Personalized vaccine

Abstract

To provide cancer vaccines based on a replication-defective vaccinia virus such as an MVA expressing antitumor neopeptide fusions, and to provide methods for preparing the same.SOLUTION: The present invention relates to a personalized cancer vaccine comprising a recombinant poxvirus encoding one or more neopeptides, or to a composition comprising such a recombinant poxvirus and a pharmaceutically acceptable vehicle, and to use of the personalized cancer vaccine to treat a cancer subject in need thereof. A particular embodiment is a method for providing the vaccine or composition, comprising the steps of: a) extracting DNA from tumor and non-tumor samples; b) selecting a target region, preferably the entire coding region of the genome (exome); c) sequencing the target region (e.g., the exome) from the extracted DNA; and d) identifying one or more tumor-specific mutations by comparing the DNA sequences obtained from the tumor and non-tumor samples.SELECTED DRAWING: None

Description

[Technical Field]

[0001] Technical field to which the invention belongs The present invention generally relates to a personalized cancer vaccine comprising a recombinant poxvirus encoding one or more neopeptides, or a composition comprising such a recombinant poxvirus and a pharmaceutically acceptable vehicle, and the use of said personalized cancer vaccine to treat a cancer subject in need thereof. Particular embodiments relate to a method of providing such a vaccine or composition, comprising: a) extracting DNA from tumor and non-tumor samples; b) selecting a target region, preferably the entire coding region of the genome (exome); c) sequencing the target region (e.g., the exome) from the extracted DNA; and d) identifying one or more tumor-specific mutations by comparing the DNA sequences obtained from the tumor and non-tumor samples. Embodiments also include a method of treating cancer or preventing its recurrence, comprising administering such a personalized cancer vaccine. The present invention is of very particular interest in the field of personalized immunotherapy, particularly for stimulating T-cell immune responses. [Background technology]

[0002] background In recent decades, numerous therapeutic vaccines expressing tumor antigens have been developed to stimulate natural and specific immune responses against tumor antigens. However, the most commonly identified tumor-associated antigens (e.g., MUC-1, WT1, PSA, CEA) are selectively overexpressed in tumors but may have residual expression in non-tumor cells. Therefore, the effectiveness of this traditional approach is expected to be limited by self-tolerance to such "self" antigens.

[0003] Furthermore, the traditional paradigm of delivering the same set of cancer antigens to everyone in a population ignores the individual variability of disease risk and immune response. Importantly, it cannot be ignored that humans respond differently to vaccines and host immune responses vary significantly within populations (Relman, 2008, J Infect Dis. 198(1):4-5; Plotkin, 2008, Clin Infect Dis.47(3):401-9). With the recent introduction of immune checkpoint blockade drugs in the clinic, it has become clear to medical professionals that certain treatments are effective for some patients but not others.

[0004] Advances in immunology, genetics, molecular biology, and bioinformatics are paving the way for more personalized approaches. Technological breakthroughs in genome sequencing (e.g., next-generation sequencing (NGS)) now make it possible to sequence the entire genome or exome (the coding region of the genome) of a tumor at unprecedented speed and cost. Molecular characterization of tumors has demonstrated that mutations arise during cancer cell oncogenesis and proliferation as a result of rapid proliferation, defective repair mechanisms, and clonal selection. The accumulation of mutations in tumor genomes generally leads to the expression of aberrant protein species specific to cancer tissues. These are called neoantigens. In contrast to most common tumor-associated antigens, tumor neoantigens are present only in tumor cells, not in normal cells, and do not induce the loss of antigen-specific T cells in the thymus. Therefore, they are expected to induce potent immune responses without the risk of self-tolerance and autoimmune reactions to self-proteins. Therefore, tumor neoantigens may represent ideal targets for the design of tumor-specifically adapted therapeutic vaccines, although their adoption in routine care will require overcoming numerous scientific and technological challenges. In particular, most cancer mutations are the result of stochastic phenomena and are specific to each patient.

[0005] The selection of an appropriate vaccine platform is a key factor for its success in clinical practice. Several technological platforms are currently envisioned for the development of neoantigen-directed vaccines, including bacteria (e.g., the Listeria-based ADXS-Neo vaccine developed by Advaxis; WO2016 / 207859; WO2016 / 191545), viral vectors (e.g., the lentiviral system ZVex from Immune Design), nucleic acid vaccines (e.g., the nanoparticle-encapsulated mRNA-based vaccine developed by Biontech; Kreiter et al., 2015, Nature 520 692-6; WO2012 / 159754, DNA encoding the vaccine complex developed by Vaccibody), and adjuvanted peptides (WO2016 / 187508). Representative examples of peptide vaccines include poly-ICLC (polyinosinic-polycytidylic acid)-stabilized peptides currently being developed by NEON Therapeutics for bladder cancer, glioblastoma, and non-small cell lung cancer (NSCLC) in combination with nivolumab; gp96 (96 kDa heat shock protein)-adjuvanted tumor neoantigens developed by Agenus; and liposome-encapsulated peptides from the University of Connecticut School of Medicine. Another recent alternative is the use of dendritic cells (DCs) as vectors after in vitro exposure to antigens before administration to patients. DCs are pulsed with peptides or tumor lysates (e.g., the activated DC vaccine sipuleucel-T marketed by Dendreon; DCs loaded with patient tumor exomes developed by Exocyte Therapeutics; and DC vaccines developed by OncoTherapeutics Science and Tella Inc.).

[0006] However, the most promising leaders in this field must address key challenges before they can achieve their promise. Among other topics, successful translation depends on identifying tumor mutations, designing neopeptides incorporating such mutations, and achieving effective manufacturing methods to ensure rapid delivery of clinically sufficient doses to patients at the bedside, as the production and testing of personalized vaccines is contingent on disease progression.

[0007] Therefore, there is a need to develop a streamlined, time-intensive method that allows for the identification of neoantigens for each patient's tumor and the production of corresponding personalized therapies using established standards for drug manufacturing.

[0008] In previous attempts to develop cancer vaccines, recombinant MVA (Modified Virus Ankara) has served as an effective vector (Acres and Bonnefoy, 2008, Expert Review of Vaccines 7, 889-93). Vaccinia-based vaccines have shown results when administered as monotherapy or in combination with chemotherapy, radiation therapy, or immune checkpoint inhibitors (Farsaci et al., 2011, In Cancer Vaccines: From Research to Clinical Practice, Ed Bot; CRC Press, pp56-77; WO2015 / 175340; WO2015 / 175334). For example, TG4010 (or its research name MVATG9931), an MVA-based cancer vaccine encoding the MUC1 tumor-associated antigen and human interleukin-2 (IL-2), has demonstrated efficacy in combination with first-line standard-of-care chemotherapy in advanced metastatic NSCLC (Quoix et al., 2011, The Lancet Oncology 12(12): 1125-33). Antigen-specific tumor control by MVATG9931 in a prophylactic RMA-MUC1 model was associated with both MUC1 and CD8 + and CD4 +It apparently relies on transient de novo expression in T cells. Other poxviruses, particularly vaccinia virus and fowlpox virus, have also been used as anti-cancer vectors (Gulley et al., 2008, Clin Cancer Res 14(10): 3060-9).

[0009] However, the ability of MVA or other poxviruses to induce potent immune responses against cancer neoantigens has not been tested and remains to be established. In addition, known methods for producing recombinant MVA rely on homologous recombination, which is inefficient and therefore very time-consuming. Summary of the Invention

[0010] We found that recombinant MVA can indeed induce immune responses against several cancer neoantigens and that personalized MVA encoding several neopeptides can be rapidly generated.

