268] The techniques and methods used herein are as described herein or known per se and are described, for example, in Sambrook et al., Molecular Cloning: A Laboratory Manual, 2 nd Edition (1989) Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY All procedures, including the use of kits and reagents, are performed according to the manufacturer's information unless otherwise indicated.

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

[0270] Mutation detection Rationale for using CTCs: Detection of circulating tumor cells (CTCs) from the peripheral blood of cancer patients is a widely accepted independent prognostic marker for the clinical course of the tumor (Pantel et al, Trends Mol Med 2010;16(9):398-406). Over the years, the clinical importance of CTCs has been the subject of intense academic and clinical research in oncology. Detection of CTCs in the blood of patients with metastatic breast, prostate and colorectal cancer has been shown to be prognostic and provide additional information to conventional imaging techniques and other prognostic tumor biomarkers. Serial blood samples taken from patients before, during early stages and after treatment with therapeutic drugs (systemic or targeted) provide information on the response / failure of the treatment. Molecular analysis of drug-resistant CTCs may provide further insight into the resistance mechanisms (e.g. mutations in specific signaling pathways or loss of target expression) in individual patients. A further possibility 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 termed "liquid tumor biopsy." This profiling is rapid, repeatable, requires only the patient's blood, and does not require surgery, providing a "real-time" indication of the tumor's status.

[0271] Mutations from tumor cells: We demonstrate our ability to identify mutations using B16 melanoma cells, exome capture to extract protein coding regions, next generation sequencing with our HiSeq 2000, and then bioinformatics analysis with our "iCAM" software pipeline (Figure 1). We identified 2448 nonsynonymous mutations and selected 50 for confirmation. All 50 somatic mutations could be confirmed.

[0272] Below is an example of the protein effect of a somatic mutation discovered in B16 melanoma cells: [ka]

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

[0274] We profiled eight individual CTCs and identified somatic mutations. Furthermore, in eight of the eight cells, we identified previously identified somatic mutations. In multiple cases, the data showed heterogeneity at the individual cell level. For example, at position 144078227 on chromosome 2 (assembly mm9), in 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, a fundamental path towards "real-time" iVAC (personalized vaccines), where patients are repeatedly profiled and the results reflect the current patient state rather than the state of the patient at an earlier time point. Furthermore, it reveals that we can identify heterogeneous somatic mutations present in subsets of tumor cells, allowing assessment of mutation frequency, e.g., to identify major and rare mutations.

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

[0277] Fluorescently labeled B16 melanoma cells were used to generate circulating tumor cells (CTCs). 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 samples were mixed gently by vortexing and then incubated at room temperature for 10 min. To stop the labeling reaction, an equal volume of PBS containing 20% ​​FSC was added to the samples and mixed gently by vortexing. After 20 min of incubation at room temperature, the cells were washed twice using PBS. Finally, the cells were resuspended in PBS and injected intravenously (iv) into mice. After 3 min, the mice were sacrificed and blood was collected.

[0278] Red blood cells from blood samples were lysed 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 samples and incubated for 5 min at room temperature. Following incubation, two washes were performed and the samples were resuspended in 500 μl of PBS.

[0279] CFSE-labeled circulating B16 cells were sorted on an Aria I cell sorter (BD). Single cells were sorted onto 96-well v-bottom plates prepared with 50 μl / well of RLT buffer (Quiagen). After sorting was completed, plates were 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] For each sorted CTC, RNA was extracted and SMART-based cDNA synthesis and non-specific amplification was performed. RNA from sorted CTC cells was extracted with 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 a long oligo(dT)-T-primer for priming the first strand synthesis reaction and TS-short (Eurogentec SA, Seraing, Belgium) that introduces an oligo(riboG) sequence to allow generation of an extended template by the terminal transferase activity of reverse transcriptase and template switching [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. PLJ Siebert, ed. BioTechniques Books, MA, Natick. 305-319]. First-strand cDNA synthesized according to the manufacturer's instructions was subjected to 35 cycles of amplification 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 s at 94°C, 30 s at 65°C, 1 min at 72°C, final extension at 72°C for 6 min). Specific primers monitoring actin and GAPDH were used to control the success of the amplification of CTC genes.

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

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

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

[0285] All cleanups were performed using 1.8 volumes of AMPure XP magnetic beads (Agencourt). All quality controls were performed using the Qubit HS assay from Invitrogen, and fragment sizes were determined using the Agilent 2100 Bioanalyzer HS DNA assay.

[0286] Exome-enriched gDNA libraries were clustered on a cBot with Truseq SR Cluster Kit v2.5 using 7 pM and 50 bp were sequenced on an Illumina HiSeq2000 using Truseq SBS Kit-HS 50 bp.

[0287] Next-generation sequencing, RNA sequencing (RNA-Seq): A barcoded mRNA-seq cDNA library was prepared from 5 μg of total RNA using a modified version of the 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, resulting in fragments ranging from 160 to 220 bp. The fragmented mRNA was converted to cDNA using random primers and SuperScriptII (Invitrogen), and then the second strand was synthesized using DNA polymerase I and RNaseH. The cDNA was end-repaired using T4 DNA polymerase, Klenow DNA polymerase, and 5' phosphorylated using T4 polynucleotide kinase. Blunt-ended cDNA fragments were 3' adenylated (3'-5' exo minus) using Klenow fragment. Single 3′T-overhang Illumina multiplex-specific adaptors were ligated at a 10:1 molar ratio of adaptor to cDNA insert using T4 DNA ligase.

[0288] cDNA libraries were purified and size selected at 200-220 bp using E-Gel 2% SizeSelect gels (Invitrogen). Enrichment, addition of Illumina 6-base index sequences and flow cell-specific sequences were performed by PCR using Phusion DNA polymerase (Finnzymes). All cleanup was performed using 1.8x volumes of Agencourt AMPure XP magnetic beads. All quality controls were 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 on a cBot with Truseq SR Cluster Kit v2.5 using 7 pM and 50 bp were sequenced on an Illumina HiSeq2000 using Truseq SBS Kit-HS 50 bp.

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

[0291] NGS data analysis, gene expression: To determine expression values, sequence reads from RNA samples output from Illumina HiSeq 2000 were preprocessed according to Illumina standard protocols, including filtering and demultiplexing low-quality reads. For RNA-Seq transcriptome analysis, sequence reads were aligned to the reference genome sequence using 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], using the "-v2-best" parameter for genome alignment and default parameters for transcript alignment [Mouse Genome Sequencing Consortium. Initial sequencing and comparative analysis of the mouse genome. Nature, 420, 520-562 (2002)]. Alignment coordinates were compared to 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 overlap alignment counts were recorded for each transcript. Sequence reads that were not alignable to the genome sequence were aligned to a database of all possible exon-exon junction sequences of RefSeq transcripts.The counts of reads aligned to splice junctions were summed with the respective transcript counts obtained in the previous step and normalized to the RPKM (number of reads which map per kilobase of exon model per million mapped reads) for each transcript [Mortazavi, A. et al. (2008). Mapping and quantifying mammalian transcriptomes by rna-seq. Nat Methods, 5(7):621-628]. Both gene and exon expression values ​​were calculated based on the normalized number of reads overlapping each gene or exon, respectively.

[0292] Mutation discovery, bulk tumors: 50 nucleotide single-end reads from an Illumina HiSeq 2000 were aligned to the 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] using default options. Ambiguous reads - reads that map to multiple locations in the genome - were removed, and the remaining alignments were curated, indexed, converted to binary compressed format (BAM), and read quality scores were converted from Illumina standard phred+64 to standard Sanger quality scores using a shell script.

[0293] For each sequencing lane, mutations were identified using three software programs, including 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 options and filtering criteria recommended by the authors were used, including first filtering, maximum coverage of 200. For the second round of samtools filtering, the indel minimum quality score was 50 and the point mutation minimum quality was 30. For GATK mutation calling, we followed the author-designed best practice guidelines 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 a hard filtering option. For SomaticSniper mutation calling, we used the default option and only predicted mutations with a "somatic score" of 30 or higher were considered further.