[0011] The present disclosure meets the aforementioned needs by providing a cancer vaccine based on MVA expressing a fusion of an antitumor neopeptide, as well as a manufacturing method that is specifically designed to maximize the production of recombinant MVA and improve scalability and time to completion of the method.

[0012] This technical problem is solved by providing the embodiments defined in the claims.

[0013] Other and further aspects, features, and advantages of the present invention will be apparent from the following description of the presently preferred embodiments of the invention, which embodiments are given for purposes of disclosure.

[0014] Summary of the Invention The present invention relates generally to personalized cancer vaccines comprising recombinant poxviruses encoding one or more neopeptides.

[0015] In one aspect, the recombinant poxvirus is a vaccinia virus, preferably a replication-deficient vaccinia virus such as MVA.

[0016] In another aspect, each of the one or more neopeptides encoded by the recombinant poxvirus comprises one or more tumor-specific mutations, and preferably, at least 60% of the neopeptides comprise missense or frameshift mutations. Desirably, the one or more neopeptides have a length of 16 to 90 amino acid residues, preferably 17 to 85 amino acid residues, and more preferably 18 to 80 amino acid residues. In a preferred embodiment, the neopeptides comprising missense mutations have a length of 18 to 29 amino acid residues, and the neopeptides comprising frameshift mutations have a length of 30 to 80 amino acid residues. At least 80% of the neopeptides comprising missense mutations have a substituted amino acid at the central position. In a preferred embodiment, some, preferably all, neopeptides are expressed by the recombinant poxvirus in the form of one or more fusions. The recombinant poxvirus may further encode one or more therapeutic genes, preferably selected from the group consisting of suicide genes and immunostimulatory genes.

[0017] In a further aspect, the present invention also provides a method for preparing a personalized cancer vaccine, comprising identifying one or more neopeptides suitable for encoding by the personalized cancer vaccine, wherein the one or more neopeptides comprise one or more tumor-specific mutations. A preferred method comprises the following substeps: a) extracting DNA from tumor and non-tumor samples; b) selecting a target region, preferably the entire coding region of the genome (exome); c) sequencing the target region (e.g., the exome) from the extracted DNA; and d) identifying one or more tumor-specific mutations by comparing DNA sequences obtained from the tumor and non-tumor samples. Such a method may further comprise one or more of the following additional substeps: e) ranking potential neopeptides by their expression level in the tumor, either at the mRNA transcription level or the protein translation level; f) selecting non-self-expressed tumor-specific mutations; or g) predicting the immunopotency of neoepitopes contained in the neopeptides.

[0018] In addition to the identification step, the method may also comprise a step of producing the recombinant poxvirus. Preferably, the nucleic acid molecule encoding the neopeptide to be inserted into the genome of the recombinant poxvirus is placed in one or more expression cassettes under the control of suitable regulatory elements that allow its expression in the subject. Preferably, the step of producing the recombinant poxvirus uses a parent poxvirus comprising a fluorescent reporter gene cloned at the insertion site selected for the nucleic acid molecule encoding the neopeptide or the expression cassette. In a preferred embodiment, the step of producing the recombinant poxvirus comprises a step of cleavage with an endonuclease capable of generating one or more double-strand breaks in the nucleotide sequence of the fluorescent reporter, wherein the endonuclease does not cleave the poxvirus genome. The method may also comprise a production step, wherein the production step comprises a step of amplifying to a suitable scale in suitable producer cells, a step of recovering the produced recombinant poxvirus from the cell culture, and an optional step of purifying the recovered recombinant poxvirus.

[0019] In yet another aspect, the personalized cancer vaccine is used in the subject who needs it to treat cancer or prevent its recurrence in the subject.The cancer is preferably solid tumor, particularly brain cancer or lung cancer.In a preferred embodiment, the personalized cancer vaccine is administered in combination with one or more additional anti-cancer therapies that are useful in treating cancer. [Brief explanation of the drawings]

[0020] [Figure 1] Figure 1 illustrates the design of the expression cassettes encoded by MVATG19022 and MVATG19023. "GS" represents a 10-amino acid linker, "Flag" represents the Flag tag, and "SR" is an acronym for the signal peptide derived from the rabies glycoprotein. FCU1, b-Gal, E16HPV16, E76HPV16, and MCU1 represent 27-amino acid-long peptides derived from these antigens. [Figures 2A-2C] Figures 2A-2C show ELISpot results after immunization of C57BL / 6 mice (5 mice / group) with peptide-encoding vectors MVATG19022 and MVATG19023 compared with MVA vectors encoding whole antigens: MUC1 (MVA-MUC1), FCU1 (MVA-FCU1), b-galactosidase (MVA-bGal), HPV-16 E7 (MVA-E7HPV16), and HPV-16 E1 (MVA-E1HPV16). Stimulation was performed with either peptides specific for each expressed antigen / peptide (FCU1-specific EG15 (A), MUC1-specific L15L3 (B), E7-specific R9F (C), E1-specific I8L (D), and b-Gal-specific I8V5E, respectively) or an irrelevant peptide ("CTRL" (control)). [Figure 2D-2E] Figures 2D-2E show ELISpot results after immunization of C57BL / 6 mice (5 mice / group) with peptide-encoding vectors MVATG19022 and MVATG19023 compared with MVA vectors encoding total antigens: MUC1 (MVA-MUC1), FCU1 (MVA-FCU1), b-galactosidase (MVA-bGal), HPV-16 E7 (MVA-E7HPV16), and HPV-16 E1 (MVA-E1HPV16). Stimulation was performed with either peptides specific for each expressed antigen / peptide (FCU1-specific EG15 (A), MUC1-specific L15L3 (B), E7-specific R9F (C), E1-specific I8L (D), and b-Gal-specific I8V5E, respectively) or an irrelevant peptide ("CTRL" (control)). [Figure 3] FIG. 3 illustrates the amino acid sequences of the first and second CT26 pentatope fusions expressed by MVATG19030. [Figure 4]Figure 4 illustrates ELISpot after immunization of Balb / c mice (5 / group) with MVATG19030 and MVATG19038 vectors. IFNγ responses were assessed after stimulation with either a pool of 10 mutant CT26 peptides or a pool of CT26 wild-type peptides (i.e., without tumor-specific mutations), with or without priming of CT26-MUC1 cells. [Figure 5] Figure 5 illustrates the IFNγ responses generated to the CT26 mutant peptide (EH27m) and the non-mutated CT26 peptide (PE27wt) after immunization of Balb / c mice with MVATG19030 and MVATG19038 (or virus formulation buffer S08 as a negative control) in both mouse models primed and not primed with CT26-MUC1 cells. Stimulation with culture medium serves as a negative control. [Figure 6] Figure 6 illustrates the IFNγ responses generated to the CT26 mutant peptide (EH27m) and the non-mutated CT26 peptide (PE27wt) after immunization of Balb / c mice with MVATG19030 and MVATG19038 (or virus formulation buffer S08 as a negative control) in both mouse models primed with and not primed with CT26-MUC1 cells. IFNγ responses are assessed in the presence (+anti-MHCII) and absence of anti-mouse MHCII antibodies. Stimulation with culture medium serves as a negative control. [Figure 7] FIG. 7 illustrates degenerate nucleotide sequences of 9 nucleotides encoding GSG, GTS or GAS linkers. [Figure 8]Figure 8 illustrates the percentage of white plaques recovered by plasmid, shown according to the number of fusion cassettes the construct carries (1, 2, or 3), the number of neopeptides contained in each fusion (Nb neo), the presence or absence of a TM segment (TM), the hydropathy score calculated for each fusion (Hydro), and the presence and length of a linker for each construct, as well as recombinant (mCherry-negative and mCherry-positive) or parental (i.e., non-recombinant) viruses identified by PCR after transfection of CEF cells. DETAILED DESCRIPTION OF THE INVENTION

[0021] Detailed Description of the Invention General definition Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0022] The terms "a" and "an" refer to "one" or "two or more" (i.e., one or more, including two, three, four, five, etc.) of the grammatical object of the article, unless the context clearly indicates otherwise.

[0023] The term "and / or" whenever used herein includes the meaning of "and", "or" and "all or any other combination of the elements connected by said term".

[0024] The terms "such as" and "for example," as used herein, are for illustrative purposes only and therefore not limiting.

[0025] The terms "about" or "approximately" are used herein to indicate that values ​​or ranges set forth herein are not definitive and may vary within 10%, preferably within 8%, and more preferably within 5% of a given value or range to include the inherent variation of error of the device or method used to determine such values ​​or ranges, or the variation that exists between test subjects.

[0026] As used herein, when used in defining products, compositions, and methods, the terms "comprising" (and any form of "comprising," such as "comprise" and "comprises"), "having" (and any form of "having," such as "have" and "has"), "including" (and any form of "including," such as "includes" and "include"), or "containing" (and any form of "containing," such as "contains" and "contain") are open-ended and do not exclude additional, unrecited elements or method steps. "Consisting essentially of" means excluding any essentially significant other ingredient or step. "Consisting of" means excluding more than trace amounts of other ingredients or steps.

[0027] The terms "polypeptide," "peptide," and "protein" are used interchangeably to refer to a polymer of amino acid residues comprising at least nine amino acids covalently linked by peptide bonds. The polymer may be linear, branched, or cyclic, may comprise naturally occurring and / or analogs of amino acids, and may be interrupted by non-amino acids. The maximum number of amino acids contained in a polypeptide is not limited. As a general indication, the term refers to both short polymers (commonly referred to in the art as peptides) and longer polymers (commonly referred to in the art as polypeptides or proteins). The term encompasses, inter alia, native polypeptides, modified polypeptides (also referred to as derivatives, analogs, variants, or mutants), polypeptide fragments, polypeptide multimers (e.g., dimers), recombinant polypeptides, and fusion polypeptides.

[0028] Within the context of the present invention, the terms "nucleic acid," "nucleic acid molecule," "polynucleotide," "nucleic acid sequence," and "nucleotide sequence" are used interchangeably to define a polymer of at least nine nucleotide residues, either deoxyribonucleic acid (DNA) or ribonucleic acid (RNA), or mixed polyribo-polydeoxyribonucleotides. These terms encompass single- or double-stranded, linear or circular, natural or synthetic, unmodified or modified versions (e.g., genetically modified polynucleotides; optimized polynucleotides), sense or antisense polynucleotides, and chimeric mixtures (e.g., RNA-DNA hybrids) thereof. Exemplary DNA nucleic acids include, but are not limited to, complementary DNA (cDNA), genomic DNA, plasmid DNA, vectors, viral DNA (e.g., viral genomes, viral vectors), oligonucleotides, probes, primers, coding DNA, non-coding DNA, or fragments of any of these. Exemplary RNA nucleic acids include, but are not limited to, messenger RNA (mRNA), precursor messenger RNA (pre-mRNA), coding RNA, non-coding RNA, and the like. The nucleic acid sequences described herein may be synthesized by standard methods known in the art, for example, by use of an automated DNA synthesizer (e.g., those commercially available from Biosearch, Applied Biosystems, etc.), or may be obtained from naturally occurring sources (e.g., genomes, cDNA, etc.) or artificial sources (e.g., commercially available libraries, plasmids, etc.) using molecular biology techniques well known in the art (e.g., cloning, PCR, etc.).

[0029] In the context of the present invention, percent identity is determined based on the optimal global alignment of the sequences being compared, i.e., the optimal alignment of the sequences taken in their entirety across their entire length, using any algorithm known to those skilled in the art, such as the algorithm of Needleman and Wunsch (1970). This sequence comparison may be performed using any software known to those skilled in the art, for example, the Needle software, using the "gap open" parameter = 10.0, the "gap extension" parameter = 0.5, the "Blosum 62" matrix, the "end gap penalty" parameter = false, the "end gap open" parameter = 10, and the "end gap extension" parameter = 0.5. The Needle software is available, for example, at https: / / www.ebi.ac.uk / tools / psa / emboss_needle / .

[0030] The terms "virus," "viral particle," "viral vector," and "virion" are used interchangeably and are broadly understood to mean a vehicle comprising one or more elements of a wild-type viral genome that can be packaged into a viral particle. This term encompasses viral genomes and viral particles.

[0031] The terms "obtained from," "originating," or "originate" are used to identify the source of a component (e.g., neoepitope, neopeptide, neoantigen, nucleic acid molecule, virus, etc.) or the source of a sample (e.g., a subject or group of subjects), but are not meant to limit the manner in which the component / sample is obtained (this may be, for example, by chemical synthesis or recombinant means).

[0032] As used herein, the term "isolated" refers to a component (e.g., a polypeptide, nucleic acid molecule, vector, etc.) that is removed from its natural environment (i.e., separated from one or more other components with which it is naturally associated or found in nature). More specifically, it refers to a component that is purified (partially or substantially). For example, a nucleic acid molecule is isolated when it is separated from sequences that are normally associated with it in nature (e.g., dissociated from a chromosome or genome), but may be associated with heterologous sequences (e.g., in a recombinant vector). A synthetic component is essentially isolated.

[0033] The term "subject" generally refers to a vertebrate organism that is in need of or may benefit from any of the products or methods disclosed herein. Generally, the organism is a mammal, particularly a mammal selected from the group consisting of livestock, farm animals, sport animals, and primates (human and non-human). The terms "subject" and "patient" can be used interchangeably when referring to a human organism, and encompass males and females, as well as fetuses, newborns, infants, young adults, adults, and the elderly.

[0034] As used herein, the term "tumor" may be used interchangeably with any of the terms "cancer," "malignant tumor," and "neoplasm," and encompasses any disease or condition resulting from uncontrolled cell growth and spread. These terms are meant to include any type of tissue, organ, or cell, and any stage of malignancy (e.g., pre-lesional to stage IV). Generally, tumors, particularly malignant tumors, exhibit a partial or complete lack of structural organization and functional integrity compared to normal tissue and generally exhibit a tendency to invade (spread) surrounding tissues and / or metastasize to distant sites. The present invention is preferably designed for the treatment of solid tumors as described herein.

[0035] "Neoplastic cell," "cancer cell," or "tumor cell" may be used interchangeably to refer to a cell that is dividing at an abnormal (i.e., increased) rate.

[0036] The term "treatment" (as well as any form of "treatment", such as "treating", "treat"), as used herein, refers to prophylaxis and / or therapy. Generally, "prophylaxis" refers to prophylaxis, e.g., to prevent, delay the onset of, or reduce the severity of, one or more clinical or biochemical symptoms (tumor size, expression level of relevant biomarkers, progression of stage, etc.), while therapy refers to a condition aimed at ameliorating one or more clinical or biochemical symptoms (tumor size, expression level of relevant biomarkers, etc.) in order to slow down or control the progression of the targeted condition, its symptoms, or conditions secondary to the condition, in a subject treated according to the present invention.

[0037] The term "administer" (or any form of "administration," such as "administered"), as used herein, refers to the delivery of components (e.g., at least a poxvirus encoding a neopeptide) to a subject according to the manner described herein.

[0038] Personalized cancer vaccines In a first aspect, the present invention relates to a personalized cancer vaccine comprising a recombinant poxvirus encoding one or more neopeptides.

[0039] "Individualization," as applied herein to characterizing the cancer vaccines of the present invention, refers either to the individual level (a particular subject) or to the subpopulation level (e.g., a small group of people who have a particular disease, a particular phenotypic characteristic, or take the same medication, or who share a common characteristic, such as exhibiting the same deficiency in the immune system).