[0294] As per the mutation discovery, CTC:bulk tumor iCAM process, 50 nucleotide single-end reads from the Illumina HiSeq 2000 were aligned to the reference mouse genome assembly mm9 using bwa (version 0.5.8c[5]) with default options. As the CTC NGS reads were derived from the RNA-Seq assay, the reads were also aligned to the transcriptome sequence including exon-exon junctions using bowtie (mentioned above). All alignments were used to compare the nucleotide sequences from the reads to both the reference genome and the B16 mutations from the bulk tumor. The identified mutations were assessed manually using perl scripts as well as using the samtools software program and IGV (Integrated Genome Viewer) to visualize the results.

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

[0296] Mutation prioritization Next, we identify a potential mutation prioritization pipeline for vaccine inclusion. This method, called the "individualized cancer mutation detection pipeline" (iCAM), identifies and prioritizes somatic mutations through a series of steps that incorporate multiple state-of-the-art algorithms and bioinformatics methods. The output of this step is a list of somatic mutations, prioritized based on their likely immunogenicity.

[0297] Somatic mutation identification: For both B16 and Black6 samples, mutations are identified using three different algorithms (mutation discovery, above). The first iCAM step is to combine the output list from each algorithm to create a high-confidence list of somatic mutations. GATK and samtools report mutations in one sample compared to the reference genome. To select high-confidence mutations with few false positives for a given sample (i.e. tumor or normal), we select mutations identified in all replicates. We then select mutations that are present in the tumor sample but not in the normal sample. SomaticSniper automatically reports potential somatic mutations from tumor and normal data pairs. We further filtered the results through the intersection of results 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 is to assign a confidence value (p-value) to each mutation based on the depth of coverage, SNP quality, consensus quality and mapping quality.

[0298] Mutation Impact: The impact of the filtered consensus somatic mutations is determined by a script within the iCAM mutation pipeline. First, sequence reads that align to multiple sites are filtered out, so that mutations occurring in non-unique genomic regions within the genome, such as occurs for some protein paralogs and pseudogenes, are excluded from the analysis. Second, it is determined whether the mutation occurs in the transcript. Third, it is determined whether the mutation occurs within a protein coding region. Fourth, the transcript sequence is translated with and without the mutation to determine whether an amino acid sequence change is present.

[0299] Expression mutations: The iCaM pipeline selects somatic mutations found in genes and exons that are expressed in tumor cells. Expression levels are determined through NGS RNA-Seq of tumor cells (above). The number of reads overlapping genes and exons indicates expression levels. 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 potential for epitopes containing mutant peptides to bind MHC molecules, the iCAM pipeline runs a modified version of 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 the B16 and Black6 data, predictions were performed using all available black6 MHC class I alleles and all epitopes for each peptide length. We select mutations contained in epitopes that rank in the 95th percentile of the prediction score distribution of the IEDB training data (http: / / mhcbindingpredictions.immuneepitope.org / dataset.html) and consider all MHC alleles and all potential epitopes that overlap the mutation.

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

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

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

[0304] For confirmation of mutations by resequencing, the genomic region containing the mutation was amplified by standard PCR from 50 ng of both tumor and normal control DNA. The size of the amplified products ranged from 150 to 400 nucleotides. The specificity of the reaction was controlled by loading the PCR products on a Qiaxel instrument (Qiagen). The PCR products were purified using the minElute PCR purification kit (Qiagen). The specific PCR products were sequenced using standard Sanger sequencing methods (Eurofins) followed by electropherogram analysis.

[0305] Mutation confirmation was also achieved through examination of tumor RNA. Tumor gene and exon expression values ​​were generated from RNA-Seq (NGS of RNA), which generates nucleotide sequences that are mapped to transcripts and counted. 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 the identified somatic mutations derived from DNA. Table 1: List of genes containing the 50 documented mutations 50 identified genes with confirmed somatic mutations, with annotation of gene symbol, gene name, and predicted location and function [Table 1]

[0306] Example 2 The IVAC selection algorithm enables detection of immunogenic mutations To test whether specific T cell responses could be induced against the identified mutations from B16F10 melanoma cells, naïve C57BL / 6 mice (n=5 / peptide) were immunized subcutaneously twice (days 0 and 7) with 100 μg peptides (+ 50 μg PolyI:C as adjuvant) containing either the mutated or wild-type amino acid sequence (see Table 2). All peptides were 27 amino acids in length and had the mutated / wild-type amino acid in the central position. Mice were sacrificed on day 12 and splenocytes were harvested. As a readout, 5 × 10 5 Splenocytes / well were used as effectors and 5 × 10 4 IFNγ ELISpot was performed using bone marrow dendritic cells as target cells. Effector splenocytes were tested against mutant, wild-type and control peptides (vesicular stomatitis virus nucleoprotein, VSV-NP).

[0307] Of the 44 sequences tested, we found that 6 of these induced T cell immunity only against the mutant sequence but not against the wild-type peptide (Figure 3).

[0308] The data demonstrate that the 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: List of mutant sequences that induced specific T cell reactivity to mutant versus wild-type peptides. Amino acid exchanges are underlined. [Table 2]

[0309] Example 3 The identified mutations can provide therapeutic antitumor immunity To test whether the identified mutations have the potential to confer antitumor immunity following vaccination in naive mice, this question was addressed using a peptide for mutation number 30 that was shown to induce mutation-selective T cell reactivity. B16F10 cells (7.5 × 10 4 ) subcutaneously on day 0. Mice were vaccinated with peptide 30 (see Table 1; peptide 100 μg + PolyI:C 50 μg sc) on days -4, +2 and +9. The control group received Poly I:C (50 μg sc) alone. Tumor growth was monitored every other day. On day +16, only 1 of 5 mice in the peptide vaccine group developed a tumor, whereas 4 of 5 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). Because B16F10 is a highly aggressive tumor cell line, the observation that the method applied to identify and prioritize mutations ultimately led to the selection of mutations that are already potent as vaccines in their own right is an important proof of concept for the entire process.

[0311] Example 4 Data supporting polyepitopic antigen presentation The verified mutations from the patient's protein coding regions constitute a pool from which candidates can be selected for the assembly of polyneoepitope vaccine templates to be used as precursors for GMP production of RNA vaccines.Vector cassettes suitable as vaccine backbones have been previously 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). The preferred vector cassette is modified in the coding and untranslated regions (UTRs) to ensure maximized translation of the encoded protein over a long period of time (Holtkamp, ​​S. et al., Blood, 108: 4009-4017, 2006; Kuhn, ANet al., Gene Ther., 17: 961-971, 2010). In addition, the vector backbone contains an antigen routing module for the simultaneous expansion of cytotoxic T cells as well as 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 demonstrate 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 mutations in the center. These sequences are linked head to tail by short linkers to form polyneoepitope vaccines encoding up to 30 or more selected mutations and their flanking regions. These patient-specific, tailored inserts are codon optimized and cloned into the RNA backbone described above. Quality control of such constructs includes in vitro transcription and expression in cells for validation of functional transcription and translation. Analysis of translation is performed using an antibody against the C-terminal targeting 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 contiguously arranged sequences encoding mutant peptides, connected by linker sequences (see Figure 6). The coding sequences are selected from nonsynonymous mutations and are always constructed with the codon for the mutated amino acid flanked by a region of 30-75 base pairs from the original sequence context. The linker sequences code for amino acids that are preferentially not processed by the cellular antigen processing machinery.

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

[0315] For the first proof of concept, a bi-epitope vector, i.e., a vector encoding one polypeptide containing two mutated epitopes, was used. Codon-optimized sequences were designed encoding (i) a mutated epitope of 20–50 amino acids, (ii) a glycine / serine-rich linker, (iii) a second mutated epitope of 20–50 amino acids, and (iv) an additional glycine / serine-rich linker, flanked by appropriate recognition sites for restriction endonucleases for cloning into a pST1-based construct as described above, and were synthesized by a commercial supplier (Geneart, Regensburg, Germany). After sequence verification, these were cloned into a pST1-based vector backbone, resulting in the construct shown in Figure 6.