[0040] poxvirus As used herein, the term "poxvirus" refers to a virus belonging to the Poxviridae family, preferably the Chordopoxvirinae subfamily, which targets vertebrate hosts and includes several genera such as Orthopoxvirus, Capripoxvirus, Avipoxvirus, Parapoxvirus, Leporipoxvirus, and Suipoxvirus. In the context of the present invention, Orthopoxvirus is preferred, and Avipoxviruses include Canarypoxvirus (e.g., ALVAC) and Fowlpoxvirus (e.g., FP9 vector). In a preferred embodiment, the cancer vaccine comprises a poxvirus vector belonging to the Orthopoxvirus genus, and even more preferably, the Vaccinia virus (VV) species. Any vaccinia virus strain can be used in the context of the present invention, examples of which include, but are not limited to, Western Reserve (WR), Copenhagen (Cop), Lister, LIVP, Wyeth, Tashkent, Tian Tan, Brighton, Ankara, MVA (Modified Vaccinia Virus Ankara), LC16M8, LC16M0 strains, etc., with WR, Copenhagen, Wyeth, and MVA vaccinia viruses being particularly preferred. The sequences of the genomes of various Poxviridae are available in the art in specialized data banks such as GenBank (e.g., accession numbers NC_006998, M35027, NC_005309, U94848 provide the sequences of the WR, Copenhagen, canarypoxvirus, and MVA genomes).

[0041] Another suitable embodiment relates to a cancer vaccine comprising a poxvirus vector belonging to the Parapoxvirus genus. Like other members of the Poxviridae family, parapoxviruses are relatively large, enveloped, double-stranded DNA viruses with an oval shape that can infect a wide range of vertebrates, including mammals and humans. Parapoxviruses have a unique helical coat that distinguishes them from other poxviruses. This genus encompasses a series of different species, including parapoxvirus ovis (ORFV), pseudocowpox virus (PCPV), and bovine papular stomatitis virus, as well as their different strains, which may have morphological, structural, and / or genetic differences from one another (e.g., ORFV01701, NZ2, NZ7, and OV-SA00 strains, and bovine papular stomatitis virus BV-AR02 strain). In the context of the present invention, the PCPV species is preferred. PCPV has a genome that is generally 130-150 kilobases long, consisting of linear and double-stranded segments of DNA.

[0042] In the context of the present invention, either wild-type strains and any derivatives thereof (i.e., poxviruses that have been modified compared to the wild-type strain, for example, by truncation, deletion, substitution, and / or insertion of one or more nucleotides, whether adjacent or not, within the viral genome) may be used. The modifications may be within endogenous viral genes (e.g., coding and / or regulatory sequences) and / or intergenic regions. Furthermore, the modifications may be silent or non-silent (e.g., resulting in a modified viral gene product). The modifications may be made in a number of ways known to those skilled in the art using conventional molecular biology techniques. The present invention includes oncolytic (e.g., designed to replicate better or selectively in tumor cells) and replication-deficient poxviruses.

[0043] Preferably, modifications encompassed by the present invention affect, for example, the virulence, toxicity, pathogenesis, or replication of the virus compared to a virus that has not been so modified. Exemplary modifications aimed at altering viral genes are preferably involved in DNA metabolism, host virulence, or the IFN pathway (see, e.g., Guse et al., 2011, Expert Opinion Biol. Ther. 11(5):595-608). A particularly preferred gene to be disrupted is the gene encoding thymidine kinase (TK) (locus J2R; GenBank accession number AAA48082). The TK enzyme is involved in the synthesis of deoxyribonucleotides. Because normal cells generally have low concentrations of nucleotides, TK is required for viral replication in normal cells, but is dispensable in dividing cells containing high nucleotide concentrations. Furthermore, inactivation of the TK gene is known to increase selectivity for tumor cells. Alternatively, or in combination, poxviruses for use herein may be modified by altering one or more genes encoding viral ribonucleotide reductase, or both genes. The viral enzyme has a subunit structure similar to that of mammalian enzymes and is composed of two heterologous subunits designated R1 and R2, encoded by the I4L and F4L loci, respectively. The sequences of the I4L and F4L genes and their locations in the genomes of various poxviruses are available in public databases. Other suitable modifications include those that alter deoxyuridine triphosphatase (F2L), viral hemagglutinin (A56R); serine protease inhibitor (B13R / B14R), and complement 4b-binding protein (C3L). The gene nomenclature used herein is that of the Copenhagen vaccinia strain.Unless otherwise specified, this also applies to homologous genes of other poxviruses herein, and the correspondence between Copenhagen and other vaccinia strains is available to those skilled in the art.For illustrative purposes, vaccinia viruses (VVs) that lack TK, TK- and F2L, and TK- and I4L have been described in the literature (see, for example, WO2009 / 065547 and WO2009 / 065546).

[0044] In a preferred embodiment, the poxvirus included in the cancer vaccine of the present invention is a replication-deficient poxvirus, preferably a replication-deficient vaccinia virus, which means that it cannot replicate to a significant extent in human cells. Poxvirus vectors can be made replication-deficient by partial or complete deletion or inactivation of regions essential for viral replication, and impaired or defective replication function can be assessed by conventional means, for example, by measuring DNA synthesis and / or viral titer in non-permissive cells.

[0045] A particularly suitable poxvirus vector for use in the context of the present invention is MVA due to its highly attenuated phenotype (Mayr et al., 1975, Infection 3: 6-14; Sutter and Moss, 1992, Proc. Natl. Acad. Sci. USA 89: 10847-51) and the more pronounced IFN1-type response that occurs upon infection compared with non-attenuated poxviruses. For illustrative purposes, MVA has been generated through serial passage in chicken embryo fibroblasts. Sequence analysis of its genome showed that it has lost the pathogenicity of its parent virus, chorioallantoic vaccinia virus Ankara, through alterations in its genome (Antoine et al., 1998, Virol. 244: 365-96 and GenBank accession number U94848). MVA has been used safely and effectively to vaccinate over 100,000 people against smallpox. This virus is also replication-deficient in human cells, but not in chicken embryo cells. Various cell lines are available in the art for producing large amounts of virus, particularly in egg-based production methods (e.g., WO2007 / 147528).

[0046] The term "recombinant" refers to a poxvirus that comprises one or more exogenous nucleic acid molecules (e.g., one or more nucleic acid molecules encoding neopeptides) inserted into its genome, as described below.

[0047] Neopeptide type and number For purposes of clarity, an "antigen" generally refers to a substance (e.g., a polypeptide) that is capable of generating a humoral or specific T cell response (or both) against that antigen, including a CD4+ (e.g., Th1, Th2, and / or Th17) and / or a CD8+ T cell response (e.g., a CTL response). As described herein, a wide variety of direct or indirect biological assays are available in the art for assessing the immunogenic properties of antigens either in vivo (animals or humans) or in vitro (e.g., biological samples).

[0048] To become detectable by the immune system, antigens are processed, and this processing requires the fragmentation of the antigen into peptides, the association of the peptides with MHC (major histocompatibility complex; HLA "human leukocyte antigen" in humans), and the presentation of peptide-MHC complexes on the cell surface where they can be recognized by T cells via the T cell receptor (TCR). These antigen-derived peptides contain the minimal immune determinants (i.e., epitopes). There are two types of MHC molecules: MHC class I and MHC class II. MHC class I molecules are present on the surface of all nucleated cells in mammalian subjects, and peptide / MHC class I complexes are expressed by CD8 +Induce CTL activation. In contrast, MHC class II molecules are usually only present on the surface of specialized antigen-presenting cells, such as dendritic cells, mononuclear phagocytes, some endothelial cells, thymic epithelial cells, and B cells, and peptide presentation via MHC class II molecules generally drives CD4+ T cell responses. Epitopes for presentation via MHC class I molecules are generally at least 8 amino acids long, preferably 8, 9, or 10 amino acids, while MHC class II epitopes are usually longer (e.g., at least 13 amino acid residues) (Rock et al., 2010, J. Immunol. 184(1): 9-15). Regarding in silico prediction of MHC class I and class II binding, many prediction algorithms exist in the art (see, for example, Nielsen et al., 2010, Immunology 130(3): 319-28). For illustrative purposes, SVMHC, NetMHCII, Tepitope / propped, syfpeithi, Epitollkit, etc. may be mentioned.

[0049] The term "neoantigen," as used herein, refers to an antigen that appears during the carcinogenesis process in cancer cells. In a preferred embodiment, the neoantigen comprises one or more non-silent mutations of amino acid residues relative to the corresponding wild-type antigen. Generally, non-silent mutations occur at the nucleotide level and translate into changes at the amino acid level. In another preferred embodiment, the neoantigen is found in cancer cells or tissues obtained from a patient, but is not found in normal cell or tissue samples obtained from the patient or healthy individuals.

[0050] The term "neopeptide" refers to a fragment of a neoantigen comprising a neoepitope and flanking sequences on one or both sides of the neoepitope. The flanking sequences are those of the neoantigen from which the neoepitope is derived. Thus, a neopeptide corresponds to a fragment of a neoantigen that is longer than the neoepitope itself presented by an MHC molecule. In a neopeptide, the neoepitope is thus present and surrounded (on one or both sides) by flanking sequences that naturally occur in its normal environment. Except for the rare case where a neopeptide is also present in another self-protein, neopeptides will in most cases be (and can be selected to be) of a non-self nature; see below. Due to this non-self nature, such neopeptides are expected to be recognized by tumor-specific T lymphocytes.

[0051] The term "neoepitope," as used herein, refers to the minimal immunodeterminant of a neoantigen that contributes to MHC-dependent T cell recognition (or is presented by an MHC molecule on the surface of a cell of a subject), and comprises one or more non-silent mutations identified in said neoantigen according to the manner described herein.

[0052] The term "mutation" refers to one or more sequence differences between a test sequence and a reference sequence.

[0053] In one embodiment, each of the one or more neopeptides encoded by the recombinant poxvirus is specific for the patient's tumor and comprises one or more tumor-specific mutations from the set of mutations described herein. In the context of the present invention, "tumor-specific mutations" refer to mutations that occur with the neoplastic transformation and / or progression of cancer cells. Generally, tumor-specific mutations are preferably present in DNA contained in cancer cells (e.g., tumor samples) but not in DNA contained in non-cancerous cells (e.g., non-tumor samples). Although occurring at the nucleotide level, in the context of the present invention, tumor-specific mutations are not silent and translate into changes at the amino acid level.

[0054] Several types of tumor-specific mutations are encompassed by the present invention, including missense mutations, deletions, insertions, frameshift mutations, and mutations at splicing sites. "Missense" mutations occur due to the substitution of one nucleotide with another within a specific codon, affecting the encoded amino acid sequence, thus resulting in a single amino acid change. Another way to affect protein sequence is the insertion or deletion of one or more nucleotides (e.g., a small piece of DNA), which changes the number of nucleotides within a nucleic acid molecule. Insertion and deletion mutations can, but do not necessarily, result in a change in the reading frame (so-called frameshift mutations) (e.g., the insertion or deletion of three nucleotides results in the addition or suppression of a codon). If a mutation occurs early in the nucleotide sequence, almost the entire polypeptide may be altered. Frameshift mutations can also result in the generation of a stop codon that is translated into a stop signal, which then truncates the resulting protein (deletion of the portion downstream of the de novo generated stop codon). Missense mutations can also be located at the splicing sites of mRNA, resulting in abnormal splicing and thus abnormal protein sequences. The term "splicing" refers to the editing of nascent precursor messenger RNA (pre-mRNA) transcripts. After splicing, introns are removed and exons are joined (ligated) together. The term "splice site" refers to the short conserved sequence located at the end of an intron, which is crucial for intron recognition and the accuracy of the splicing reaction.

[0055] The present invention encompasses neopeptides derived from all protein classes, including enzymes, receptors, transcription factors, etc. Any type of mutation may be present; however, tumor-specific mutations are preferably missense or frameshift. Advantageously, at least 60%, desirably at least 65%, preferably at least 70%, more preferably at least 75%, at least 80% (e.g., 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, etc.), at least 85% (e.g., 85%, 86%, 87%, 88%, 89%, 90%, 91%, etc.), of the neopeptides for use herein. At least 90% (e.g., 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or even 100%), at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or even 100% comprise missense or frameshift mutations. Of the missense and frameshift mutations, missense mutations are more frequent. In certain embodiments, at least 60%, desirably at least 65%, preferably at least 70%, more preferably at least 75% and even more preferably at least 80% (e.g., 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, etc.), at least 85% (e.g., 85%, 86%, 87%, 88%, 89%, 90%, 91%, etc.), of the neopeptides for use herein. 9%, 90%, 91%, etc.), at least 90% (e.g., 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or even 100%), at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or even 100% comprise missense mutations. However, if a frameshift mutation is identified, it is also preferred that one or more of the neopeptides encoded by the recombinant poxvirus comprise a frameshift mutation.For example, for 10 neopeptides encoded by a recombinant poxvirus, preferably at least 6, at least 7, at least 8, at least 9, or even 10 of the neopeptides are missense or frameshift mutations, and more preferably at least 6, at least 7, at least 8, at least 9, or even 10 of the neopeptides are missense mutations. Similarly, for 30 neopeptides encoded by a recombinant poxvirus, preferably at least 18, at least 19, at least 20, at least 21, at least 22, at least 23, at least 24, at least 25, at least 26, at least 27, at least 28, at least 29, or even 30 of the neopeptides are missense or frameshift mutations, and more preferably at least 18, at least 19, at least 20, at least 21, at least 22, at least 23, at least 24, at least 25, at least 26, at least 27, at least 28, at least 29, or even 30 of the neopeptides are missense mutations. Those skilled in the art will readily adapt these examples to poxviruses encoding other numbers of neopeptides.

[0056] The length of neopeptides encoded by recombinant poxviruses generally ranges from 13 to about 151 amino acid residues, although the length may vary depending on the neopeptide. Advantageously, each neopeptide for use herein has a length of 15 to 101 amino acid residues. Desirably, it has a length of 16 to 90 amino acid residues, preferably 17 to 85 amino acid residues, and more preferably 18 to 80 amino acid residues. Advantageously, at least 70% of neopeptides for use herein have an odd number of amino acids, since in the case of a single missense mutation, this allows for the insertion of the mutated amino acid in the center of the neopeptide with the same number of flanking amino acids on either side of the mutated amino acid. Of course, the length of the neopeptide may depend on the type of mutation. For illustrative, but not limiting purposes, neopeptides comprising missense mutations preferably have a length of 18 to 29 residues, while neopeptides comprising frameshift mutations preferably have a length of 30 to 80 amino acid residues. In the context of the present invention, at least 60%, preferably at least 70%, more preferably at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or even 100% of the neopeptides encoded by the poxvirus cancer vaccines of the present invention preferably have a length of 18 to 29 residues. Individual neopeptides of 19, 21, 26, 27, or 29 residues are particularly preferred. In particularly preferred embodiments, at least 60%, preferably at least 70%, more preferably at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or even 100% of the neopeptides encoded by the poxvirus cancer vaccines of the invention comprise missense mutations and have a length of 18 to 29 residues, particularly 19, 21, 26, 27, or 29 residues.