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

[0317] To prove that sequences incorporating mutant amino acids and flanked at 5' and 3' by linker sequences can be processed, presented and recognized by antigen-specific T cells, we used T cells from peptide-vaccinated mice as effector cells. In IFNγ ELISpot, we tested whether T cells induced by peptide vaccination as described above could recognize target cells (myeloid dendritic cells, BMDCs) pulsed with peptide (2 μg / ml, 2 h at 37° C. and 5% CO2) or transfected with RNA (20 μg generated as described above) by electroporation. As illustrated in FIG. 7 for mutations 12 and 30 (see Table 2), we could observe that the RNA constructs were able to generate epitopes recognized by mutant-specific T cells.

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

[0319] Example 6 Design of polyneoepitope vaccines - the relevance of linkers The polyneoepitope RNA construct comprises a backbone construct in which multiple somatically mutated coding peptides are arranged linked by linker peptide sequences. In addition to codon optimization and increased RNA stability and translation efficiency through the backbone, one embodiment of the RNA polyneoepitope vaccine comprises a linker designed to increase MHC class I and II presentation of antigenic peptides and decrease presentation of harmful epitopes.

[0320] Linker: A linker sequence was designed to link multiple mutation-containing peptides. The linker should allow for the generation and presentation of mutant epitopes and prevent the generation of harmful epitopes such as those generated at the junction sutures between adjacent peptides or between the linker sequence and endogenous peptides. These "junction" epitopes may not only compete with the intended epitopes to be presented on the cell surface, reducing the efficacy of the vaccine, but also may result in undesirable autoimmune reactions. Therefore, we designed the linker sequence to a) avoid generating "junction" peptides that bind to MHC molecules, b) avoid proteasomal processing that generates "junction" peptides, and c) be efficiently translated and processed by the proteasome.

[0321] To avoid the generation of "junction" peptides that bind to MHC molecules, we compared various linker sequences. For example, glycine inhibits strong binding at the MHC binding groove [Abastado JP.et al.,J Immunol.1993 Oct 1;151(7):3569-75]. We considered multiple linker sequences and multiple linker lengths and calculated the number of "junction" 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 MHC class I molecules.

[0322] In the B16 model, we identified 102 expressed nonsynonymous somatic mutations predicted to be presented on MHC class I molecules. The 50 confirmed mutations were used to computationally design various vaccine constructs, including the use of no linker or different linker sequences, and the number of deleterious "junction" peptides was computationally calculated using the IEDB algorithm (Figure 8).

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

[0324] Linker 1 and Linker 2, used in experimentally tested RNA polyneoepitope vaccine constructs (see below), also predicted a good low number of junctional neoepitopes, and this was also true for the predictions of the nonamers and decamers.

[0325] This reveals that the sequence of the linker is crucial for the generation of unwanted MHC-binding epitopes. Moreover, the length of the linker sequence influences the number of unwanted MHC-binding epitopes. We find that G-rich sequences hinder the generation of MHC-binding ligands. Table 3: Influence of the linker (10 amino acid epitope). For each peptide linker, the predicted number of unwanted epitopes, defined as MHC class I binding epitopes that contain junction sequences. Here, 10 amino acid epitopes are considered. The glycine-rich linker has the fewest junction epitopes. [Table 3] Table 4: Influence of the linker moiety (9 amino acid epitopes). For each peptide linker, the predicted number of unwanted epitopes, defined as MHC class I binding epitopes that contain junction sequences. Here, 9 amino acid epitopes are considered. The glycine-rich linker has the fewest junction epitopes. [Table 4] Table 5: Influence of the linker moiety. Predicted number of unwanted epitopes, defined as MHC class I binding epitopes containing junction sequences, for each peptide linker. Here, 9 amino acid epitopes are considered. Top: Number of 9 amino acid junction epitopes for no linker and for each of the 5 different linkers. Middle: Number of 10 amino acid junction epitopes for no linker and for each of the 5 different linkers. Bottom: Number of 99 amino acid junction epitopes for similar linkers of different lengths. Glycine-rich linkers have the fewest junction epitopes. [Table 5]

[0326] To avoid proteasomal processing that may generate "junction" peptides, we considered using different amino acids in the linker. Glycine-rich sequences reduce proteasomal 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]. Thus, glycine-rich linker sequences serve 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]; their inclusion 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 allow for more efficient translation and processing by the proteasome, which in turn should allow better access to the encoded antigenic peptide.

[0328] Thus, the linker should be glycine-rich to prevent the generation of undesired epitopes that bind to MHC, hinder the ability of the proteasome to process the linker peptide (which can be achieved by including glycine), and be flexible to increase access to the mutation-containing peptide (which can be achieved by a combination of glycine and serine amino acids).Therefore, 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, tandem β-globin 3'UTR sequence and a 120bp poly(A) tail, which have been shown to increase the stability and translation efficiency of RNA, thereby enhancing the T cell stimulating ability of the encoded antigen (Holtkamp S. et al., Blood 2006; PMID:16940422). In addition, a transmembrane and cytosolic domain containing a termination codon (MHC class I transport signal or MITD) adjacent to a polylinker sequence for cloning the epitope was inserted (Kreiter S. et al., J.Immunol.,180:309-318,2008). The latter enhances antigen presentation, thereby enhancing antigen-specific CD8 + and has been shown to enhance CD4+ T cell proliferation and improve effector function.

[0330] Three RNA constructs were generated to provide RNA polyneoepitope constructs for the 50 identified and verified mutations of B16F10. The constructs consist of codon-optimized sequences encoding (i) the 25 amino acid mutated epitope, (ii) a glycine / serine-rich linker, and (iii) a repeat of the mutated epitope sequence followed by a glycine / serine-rich linker. The mutated epitope-containing sequence and the linker strand are flanked by appropriate recognition sites for restriction endonucleases for cloning into the pST1-based construct described above. Vaccine constructs were designed and synthesized by GENEART. After sequence verification, they were cloned into the pST1-based vector backbone to obtain the RNA polyneoepitope vaccine constructs.

[0331] Description of clinical approach Clinical application involves the following steps: Eligible patients must consent to DNA analysis by next-generation sequencing. Tumor specimens (paraffin-embedded, formalin-fixed tissue) and peripheral blood cells obtained from routine diagnostic procedures are obtained and used for mutation analysis as described above. · Verify the mutations found. Based on the prioritization, vaccines will be designed. For RNA vaccines, master plasmid templates will be generated by gene synthesis and cloning. The plasmids are used for the production of clinical grade RNA, quality control and release of RNA vaccines. · Sending the vaccine drug products to the respective clinical trial sites for clinical application. RNA vaccines can be used as naked vaccines in formulation buffer or encapsulated in nanoparticles or liposomes for direct injection, e.g., into lymph nodes, subcutaneous, intravenous, intramuscular injection, or for in vitro transfection of dendritic cells for adoptive transfer.

[0332] The entire clinical process takes less than six weeks. The "lag period" between patient informed consent and drug availability is carefully addressed by the clinical trial protocol, including allowing the standard treatment regimen to be continued until the investigational drug is available.

[0333] Example 8 Identification of tumor metastases and their use for tumor vaccination We applied NGS exome resequencing for mutation discovery in the B16F10 mouse melanoma cell line and identified 962 nonsynonymous somatic point mutations, 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 controlling 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 Trrap, previously described as frequently altered in human melanoma.

[0334] The immunogenicity and specificity of the 50 validated mutations were assayed 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% elicited an immune response preferentially against the mutant sequence compared to the wild-type sequence.

[0335] We tested the hypothesis in tumor xenograft models: immunization with the peptide conferred in vivo tumor suppression in protective and therapeutic settings, and mutant epitopes containing single amino acid substitutions were identified as effective vaccines.

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

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

[0338] Next-generation sequencing Nucleic Acid Extraction and Sample Preparation: DNA and RNA from bulk B16F10 cells and DNA from C57BL / 6 tail tissue were extracted in triplicate using Qiagen DNeasy Blood and Tissue kits (for DNA) and Qiagen RNeasy Micro kits (for RNA).