[0057] With regard to the position of the mutation, the present invention contemplates any position within the neopeptide. However, a central position is particularly preferred for missense mutations. By "central position" is meant that the mutated amino acid is located exactly in the center of the neopeptide (if the neopeptide has an odd number of amino acids), or in one of two central positions (if the neopeptide has an even number of amino acids), or exactly 2 to 5 amino acids to either side of the central position, depending on the length of the neopeptide. Desirably, at least 80%, preferably at least 85%, more preferably at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99%, or even 100% of neopeptides with missense mutations have a substituted amino acid at the central position as defined above. Specifically preferred are neopeptides of 19, 21, 27 or 29 amino acids (even more preferably 27 or 29 amino acids) that have a missense mutation at the exact central position (e.g., position 10 for a 19mer neopeptide, position 11 for a 21mer neopeptide, position 14 for a 27mer neopeptide, and position 15 for a 29mer neopeptide).

[0058] However, mutations may be located near the N- or C-terminus in at least some neopeptides, particularly for frameshift mutations or when missense mutations occur at or near the N- or C-terminus of the neoantigen.

[0059] Some embodiments also contemplate expression of neopeptides by recombinant poxviruses in the form of one or more fusions. The term "fusion," as used herein, refers to the combination of two or more neopeptides in a single polypeptide chain. Fusions can be direct (i.e., no additional amino acid residues between) or via a linker to improve accessibility of the neopeptides. In addition to a series of unique neopeptides expressed individually, the present invention therefore also encompasses recombinant poxviruses that express several neopeptides in the form of fusions and several others individually. Alternatively, all neopeptides expressed by a poxvirus vector are clustered into one or more fusions.

[0060] Certain embodiments contemplate the presence of a signal peptide at the N-terminus of the encoded neopeptide (or a fusion thereof) to enhance processing and / or secretion through the endoplasmic reticulum (ER). Briefly, a signal peptide, typically comprising 15-35 essentially hydrophobic amino acids, is inserted at the N-terminus of a polypeptide downstream of the codon for initiating translation and is then removed by specific ER-located endopeptidases to yield the mature polypeptide. Membrane anchoring can also be used to enhance MHC class I and / or MHC class II presentation of the encoded neopeptide (or a fusion thereof) by further incorporating a membrane anchor sequence. Transmembrane peptides are highly hydrophobic in nature and serve to anchor the polypeptide within the cell membrane. They are preferably inserted within the C-terminal end of the polypeptide, preferably immediately upstream of the stop codon. Suitable transmembrane and / or signal peptides are known in the art. They may be derived from cellular or viral polypeptides, such as those of immunoglobulins, tissue plasminogen activator, insulin, rabies glycoprotein, HIV virus envelope glycoprotein, or measles virus F protein, or may be synthetic. Preferred signal peptide and membrane anchor sequences are described in WO2008 / 138649, with those derived from rabies or measles F glycoproteins being preferred. If more than one signal peptide and / or transmembrane sequence is used in the recombinant poxvirus, those of different origin may be selected, and homologous sequences showing a high degree of sequence identity (e.g., greater than 75%) may be modified to limit homologous recombination events that could compromise the production process.

[0061] Alternatively or in combination with the preceding embodiments, the present invention contemplates the presence of one or more linkers (also called spacers), particularly in neopeptide fusions. Generally, linkers are peptides of 1 to 30 amino acids in length composed of amino acid residues such as glycine, serine, threonine, asparagine, alanine, and / or proline. Preferred linkers in the context of the present invention comprise 2 to 15 amino acids, preferably 3, 5, or 10 amino acids, primarily glycine and serine (e.g., GSG, GSGSG (SEQ ID NO: 66), SGSGS (SEQ ID NO: 67), GSGSGSGSGS (SEQ ID NO: 68)), or glycine, serine, and threonine (e.g., GSTSG (SEQ ID NO: 69), SGTGS (SEQ ID NO: 70)), or glycine, serine, and threonine and / or alanine (e.g., GAS, GTS). Preferred three-amino acid linkers include, inter alia, linkers of the sequences GSG, GAS, and GTS. It is within the skill of one in the art to assess whether or not it is necessary to include a linker between the two fusion neopeptides.

[0062] In a preferred embodiment, the neopeptides are arranged in fusions with linkers between each neopeptide (e.g., between neopeptide 1 and neopeptide 2, between neopeptide 2 and neopeptide 3, etc.) and, optionally, at the N-terminus of the first neopeptide and / or the C-terminus of the last neopeptide. When several linkers are used within one construct, the amino acid sequence or codon nucleic sequence may be varied to reduce the percentage of identity at the nucleic acid level (e.g., to less than 77% identity, desirably less than 75%, less than 70%, preferably less than 60%, less than 50%, and even more preferably less than 35%, especially for linkers of 10 amino acids), thus limiting recombination events during the production process. In a preferred embodiment, the linker nucleic acid sequences included in a single recombinant poxvirus are designed (by altering the amino acid sequence or codon nucleic acid sequence) so that the portion of the nucleic acid sequence identical between any two linkers is reduced to a maximum of eight contiguous nucleotides (e.g., preferably, a maximum of seven, more preferably, a maximum of six contiguous nucleotides), thereby contributing to reducing the identity of the nucleic acid sequences and thus limiting undesired recombination events during production. As described in the Examples section, short linkers of 2 to 5 amino acid residues (e.g., 2, 3, 4, or 5 amino acids) are advantageous, especially when the recombinant poxvirus encodes 10 or more neopeptides. Three-amino acid linkers, such as GSG, GAS, and / or GTS, are particularly preferred in the context of the present invention to separate neopeptides from each other. Exemplary nucleotide sequences encoding three-amino acid linkers for use individually or in combination with the neopeptide fusions described herein are shown in Figure 7.For illustrative purposes, 15 GSG-encoding nucleotide sequences have been generated and modified by taking advantage of codon degeneracy (four possible codons for G and six for S) to reduce the risk of homologous recombination in recombinant poxviruses that can occur between stretches of identical or similar sequences. Similarly, 10 and 5 modified nucleotide sequences encoding the GTS and GAS linkers, respectively, are shown in Figure 7.

[0063] Certain embodiments of the present invention contemplate the presence of a tag to facilitate detection of expression of a neopeptide (or fusion thereof) or infected host cells expressing such a neopeptide or fusion. The tag peptide can be detected by immunodetection assays using anti-tag antibodies. A wide variety of tag peptides can be used in the context of the present invention, including, but not limited to, PK tags, FLAG tags (DYKDDDK; SEQ ID NO: 6), MYC tags (EQKLISEEDL; SEQ ID NO: 11), polyhistidine tags (usually a stretch of 5-10 histidine residues), HA tags (YPYDVPDYA; SEQ ID NO: 18), HSV tags (QPELAPEDPED; SEQ ID NO: 19), and VSV tags (YTDIEMNRLGK; SEQ ID NO: 20). The tag peptide may be independently located at the N-terminus of the neopeptide or its fusion (tag-polypeptide), at its C-terminus (polypeptide-tag), or internally.

[0064] The number of neopeptides that can be encoded by a recombinant poxvirus is not limited, depending on the type of expression selected (individual expression, expression in short or large fusions as described below) and the type of poxvirus. For illustrative purposes, 1 to 50 neopeptides can be expressed by a recombinant poxvirus, more preferably 5 to 45, even more preferably 6 to 35, with 10 to 30 neopeptides being most preferred.

[0065] In a preferred embodiment, some, preferably all, neopeptides are expressed by the recombinant poxvirus in the form of one or more fusions. Preferably, the poxvirus included in the personalized cancer vaccine of the present invention encodes 1 to 5 fusions of two or more neopeptides, preferably 1 to 4 fusions of two or more neopeptides, more preferably 1 to 3 fusions of two or more neopeptides, and particularly preferably 2 or 3 fusions of two or more neopeptides. More preferably, each fusion comprises 2 to 15 neopeptides, preferably 3 to 12 neopeptides, more preferably 4 to 11 neopeptides, and even more preferably 5 to 10 neopeptides (e.g., 5, 6, 7, 8, 9, or 10). The number of neopeptides can vary depending on the fusion encoded by the recombinant poxvirus. However, in a particularly preferred embodiment, the poxvirus included in the personalized cancer vaccine of the present invention encodes 2 or 3 fusions of 5 to 10 neopeptides.

[0066] In the context of the present invention, each fusion may be designed differently from the others and may be distinguished from one another by the presence and / or sequence and / or number and / or location of elements such as peptide signals, linkers, tags, etc. According to a preferred embodiment, however, each fusion comprises a) a signal peptide at its N-terminus, b) linkers at the N-terminus of the first neopeptide, between each neopeptide, and at the C-terminus of the last neopeptide, and c) a tag at its C-terminus.

[0067] Prediction of the presence of TM (transmembrane) segments in neopeptides and their fusions. In one embodiment, the amino acid sequence of one or more neoepitopes or fusions thereof encoded by the recombinant poxvirus does not contain any potential transmembrane segments. A TM segment can be defined as a short, hydrophobic α-helix of approximately 20 residues (e.g., 18-30 residues, preferably 19-21 residues). Several prediction tools can be used to predict the probability that a given sequence (e.g., a specific neopeptide or neopeptide fusion) contains a TM segment, including TMHMM (Transmembrane Hidden Markov Model; Krogh et al., 2001, J. Mol. Biol. 305: 567-80) and DAS (Dense Alignment Surface). For example, the DAS-TM selection algorithm provides a highly accurate hydrophobic profile for a query, from which the location of potential transmembrane segments can be obtained.

[0068] In a preferred embodiment, at least 80% (e.g., at least 85%, at least 90%, at least 95%, at least 98%, and even 100%) of the neopeptides encoded by the poxviruses included in the personalized cancer vaccine of the present invention do not contain potential TM segments (intra-peptide TM). In this context, neopeptides predicted to comprise a TM segment are preferably not selected.

[0069] Even more preferably, the prediction of the presence of a TM segment is also applied to each neopeptide fusion. It has been observed in practice that a TM segment can arise from the junction of two specific neopeptides (interpeptide TM). In this case, modifying the order of the neopeptides in the fusion is an option to eliminate the presence of a TM. For example, if a TM segment arises from the fusion of neopeptide 1 with the N-terminus of neopeptide 2, inverting the neopeptides (fusion of neopeptide 2 with the N-terminus of neopeptide 1) can eliminate the risk of having a TM segment. Another option could be to add a suitable linker between the two fused neopeptides.

[0070] Prediction of hydrophobicity and hydropathy scores of neopeptides and their fusions In one embodiment, the amino acid sequence of one or more neoepitopes or fusions thereof encoded by the recombinant poxvirus is hydrophilic in nature. The hydrophilic or hydrophobic nature of a given sequence can be easily determined by numerous methods and algorithms available in the art. Calculating the hydrophobicity score and / or hydropathy score of a particular sequence is within the skill of a person skilled in the art. For example, these scores can be determined using the Kyte-Doolittle method (Kyte and Doolittle, 1982, J. Mol. Biol. 157: 105-32) or any other suitable method (e.g., Rose et al., 1993, Ann. Rev. Biomol. Struc. 22: 381-415; Kallol et al., 2003, J. Chromat. 1000: 637-55; Sweet et al., 1983, J. Mol. Biol. 171: 479-88, among others) or algorithm (e.g., ExPAsy Prot Scale Protein; Protein Hydrophobicity Plots developed by Colorado State or World of Bioinformatics, etc.). In a general method, the hydrophobicity score of a given sequence is determined by the sum of the hydrophobicity / hydrophilicity values ​​of each amino acid residue, as summarized in Table 1 below.

[0071] [Table 1]

[0072] As illustrated in Table 1, amino acids I, V, L, F, C, M and A exhibit positive hydrophobicity scores (e.g., 4.5 for I) that correlate with the hydrophobic nature of these residues, while G, T, S, W, Y, P, H, E, Q, D, N, K and R exhibit negative hydrophobicity scores (e.g., −4.5 for R) that correlate with their hydrophilic nature.

[0073] The hydrophobicity score of a given neopeptide is determined by the sum of the hydrophobicity / hydrophilicity values ​​of each amino acid residue contained in the neopeptide, and the hydropathy score is calculated by dividing the hydrophobicity score determined for the neopeptide by the number of residues present in the peptide. The hydrophobicity score of a given neopeptide fusion corresponds to the sum of the hydrophobicity scores determined for each neopeptide contained in the fusion. The hydropathy score of a given neopeptide fusion is calculated by dividing the hydrophobicity score determined for the neopeptide fusion by the number of residues present in the neopeptide fusion.

[0074] In a preferred embodiment, at least 60% of the neopeptides encoded by the recombinant poxviruses included in the personalized cancer vaccine of the present invention exhibit a negative hydrophobicity score and / or a hydropathy score of 0.1 or less (e.g., 0.1, 0.09, 0.08, etc., or even a negative score, e.g., -0.019, -1.5, etc.). Similarly, each fusion encoded by a recombinant poxvirus for use herein preferably exhibits a negative hydrophobicity score and / or a hydropathy score of 0.1 or less. "At least 60%" encompasses at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 98% and even 100%.

[0075] For illustrative purposes, the neopeptide of sequence GQSLPMTHSLKLSKTNRTLFLLGVTKY (SEQ ID NO: 58) has a negative hydrophobicity score of −3.6 (plus −0.4 for G, −3.5 for Q, −0.8 for S, +3.8 for L, etc.) and a hydropathy score of −0.13 (−3.6 / 29). This peptide is therefore suitable for expression by a recombinant poxvirus included in a personalized cancer vaccine of the present invention. In contrast, the neopeptide GLMGIVVGTVFIIRGLRSVGASRHQGL (SEQ ID NO: 59) has a positive hydrophobicity score (23.3) and a hydropathy score of 0.86. This peptide is therefore not suitable for expression by a recombinant poxvirus included in a personalized cancer vaccine of the present invention, unless other neopeptides included in the fusion complement it, allowing the global score of the fusion to be reduced to the appropriate thresholds described herein.

[0076] In a particularly preferred embodiment, the recombinant poxvirus comprised in the personalized cancer vaccine of the present invention encodes two or three fusions, wherein: Each fusion comprises 5 to 10 neopeptides; with a linker, preferably 3 amino acids in length, at the N-terminus of the first neopeptide and between each neopeptide of the fusion (e.g., one or more of those shown in Figure 7); each fusion comprises a signal peptide at the N-terminus of the fusion; each fusion optionally comprises a tag sequence at the C-terminus of the fusion; and Each fusion is exhibit a global negative hydrophobicity score and / or a global hydropathy score of 0.1 or less, and Does not contain any potential TM segments.

[0077] Identification of neoantigens and neopeptides encoded by personalized cancer vaccines The development of personalized cancer vaccines requires the identification and selection of neopeptides for each patient.

[0078] Thus, the present invention relates to a method for preparing a personalized cancer vaccine comprising identifying one or more neopeptides suitable to be encoded by a recombinant poxvirus, wherein each of said neopeptides comprises one or more tumor-specific mutations.