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

[0340] RNA gene expression, "transcriptome" profiling (RNA-Seq): Barcoded mRNA-seq cDNA libraries were prepared in triplicate 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 SuperScriptII (Invitrogen), and then the second strand was synthesized using DNA polymerase I and RNaseH. cDNA was end-repaired, 5' phosphorylated, and 3' adenylated according to the manufacturer's instructions. Single 3' T-overhanging Illumina multiplex-specific adapters were ligated with T4 DNA ligase using a 10:1 molar ratio of adapter to cDNA insert. cDNA libraries were purified and size-selected at 200-220 bp (E-Gel 2% SizeSelect gel, Invitrogen). Enrichment, addition of Illumina 6-base index sequences and flow cell-specific sequences were performed by PCR using Phusion DNA polymerase (Finnzymes). All cleanup up to this step was performed with 1.8x volume of Agencourt AMPure XP magnetic beads. All quality controls were performed using Invitrogen's Qubit HS assay and fragment sizes were determined using Agilent's 2100 Bioanalyzer HS DNA assay. Barcoded RNA-Seq libraries were clustered and sequenced as described above.

[0341] NGS data analysis, gene expression: Outputted sequence reads from RNA samples were preprocessed according to Illumina standard protocols, 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 alignments, two mismatches were allowed and only the best alignment ("-v2-best") was recorded; for transcriptome alignments default parameters were used. Reads not alignable to genome sequences 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 values ​​were determined by intersecting read coordinates with those of RefSeq transcripts, and overlapping exon 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, somatic mutation discovery: Somatic mutations were identified as described in Example 9. Single-end reads of 50 nucleotides were aligned to the mm9 reference mouse genome using bwa (default options, 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). Potential mutations identified in all B16F10 triplicates were assigned a "false discovery rate" (FDR) confidence value (see Example 9).

[0343] Mutation selection, validation and function Selection: Mutations had to meet the following criteria to be selected: (i) present in all B16F10 triplets and absent in all C57BL / 6 triplets, (ii) FDR≦0.05, (iii) uniform in C57BL / 6, (iv) occurring in RefSeq transcripts, and (v) causing a nonsynonymous change that was scored as a true mutation. Selection for validation and immunogenicity testing required that the mutation was in an expressed gene (average RPKM>10 across replicates).

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

[0345] Functional impact: The impact of selected mutations was assessed 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 cross-species 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 the nucleotide sequence of ... 366-374 Peptide-specific MHC pentamers were purchased from ProImmune Ltd.

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

[0348] Alternatively, age-matched female C57BL / 6 mice were injected intravenously with 20 μg of in vitro transcribed RNA formulated with 20 μl of Lipofectamine® RNAiMAX (Invitrogen) in PBS for 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 sacrificed 23 days after the first injection and splenocytes were isolated for immunological studies. DNA sequences displaying one (monoepitope), two (biepitopes) or 16 (polyepitope) mutations were constructed using 50 amino acids with a mutation at position 25 (biepitopes) or 27 amino acids with a mutation at position 14 (monoepitopes), 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 described previously (Kreiter et al., Cancer Immunol Immunother 2007;56:1577-87).

[0349] Enzyme-linked immunospot assay Enzyme-linked immunospot (ELISPOT) assays (Kreiter S et al., Cancer Res 2010;70:9031-40) and the generation of syngeneic bone marrow-derived dendritic cells (BMDCs) as stimulators have been described previously (Lutz MB et al., J Immunol Methods 1999;223:77-92). BMDCs were pulsed with peptides (2 μg / ml) or transfected with in vitro transcribed (IVT) RNAs encoding the indicated mutations or control RNA (eGFP-RNA). Sequences displaying the two mutations, 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 has been described previously (Kreiter S et al., Cancer Immunol Immunother 2007;56:1577-87). For the assay, 5×10 4 Peptide- or RNA-engineered BMDCs were cultured at 5 × 10 in microtiter plates coated with anti-IFN-γ antibody (10 μg / mL, clone AN18; Mabtech). 5 Freshly isolated splenocytes were co-incubated with 1000 cells / ml of 1000 freshly isolated splenocytes. After 18 h at 37°C, cytokine secretion was detected with an anti-IFN-γ antibody (clone R4-6A2; Mabtech). Spot numbers were counted and analyzed with an 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 test (non-parametric test). A p value of <0.05 was given for the test or a mean spot count of >30 spots / 5 × 10 5 Responses were considered significant if they were effector cells. Reactivity was assessed by the mean number of spots (-: <30; +: >30; ++: >50; +++ >200 spots / well).

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

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

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

[0353] Identification of nonsynonymous mutations in B16F10 mouse melanoma. Our aim was to identify potentially immunogenic somatic point mutations in B16F10 mouse melanoma by NGS and test these for in vivo immunogenicity by peptide vaccination of mice and measure the elicited T cell responses by ELISPOT assay (Figure 9A). We extracted, captured, and sequenced the C57BL / 6 wild-type background genome and the exome of B16F10 cells, each in triplicate. Over 100 million single-end 50-nucleotide reads were generated for each sample. Of these, 80% aligned specifically to the mouse mm9 genome and 49% aligned on target, revealing successful target enrichment, resulting in 20x or greater coverage for 70% of the targeted nucleotides in each of the triplicate samples. RNA-Seq of B16F10 cells, also profiled in triplicate, generated an average of 30 million single-end 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 DNA amplifications in B16F10 and deletions, including a homozygous deletion of the tumor suppressor Cdkn2a (cyclin-dependent kinase inhibitor 2A, p16Ink4A). Focusing on point mutations to identify potentially immunogenic mutations, we identified 3570 somatic point mutations with FDR ≤ 0.05 (Figure 9B). The most frequent class of mutations were C>T / G>A transitions, typically resulting from UV radiation (Pfeifer GP et al., Mutat Res 2005;571:19-31). Of these somatic mutations, 1392 occur in transcripts and 126 mutations occur in untranslated regions. Of the 1266 mutations in coding regions, 962 cause nonsynonymous protein changes, and 563 of these occur in expressed genes (Figure 9B).

[0355] Assignment and validation of identified mutations to carrier genes Of note, many of the mutated genes (962 genes including nonsynonymous somatic point mutations) have been previously associated with cancer phenotypes. Mutations were found in established tumor suppressor genes, including Pten, Trp53 (also called p53) and Tp63. In Trp53, the most widely established tumor suppressor (Zilfou JT et al., Cold Spring Harb Perspect Biol 2009;1:a001883), an asparagine to aspartic acid mutation at protein position 127 (p.N127D) is localized within the DNA binding domain and predicted to alter function by SIFT. Pten contains two mutations (p.A39V, p.T131P), both predicted to have deleterious 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). In addition, mutations were found in genes related to DNA repair pathways, such as Brca2 (breast cancer 2, juvenile), Atm (ataxia telangiectasia mutated), Ddb1 (damage-specific DNA-binding protein 1) and Rad9b (RAD9 homolog B). In addition, mutations occur in other tumor-associated 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 nuclear protein), Mta3 (metastasis-associated 1 family, member 3), and Alk (anaplastic lymphoma receptor tyrosine kinase). We found a mutation at p.S144F in Pdgfra (platelet-derived growth factor receptor alpha polypeptide), a cell membrane-bound receptor tyrosine kinase of the MAPK / ERK pathway, previously identified in tumors (Verhaak RG et al., Cancer Cell 2010;17:98-110). The mutation occurs at p.L222V in Casp9 (caspase 9, apoptosis-associated cysteine ​​peptidase).CASP9 proteolytically cleaves poly(ADP-ribose) polymerase (PARP), regulates apoptosis, and has been implicated in several cancers (Hajra KM et al.,Apoptosis 2004;9:691-704). The mutations we found 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-associated protein) (p.S90R), Ksr1 (kinase suppressor of ras 1) (p.L301V), and Atm (PI3K pathway) (p.K91T), all of which are predicted to have a significant effect on protein function. Trrap (transformation / transcription domain-associated protein) was identified in human melanoma specimens earlier this year as a novel potential melanoma target (Wei X et al., Nat Genet 2011;43:442-6). In B16F10, the Trrap mutation occurs at p.K2783R and is predicted to disrupt the overlapping phosphatidylinositol kinase (PIK)-related kinase FAT domain.

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

[0357] FIG. 9C shows B16F10 chromosomal location, gene density, gene expression, mutations, and filtered mutations (inner ring).

[0358] In vivo immunogenicity testing of long peptides exhibiting mutations To provide antigens for immunogenicity testing of these mutations, we used long peptides, which have many advantages over other peptides for immunization (Melief CJ and van der Burg SH, Nat Rev Cancer 2008;8:351-60). Long peptides stimulate antigen-specific CD8 +and CD4+ T cells (Zwaveling S et al., Cancer Res 2002;62:6187-93; Bijker MS et al., J Immunol 2007;179:5033-40). Furthermore, long peptides require processing to be presented on MHC molecules. Such uptake is most efficient by dendritic cells, which are best suited to prime strong T cell responses. Matching peptides, in contrast, do not require trimming and can be exogenously loaded onto all cells expressing MHC molecules, including non-activated B and T cells, resulting in the induction of immunological tolerance and fratricide (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 long peptide with the mutation or wild type amino acid located in the central region. Thus, any potential MHC class I and class II epitope of 8-14 amino acids long carrying the mutation could be processed from this precursor peptide. As an adjuvant for peptide vaccination, we used poly(I:C), which is 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 the induction of T cells. Strikingly, 16 of the peptides encoding the 50 mutations were found to elicit an immune response in immunized mice. The induced T cells showed various reactivity patterns (Table 7). Table 7: Summary of T cell reactivity measured following vaccination with peptides encoding the mutations. Statistical analysis was performed by Student's t-test and Mann-Whitney test (non-parametric test). A p-value of <0.05 was given for the test or a mean spot count of >30 spots / 5 x 10 5Responses were considered significant if they were effector cells. Reactivity was assessed by the mean number of spots (-:<30; +:>30; ++:>50; +++>200 spots / well). [Table 7]

[0359] Eleven peptides induced immune responses that selectively recognized the mutant epitopes. This is exemplified for mice immunized with mutations 30 (MUT30, Kif18b) and 36 (MUT36, Plod2) (Figure 10A). ELISPOT studies revealed strong mutant-specific immune responses without cross-reactivity to the wild-type peptide or an irrelevant control peptide (VSV-NP). Five peptides, including mutations 05 (MUT05, Eef2) and 25 (MUT25, Plod2) (Figure 10A), yielded immune responses that recognized both mutant and wild-type peptides equally well. The majority of mutant peptides failed to induce significant T cell responses as exemplified by mutations 01 (MUT01, Fzd7), 02 (MUT02, Xpot) and 07 (MUT07, Trp53). The immune responses induced by some of the discovered mutations were sufficient to confirm the positive control, the mouse melanoma tumor antigen tyrosinase-related protein 2 (Trp2 180-188 , Fig. 10A ) (500 spots / 5 × 10 5The BMDCs transfected with the mRNA encoding the mutations or with irrelevant RNA were used as antigen-presenting cells (APCs) and the splenocytes of immunized mice were used as effector cell populations. BMDCs transfected with the mRNA encoding MUT17, MUT30 and MUT44 were specifically and potently recognized by splenocytes of mice immunized with the respective long peptides (Figure 10B). A significantly lower reactivity was recorded against BMDCs transfected with control RNA, possibly due to non-specific 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 mutation-specific T cells indeed recognize endogenously processed epitopes. Two mutations inducing the preferred recognition of mutant epitopes are present in the genes Actn4 and Kif18b. The somatic mutation in ACTN4 (actinin, alpha 4) is located at p.F835V within the calcium-binding "EF-hand" protein domain. Although both SIFT and POLYPHEN predict a significant impact of this mutation on protein function, this gene is not an established oncogene. However, mutation-specific T cells against ACTN4 have recently been associated with favorable patient outcome (Echchakir H et al., Cancer Res 2001;61:4078-83).KIF18B (kinesin family member 18B) is a kinesin with ATP and nucleotide binding that is involved in microtubule motor activity as well as the regulation of cell division (Lee YM et al., Gene 2010;466:16-25) (Figure 10C). The DNA sequence of the position encoding p.K739 is homogeneous in the reference C57BL / 6, but B16F10 DNA reads reveal a heterozygous somatic mutation. Both nucleotides were detected in B16F10 RNA-Seq reads and verified by Sanger sequencing. KIF18B has not been previously associated with a cancer phenotype. The mutation 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 that recognizes the mutation and functional or immunological relevance.

[0360] In vivo evaluation of antitumor activity of vaccine candidates To evaluate whether the immune responses elicited in vivo could translate into antitumor effects in tumor-bearing mice, we chose MUT30 (mutation in Kif18b) and MUT44 as examples. These mutations were shown to selectively induce strong immune responses against mutant peptides and to be endogenously processed (Fig. 10A,B). The therapeutic potential of vaccination with mutant peptides was demonstrated by comparing the 7.5 × 10 5 This was studied by immunizing mice with either MUT30 or MUT44 plus adjuvant 3 and 10 days after implantation of B16F10. Tumor growth was inhibited by both peptide vaccinations compared to the control group (Figure 11A). Since B16F10 is a very aggressively growing tumor, we also tested protective immune responses. Mice were immunized with MUT30 peptide and 4 days later received 7.5 × 10 5B16F10 cells were inoculated subcutaneously and boosted with MUT30 2 and 9 days after tumor challenge. Complete tumor protection and 40% survival of mice treated with MUT30 was observed, whereas all mice in the control treatment group died within 44 days (Figure 11B, left). In mice that developed tumors despite immunization with MUT30, tumor growth was slower, resulting in an average survival increase of 6 days compared to the control group (Figure 11B, right). These data suggest that vaccination against a single mutation can already confer an antitumor effect.

[0361] Immunization with RNA encoding the mutation Various RNA vaccines were constructed using 50 validated mutations from the B16F10 melanoma cell line. DNA sequences displaying one (monoepitope), two (biepitopes) or 16 (polyepitopes) different mutations were constructed using 50 amino acids with a mutation at position 25 (biepitopes) or 27 amino acids with a mutation at position 14 (monoepitopes) and separated 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 the various RNA vaccines, groups of three C57BL / 6 mice were immunized with RNA and RNAiMAX Lipofectamine formulations followed by intravenous injection. After five immunizations, mice were sacrificed and splenocytes were analyzed for mutation-specific T cell responses after restimulation with the corresponding mutant-encoded peptide or a control peptide (VSV-NP) using intracellular cytokine staining and IFN-γ ELISPOT analysis.

[0363] Figure 12 shows an example for each vaccine design. In the top row, mice were vaccinated with monoepitope-RNA encoding MUT30 (mutation in Kif18b), which induces MUT30-specific CD4+ T cells (see example FACS plot). In the middle row, graphs and FACS plots show the induction of MUT08 (mutation in Ddx23)-specific CD4+ T cells after immunization with bi-epitope encoding MUT33 and MUT08. In the bottom row, mice were immunized with poly-epitopes encoding 16 different mutations including MUT08, MUT33 and MUT27 (see Table 8). Graphs and FACS plots illustrate that MUT27-reactive T cells are of CD8 phenotype. Table 8: Summary of mutations and gene names encoded by monoepitope, biepitope, and polyepitope RNA vaccines [Table 8]

[0364] The same polyepitope was used to generate the data shown in Figure 13. The graph shows ELISPOT data following restimulation of splenocytes with control (VSV-NP), MUT08, MUT27 and MUT33 peptides, demonstrating that the polyepitope vaccine is able to induce specific T cell responses against several different mutations.

[0365] Taken together, the data indicate the feasibility of using RNAs encoding mono-, bi-, and polyepitopes to induce mutation-specific T cells. Furthermore, the data demonstrate the feasibility of using RNAs encoding mono-, bi-, and polyepitopes to induce CD4 T cell proliferation from a single construct. + and induction of CD8+ T cells and of several different specificities.

[0366] Immunization with model epitopes To further characterize the polyepitope RNA vaccine design, a DNA sequence was constructed 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)). The epitopes were separated by the same 9 amino acid glycine / serine linker used for the mutant polyepitope. This construct was cloned into the pST1-2BgUTR-A120 backbone for in vitro transcription of mRNA.

[0367] In vitro transcribed RNA was used to vaccinate five C57BL / 6 mice by intranodal immunization (four immunizations with 20 μg RNA into the inguinal lymph node). Five days after the last immunization, blood samples and splenocytes were taken from the mice for analysis. Figure 14A shows IFN-γ ELISPOT analysis of splenocytes restimulated with the indicated peptides. It can be clearly seen that all three MHC class I epitopes (SIINFEKL, Trp2 and VSV-NP) induce very high numbers of antigen-specific CD8+ T cells. The MHC class II epitope OVA class II also induces a strong CD4+ T cell response. The fourth MHC class I epitope was analyzed by staining Inf-NP-specific CD8+ T cells with fluorescently labeled pentameric MHC peptide complexes (pentamers). (Figure 14B).

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

[0369] Antitumor responses following treatment with a polyepitope RNA vaccine encoding mutations The same polyepitope analyzed in Figure 13 for immunogenicity was used to study the antitumor activity of mutant coding RNA against B16F10 tumor cells. In particular, groups of C57BL / 6 mice (n=10) were injected with 1x10 5 B16F10 melanoma cells were inoculated subcutaneously in the flank. Mice were immunized with polyepitope RNA using a liposomal transfection reagent on days 3, 6, 10, 17, and 21. The control group was injected with liposomes alone.

[0370] FIG. 21 shows the survival curves of both groups, revealing a strongly improved median survival time of 27 days compared to 18.5 days in the control group, with 1 in 10 mice surviving tumor-free.

[0371] Antitumor responses following treatment with a combination of mutant and normal peptides The antitumor activity of the demonstrated mutations was evaluated by therapeutic in vivo tumor experiments using MUT30 as a peptide vaccine. In detail, groups of C57BL / 6 mice (n=8) were inoculated with 1×10 5 B16F10 melanoma cells were inoculated subcutaneously into the flank. On days 3, 10, and 17, polyI:C was used as an adjuvant to inject MUT30, tyrosinase-related protein 2 (Trp2 180-188 ) 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 average tumor growth of each group. It can be clearly seen that the group immunized with the combination of the known CD8+ T cell epitope and MUT30 inducing CD4+ T cells has almost complete inhibition of tumor growth until day 28. Although the known Trp2 epitope alone is not enough to provide good antitumor activity in this setting, both single treatment groups (MUT30 and Trp2) still provide inhibition of tumor growth compared to the untreated group from the start of the experiment until day 25. These data are supported by the survival curves shown in Figure 15B. Clearly, mice injected with a single peptide increased the average survival rate, with 1 / 8 mice surviving in the group vaccinated with Trp2. In addition, the group treated with both peptides showed even better survival, with 2 / 8 mice surviving.

[0373] Taken together, the two epitopes act synergistically to provide potent anti-tumor activity.

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

[0375] However, although groundbreaking, NGS platforms are still prone to errors that lead to erroneous mutation calls. Moreover, the quality of the results depends on the experimental design parameters and analysis methods. Mutation calling typically involves scores designed to distinguish true mutations from errors, but the utility of these scores is not well understood, as is their interpretation with respect to experimental optimization. This is especially true when comparing tissue states, i.e., tumor and normal tissues with respect to somatic mutations. As a result, researchers must rely on personal experience in determining experimental parameters and arbitrary filtering thresholds for mutation selection.

[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 B16F10 melanoma cell lines. These data are used to develop a false discovery rate for detected somatic mutations, a measure that is then used to evaluate existing mutation discovery software and experimental protocols.

[0377] A variety of experimental and algorithmic factors contribute to the false positive rate for mutations found 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)] and bias in target enrichment [Bainbridge, MNet al., Genome Biol. 11, R62 (2010)], sequence effects [Nakamura, K. et al., Acids Res. (2011) first published online May 16, 2011 doi:10.1093 / nar / gkr344], base calling leading to sequence errors [Kircher, M. et al., Genome Biol. 10, R83 (2009). Epub 2009 Aug 14] as well as further downstream analysis, e.g. mutation calling around indels [Li, H., Bioinformatics 27, 1157-1158 (2011)] and read alignment that introduces sequencing errors [Lassmann, T. et al., Bioinformatics 27, 130-131 (2011)].

[0378] No general statistical model accounting for the influence of different error sources on somatic mutation calling has been described; only individual aspects are covered without eliminating all biases. Recent computational methods for measuring the expected amount of false positive mutation calls include the use of transition / transversion ratios for a set of mutations [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 inherited errors when working on family genomes [Ewen, KR et al., Am. J. Hum. Genet. 67, 727-736 (2000)] or on 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. For the set of single nucleotide variants (SNVs) and selected experiments, comparison with SNVs identified by other techniques is feasible [Van Tassell, CP et al., Nature Methods 5, 247-252 (2008)], but it is difficult to assess for novel somatic mutations.

[0379] Using the exome sequencing project as an example, we propose to calculate the false discovery rate (FDR) based solely on the NGS data. This method is not only applicable to the selection and prioritization of diagnostic and therapeutic targets, but also aids in the development of algorithms and methods by allowing us to define confidence-driven recommendations for similar experiments.

[0380] To discover mutations, DNA from tail tissues of three C57BL / 6 (black6) mice (littermates) and DNA from B16F10 (B16) melanoma cells were enriched for protein-coding exons separately in triplicate (Agilent Sure Select Whole Mouse Exome), yielding six samples. RNA was extracted in triplicate from B16 cells. Single-end 50-nucleotide (1 x 50 nt) reads and paired-end 100-nucleotide (2 x 100 nt) reads were generated on an Illumina HiSeq 2000. Each sample was loaded on a separate lane, generating 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)]. A 38-fold average coverage of 97% of the targeted regions was achieved for the 1 × 50 nt library, and the 2 × 100 nt experiment yielded a 165-fold average coverage of 98% of the targeted regions.

[0381] Somatic mutations were independently identified by comparing single nucleotide mutations found in B16 samples with the corresponding loci in black6 samples (B16 cells were originally derived from black6 mice) 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). Potential mutations were filtered according to the recommendations by the respective software authors (SAMtools and GATK) or by selecting the appropriate lower threshold for the somatic score of SomaticSNiPer, respectively.

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

[0383] FIG. 18A shows examples of mutations found in the black6 / B16 data, including somatic mutations (left), non-somatic mutations (center), and possible false positives (right) relative to the reference. Each somatic mutation can be associated with a quality score Q. The number of false positives in the tumor comparison indicates the number of false positives in the pairwise comparison. Thus, for a given mutation with a quality score Q detected in the tumor comparison, we estimate the false discovery rate by computing the ratio of pairwise mutations with Q or better to the total number of mutations found in the tumor comparison with Q or better.

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

[0385] A potential bias in comparative methods is differential coverage; we therefore normalize the false discovery rate with respect to coverage:

number

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

[0387] By estimating the number of false positives and positives at each FDR (see Methods), receiver operating characteristic (ROC) curves were constructed and the AUC (area under the curve) was calculated for each mutation discovery method, thus allowing a comparison of strategies for mutation discovery (Figure 18B).

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

[0389] Using this definition of the false discovery rate, we establish a general framework for evaluating the influence of numerous experimental and algorithmic parameters on the set of resulting somatic mutations. We then apply this framework to examine the effects of software tools, coverage, paired-end sequencing, and number of technical replicates on somatic mutation identification.

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

[0391] For the B16 sequencing experiment described above, we sequenced each sample in an individual flow cell lane, achieving a target region average base coverage of 38x for each individual sample. However, this coverage is not believed to be necessary to obtain an equally good set of somatic mutations, presumably to reduce costs. Also, the impact of coverage depth on genome-wide SNV detection has been recently discussed [Ajay, Sset et 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 1x50nt libraries, generating approximate coverages of 5, 10 and 20x, respectively, and then reapplied the mutation calling algorithm. As expected, higher coverage results in a better (i.e., fewer false positives) set of somatic mutations, but the improvement from 20x coverage to the maximum is small (Figure 19B).

[0392] It is straightforward to simulate and rank different experimental settings using the available data and framework. Comparing duplicates to triplicates, triplicates do not confer any benefit over duplicates (Figure 19C), but duplicates provide a clear improvement over tests without replicates. In terms of the proportion of somatic mutations in a given set, we observe an enrichment from 24.2% to 71.2% for duplicates and 85.8% for triplicates at an FDR of 5%. Despite the enrichment, using the intersection of triplicates removes more mutations with low FDR than those with high FDR, as shown by the lower ROC AUC and the shift of the curve to the left (Figure 19C): specificity is slightly increased at the cost of lower sensitivity.

[0393] Using the additionally sequenced 2x100nt library, we simulated 1x100nt, two 2x50nt and two 1x50nt libraries, respectively, by in silico removing the second read and / or the 3' and 5' ends of the read, making a total of five simulated libraries. These libraries were compared using the calculated FDR of predicted mutations (Figure 19D). Despite a much higher average coverage (>77 vs. 38), the somatic mutations found using the 2x50nt 5' and 1x100nt libraries have a lower ROC AUC and therefore a poorer FDR distribution than the 1x50nt library. This phenomenon results from the accumulation of high FDR mutations in low coverage regions, as the sets of low FDR mutations found are highly similar. As a result, the optimal sequencing length must be either small so that the sequenced bases are concentrated near the capture probe sequence (although potentially losing information about the somatic mutation status in the uncovered regions) or close to the fragment length (in our case 2 × 100 nt = 200 nt total length for a ∼250 nt fragment) to effectively fill the coverage gap. This is also supported by the fact that the ROC AUC of the 2 × 50 nt 3' library (simulated by using only the 3' end of the 2 × 100 nt library) is higher than that of the 2 × 50 nt 5' library (simulated by using only the 5' end of the 2 × 100 nt library), despite the lower base quality of the 3' read end.

[0394] These observations allow us to define the best practice procedure for somatic mutation discovery. Using 20x coverage in both samples and technical duplicates across all evaluation parameters achieves near-optimal results in these relatively homogenous samples, while also taking into account costs. A 1x50nt library yielding approximately 100 million reads appears to be the most practical choice to achieve this coverage. This is true across all possible dataset pairs. We applied these parameter settings retrospectively and calculated the FDR for 50 selected mutations from the intersection of all three methods shown in Figure 17 without additional filtering of raw mutation calls. All mutations were confirmed by a combination of Sanger resequencing and B16 RNA-Seq sequence reads. 44 of these mutations would have been found using a 5% FDR cutoff value (Figure 20). As negative controls, we resequenced the loci of 44 predicted mutations with high FDR (>50%) and examined the respective sequences in the RNA-Seq data. We found that 37 of these mutations were not substantiated, and the remaining 7 loci of potential mutations were not covered by RNA-Seq reads or yielded any sequencing reactions.

[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 consider the influence of all experimental or algorithmic parameters, e.g., the influence of the alignment software, the choice of mutation metric, or the choice of vendor for exome selection.

[0396] We performed all experiments on a set of B16 melanoma cell experiments; however, the method is not limited to these data. The only requirement is that a "paired" reference dataset is available, meaning that at least one technical replicate of a non-tumor sample should be performed for each new protocol. Our experiments show that the method is robust to the selection of technical replicates within certain limits, so that replicates are not necessarily required in every single experiment. However, the method requires that various quality assessment measures are comparable between the reference dataset and the remaining datasets.

[0397] Within this contribution, we developed a statistical framework for false discovery rate-driven detection of somatic mutations. This framework is not only applicable to diagnostic or therapeutic target selection, but also allows for a general comparison of experimental and computational protocol steps on the pseudo-true data generated. Here, we applied this idea to determine protocols in terms of software tools, coverage, replicates as well as paired-end sequencing.

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

[0399] Three micrograms of purified genomic DNA were fragmented to 150-200 nt using a Covaris S2 sonicator. gDNA fragments were end-repaired using T4 DNA polymerase, Klenow DNA polymerase, and 5' phosphorylated using T4 polynucleotide kinase. Blunt-ended gDNA fragments were 3' adenylated (3'-5' exo minus) using Klenow fragment. Single 3' T-overhang Illumina paired-end adapters were ligated to the gDNA fragments using a 10:1 molar ratio of adapter to genomic DNA insert using T4 DNA ligase. Adapter-ligated gDNA fragments were enriched prior to capture, and flow cell-specific sequences were added using four PCR cycles using Illumina PE PCR primers 1.0 and 2.0 and Herculase II polymerase (Agilent).

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

[0401] Cleanup was performed using 1.8 volumes of AMPure XP magnetic beads (Agencourt). For quality control we used the Qubit HS assay from Invitrogen, and fragment size was determined using the 2100 Bioanalyzer HS DNA assay from Agilent.

[0402] Exome-enriched gDNA libraries were clustered on a cBot with the Truseq SR Cluster Kit v2.5 using 7 pM and sequenced on an Illumina HiSeq2000 using the Truseq SBS kit.

[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)] using default options. Ambiguous reads - reads mapping to multiple locations in the genome as provided by the bwa output were removed. The remaining alignments were curated, indexed, and converted to a binary compressed format (BAM), and read quality scores were converted from Illumina standard phred+64 to standard Sanger quality scores using shell scripts.

[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 options and filtering criteria recommended by the authors were used, including 1st filtering, maximum coverage of 200 (http: / / sourceforge.net / apps / mediawiki / SAMtools / index.php?title=SAM_FAQ; accessed September 2011). For the second round of filtering in SAMtools, the indel minimum quality score was 50 and the point mutation minimum quality was 30. For GATK mutation calling, we followed the best practice guidelines designed by the authors 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, a local realignment near the indel site was performed, followed by a recalibration of the base quality. The UnifiedGenotyper module was applied to the resulting aligned data files. When necessary, known polymorphisms from dbSNP [Sherry, ST et al., Nucleic Acids Res. 29, 308-311 (2009)] (version 128 for mm9) were supplied to the individual steps. The variant score recalibration step was omitted and replaced with a hard filtering option. For SomaticSniper mutation calling, default options were used and only predicted mutations with a "somatic score" of 30 or greater were considered further.In addition, for each potentially mutated locus, we required non-zero coverage in normal tissues and removed all mutations located in repeat sequences defined by the RepeatMasker track of the UCSC Genome Browser for the mouse genome assembly mm9 [Fujita, PA et al., Nucleic Acids Res. 39, 876-882 (2011)].

[0405] RNA-Seq A barcoded mRNA-seq cDNA library was 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). Isolated mRNA was fragmented using divalent cations and heat, resulting in fragments ranging from 160 to 200 bp. Fragmented mRNA was converted to cDNA using random primers and SuperScriptII (Invitrogen), and then the second strand was synthesized using DNA polymerase I and RNaseH. cDNA was end-repaired using T4 DNA polymerase, Klenow DNA polymerase, and 5' phosphorylated using T4 polynucleotide kinase. Blunt-ended cDNA fragments were 3' adenylated (3'-5' exo minus) using Klenow fragment. Single 3' T-overhang Illumina multiplex-specific adaptors were ligated to the cDNA fragments using T4 DNA ligase. The cDNA library was purified and size selected at 300 bp using E-Gel 2% SizeSelect gel (Invitrogen). Enrichment, addition of Illumina 6-base index sequences and flow cell specific sequences were performed by PCR using Phusion DNA polymerase (Finnzymes). All cleanup was performed using 1.8x volume of Agencourt AMPure XP magnetic beads.

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

[0407] The raw output data from HiSeq was processed according to Illumina standard protocols, including removal of low-quality reads and demultiplexing. Sequence reads were then aligned to the reference genome sequence [Mouse Genome Sequencing Consortium, Nature 420, 520-562 (2002)] using bowtie [Langmead, B. et al., Genome Biol. 10, R25 (2009)]. Alignment coordinates were compared to exon coordinates of RefSeq transcripts [Pruitt, KD et al., Nucleic Acids Res. 33, 501-504 (2005)], and overlap alignment counts were 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, KD et al., Nucleic Acids Res. 33, 501-504 (2005)]. Alignment coordinates were compared to RefSeq exon and junction coordinates, and reads were counted and normalized to RPKM (number of reads which map per nucleotide kilobase of transcript per million mapped reads) for each transcript [Mortazavi, A. et al., Nat. Methods 5, 621-628 (2008)].

[0408] Validation of SNVs We selected SNVs for validation by Sanger sequencing and RNA. We identified SNVs predicted by all three programs, nonsynonymous, and found in transcripts with a minimum of 10 RPKM. Among 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% or higher, present only in one cell line sample, and predicted only by 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 and 32 loci in the positive and negative controls, respectively. Validation was also performed by examination of 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 affect the quality of the mutation call. These include, but are not limited to, the base quality of interest assigned by the instrument, the quality alignment for this position, the number of reads covering this position, or the score for the difference between the two genomes compared at this position. For the calculation of the false discovery rate, a ranking of the mutations is required, but this cannot be performed directly for all mutations, as the various quality scores may give conflicting information.

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

number

[0411] However, we consider IFDR merely an intermediate step, since in many closely related cases comparison is not feasible and therefore does not benefit from the vast amount of available data. Therefore, we take advantage of 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 with n quality characteristics, a value range for each characteristic was determined and up to p values ​​were sampled at uniform intervals from this range; if the set of values ​​for a quality characteristic was smaller than p, this set was used instead of the sample set. Then, in the n-dimensional quality space, n"Create each possible combination of sampled or selected quality values ​​that results in the maximum value of the data points. A 1% random sample of these points and the 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: it can be viewed as a locally weighted combination of the individual quality scores. This allows for a direct single-valued comparison of any two mutations and the calculation of the actual false discovery rate:

number

[0414] For training the random forest model used to generate the results of this study, we calculate the sample IFDR for somatic mutations for all samples before selecting a random 1% subset. This ensures that we map the entire available quality space to FDR values. We used the quality attributes "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), respectively. Different values ​​of p ensure comparable large set sizes.

[0415] Calculation of common coverage: The number of possible mutation calls can introduce an important bias in the definition of the false discovery rate. Only if the number of possible positions where mutations occur for our tumor and pairwise comparisons is the same, the number of called mutations is comparable and can be used as the basis for the false discovery rate calculation. To correct for this potential bias, we use the common coverage ratio. As common coverage, we define the number of bases with at least one coverage in both samples used for mutation calling. We calculate the common coverage separately for tumor and pairwise comparisons.

[0416] ROC Estimation Receiver Operating Characteristic (ROC) curves and the 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 a ROC graph requires knowledge of all true positive and false positive (TP and FP) examples in the dataset, information that is usually not given and is difficult to establish for high-throughput data (such as NGS data). Therefore, we use the calculated FDR to estimate the respective TP and FP rates, plot the ROC graph, and calculate the AUC. The central idea is that the FDR of a single mutation in the dataset gives the proportion to which this mutation contributes to the total TP / FP mutations, respectively. Also, for a list of random assignments to TP and FP, the resulting ROC AUC is equal to 0.5 for our method, indicating a completely random prediction.

[0417] We have two conditions:

number

number

[0418] To obtain the estimated ROC curve, the mutations in the data set are sorted by FDR and for each mutation, a point is plotted on the cumulative TPR and FPR values ​​up to this mutation divided by the sum of all TPR and FPR values, respectively. The AUC is calculated by summing the areas of all successive trapezoids between the curve and the x-axis.

[0419] Example 10 Selection of combinations of tumor antigens as targets for cancer therapy In this example, it was assessed whether it was possible to establish a set of tumor antigens that are at least partially shared by a large proportion of tumor patients, thereby providing a set of vaccine products that could be applied to a broad spectrum of cancer patients.

[0420] For this, 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 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 only three tumor antigens, namely DCT (isoform 1), TYR and TPTE, was sufficient to represent 88% of the analyzed patient samples (Figure 22).

Claims

1. 1. A personalized vaccine for use in a method for preventing or treating cancer in a patient, comprising: The personalized cancer vaccine is (A) a polyepitope polypeptide comprising a neoepitope based on the patient's mutation; or (B) a nucleic acid encoding the polyepitope polypeptide of (A); Including, The method comprises the following steps: (i) administering a vaccine to induce a first immune response, The vaccine for inducing the first immune response comprises: (a) a peptide or polypeptide comprising a combination of common tumor antigens; (b) a peptide or polypeptide comprising a T cell epitope of a combination of common tumor antigens, or (c) a nucleic acid encoding the peptide or polypeptide of (a) or (b), the first immune response is induced against the tumor antigen in the patient; and (ii) administering a vaccine to induce a second immune response, wherein the second immune response is induced against a neoepitope based on the mutation in the patient, and the second immune response is specific to a somatic mutation present in cancer cells of the patient; Personalized vaccines.

2. A vaccine for use according to claim 1, comprising: the primary and / or secondary immune response is a cell-mediated response, and / or the first immune response comprises a CD8+ T cell response; and / or administration of the personalized vaccine may provide MHC class II-presented neoepitopes; The secondary immune response comprises a CD4+ T cell response, the vaccine.

3. A vaccine for use according to claim 1 or 2, comprising the first immune response is not specific for a cancer-specific somatic mutation present in the patient's cancer cells, and / or the tumor antigen is common in the cancer being treated, and / or The tumor antigens are common in a variety of cancer vaccines.

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

5. A vaccine for use according to claim 3, comprising: the cancer-specific somatic mutation is present in the exome of the patient's cancer cells; and / or The cancer-specific somatic mutation is a non-synonymous mutation.

6. A vaccine for use according to any one of claims 1 to 5, comprising each of said polyepitope polypeptides comprising mutation-based neoepitopes comprises up to 30 mutation-based neoepitopes; Each of the polyepitopic polypeptides further comprises an epitope that does not contain a cancer-specific somatic mutation expressed by the cancer cell.

7. A vaccine for use according to claim 6, comprising: A vaccine wherein said mutation-based neoepitope and said epitope are present in their natural sequence context to form a vaccine sequence, said vaccine sequence being 30 amino acids in length.

8. A vaccine for use according to claim 6 or 7, comprising the mutation-based neoepitope, the epitope, and / or the vaccine sequence are arranged in a head-to-tail direction and / or separated by a linker; The linker has the amino acid sequence (GGS) ａ (GSS) ｂ (GGG) ｃ (SSG) ｄ (GSG) ｅ wherein a, b, c, d and e are independently numbers selected 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 vaccine for use according to any one of claims 1 to 8, comprising A vaccine, wherein the primary and / or secondary immune response is induced by administration of an RNA vaccine.

10. A vaccine for use according to claim 9, comprising: A vaccine in which cytidine is partially or completely substituted with 5-methylcytidine or uridine is partially or completely substituted with pseudouridine.

11. A vaccine for use according to claim 9 or 10, comprising The 5' end of the RNA vaccine comprises a cap structure having the general formula: 【Chemical 1】 In the formula, R １ and R ２ is independently hydroxy or methoxy, and W － , X － and Y － are independently oxygen, sulfur, selenium or BH ３ That is, the vaccine.

12. A vaccine for use according to any one of claims 9 to 11, comprising The vaccine, wherein the RNA is an mRNA comprising a 5'-UTR and a 3'-UTR.

13. A vaccine for use according to any one of claims 9 to 12, comprising The RNA comprises a poly(A) tail having a length of 100 to 150 adenosine residues.

14. 14. A vaccine for use according to claim 12 or 13, comprising The vaccine, wherein the 3'-UTR comprises two copies of the 3'-UTR derived from a globin gene.

15. A vaccine for use according to any one of claims 9 to 14, comprising The vaccine, wherein the RNA is associated with a carrier comprising a lipid-containing carrier, a cationic lipid, a liposome, a micelle, or a nanoparticle.

Images