[0079] For this purpose, tumor biological sample and non-tumor biological sample are obtained from the subject to be treated.The term " biological sample " as used herein refers to the sample obtained from subject that can be collected or recovered in vivo.In the context of the present invention, the biological sample that is collected or recovered from subject contains DNA, and preferably also contains mRNA or protein.

[0080] In one embodiment, the biological sample is obtained from a bodily fluid or tissue sample. Exemplary bodily fluid samples include, but are not limited to, blood, serum, plasma, urine, feces, cerebrospinal fluid, and bronchial fluid, as well as their derivatives (e.g., partially purified blood, PBMCs). Tissue samples can be, but are not limited to, organs, tissues, fractions, and cells isolated from mammals of any origin or cell type (e.g., samples comprising tissues from the oral cavity, gastrointestinal tract, skin, lungs, head, etc.). Exemplary tissue samples include, for example, cell lysates, cell cultures, tissues, fresh-frozen tissues, cytological materials (e.g., cervical smears, vaginal swabs, bronchial scrapings, pleural fluid, colonoscopy, etc.), tissues obtained after surgical tumor removal, and tissue biopsies. In the context of the present invention, the biological sample may be in suspension or solution, or may remain attached to a solid support (e.g., a slide, beads, chips, or nanoparticles). The present invention also encompasses "purified" or semi-"purified" biological samples, such as isolated RNA, cDNA, isolated protein, and the like.

[0081] In a preferred embodiment, the tumor sample is preferably a tumor biopsy, which preferably comprises more than 20% tumor cells. It can be prepared in any manner, including but not limited to, dissociated tissue specimens such as fine needle aspirates or fresh solid tumors (which can be enzymatically digested), or microtome sections of biopsies. In another embodiment, the tumor sample is cytological material (e.g., cervical or vaginal swabs).

[0082] In yet another embodiment, the non-tumor sample is obtained from a biological fluid (e.g., blood with or without PBMC isolation), cytological material (e.g., a buccal swab), a biopsy (e.g., a biopsy of healthy tissue), or a small piece of surgical resection comprising non-tumor cells. The non-tumor sample can be taken or collected from the same patient or a reference healthy subject or group of subjects.

[0083] If tumor and non-tumor samples are from the same sampling, the skilled artisan knows how to recognize and separate tumor cells from non-tumor cells (eg, by immunohistology).

[0084] In one embodiment, the identifying step comprises: a) extracting DNA from said tumor sample and said non-tumor sample; b) selecting a target region, preferably the entire coding region of the genome (exome); c) sequencing the target region (e.g., exome) from the extracted DNA; and d) identifying one or more tumor-specific mutations by comparing DNA sequences obtained from said tumor and non-tumor samples; The compound comprises:

[0085] The term "exome" refers to the portion of the genome formed by exons (i.e., the sequence remaining in mRNA after intron removal by RNA splicing when transcribed). For illustrative purposes, the exome of the human genome consists of approximately 180,000 exons, which corresponds to approximately 1% of the whole genome. The technology for exome sequencing is well known in the art (Anderson and Schrijver, 2010, Genes 1(1): 38-69). Generally, the method involves the following two-step approach: first, targeted capture of coding sequences using array-based hybridization (commercially available from Agilent technologies) or liquid-based hybridization (commercially available from Roche as NimbleGen; Bainbridge et al., 2010, Genome Biol.11:R62), and this capture step is followed by a sequencing step (Ng SB et al., 2009, Nature 461: 272-6). A number of technical approaches are commercially available for sequencing DNA sequences, including pyrosequencing (commercially available as Roche454), dye sequencing (commercially available through Illumina), and digital sequencing (commercially available as IonTorrent sequencing), among others.

[0086] Although the step of identifying tumor-specific mutations is essential, further optional selection may be used to select optimal neopeptides encoded by the personalized recombinant poxvirus and / or to combine them in suitable neopeptide fusions. Indeed, many tumor-specific mutations may accumulate during transformation and cancer progression, and some may have great potential to stimulate the immune system of a subject being treated for their cancer. If additional selection is used, the identification step preferably comprises, in addition to the above substeps a) to d), one or more of the following further substeps e) to i): e) ranking potential neopeptides by their expression level in the tumor (or the expression level of the corresponding neoantigen), either at the mRNA transcription level or at the protein translation level; f) selecting non-self-expressing tumor-specific mutations; g) predicting the immunopotency of neoepitopes contained in neopeptides either by in silico prediction or immunogenicity biological testing; h) predicting the presence of potential TM segments in the neoepitope itself (intra-peptide TM) and / or in neopeptide fusions comprising several neopeptides (inter-peptide TM in the fusion region, i.e., the region covering the C-terminus of the first neopeptide and the N-terminus of the second neopeptide, and optionally a linker between the first and second neopeptides); i) Ranking potential neopeptides and / or neopeptide fusions by degree of hydrophobicity, and in particular selecting neopeptides and / or neopeptide fusions that exhibit a negative global hydrophobicity score and / or a global hydropathy score of 0.1 or less.

[0087] As mentioned above, one or more of the further sub-steps e) to i) may be present in the identification step. Several of the further sub-steps e) to i) may also be present. In particular, in a preferred embodiment, the identification step comprises, in addition to the above sub-steps a) to d), the above sub-steps e) and f) (these sub-steps are described in more detail below), and optionally Sub-step g) (also described in more detail below), Sub-step h) (also described in more detail below), Sub-step i) (also described in more detail below), or Any combination of substeps g), h) and i) (e.g., substeps g) and h), substeps g) and i), substeps h) and i), or substeps g), h) and i)). It further comprises:

[0088] Sub-process e) Further selection can be achieved by ranking neoantigens by their expression level in tumors. Thus, in a preferred embodiment, the identification step further comprises substep e) of selecting neopeptides that are at least expressed, preferably highly expressed, in tumors. Expression can be detected or quantified at the mRNA or protein level. In one embodiment, neopeptides can be selected based on their detection that they are at least transcribed, preferably highly transcribed, in tumors, i.e., expressed tumor-specific mutations (RefSeq transcripts or GenBank) found in RNA tumor samples and occurring in ORFs (if translated). For illustrative purposes, the expressed tumor-specific mutations are generally identified by sequencing complementary DNA (cDNA) derived from RNA extracted from the tumor biological sample obtained from the patient, i.e., the transcriptome. As used herein, transcriptome refers to the entire range of messenger RNA (mRNA) expressed by an organism. Techniques for transcriptome sequencing are well known in the art (e.g., Chu and Corey, 2012, Nucleic Acid Ther. 22: 271-4; Wang et al., 2009, Nat Rev Genet. 10(1): 57-63). Preferably, the mRNA encoding the mutant neoantigen is abundantly expressed in RNA extracted from the tumor sample. It is not possible to give an absolute expression preference for neopeptide selection, but rather the selection is relative, and potential neopeptide expression is detected / quantified and ranked in order of priority. Those skilled in the art know how to compare the expression levels of various candidate neopeptides in the tumor of a subject to be treated and rank them in descending order of expression level, with highly expressed neopeptides ranked first and non-expressed neopeptides at the end of the list.

[0089] Alternatively or in combination with mRNA expression, neopeptide selection may be performed based on the expression levels of peptides resulting from tumor-specific mutations using well-described proteomic techniques such as Western blot, immunoassays, and mass spectrometry. Here too, the selection is rather relative, and potential neopeptide expression is detected / quantified and ranked by priority, which is also easy for one skilled in the art.

[0090] Sub-process f) Alternatively or in combination with substep e), further selection can be performed by selecting non-self neopeptides, meaning neopeptides that do not exist as such in the subject's proteome. While the presence of tumor-specific mutations ensures that the same peptide does not exist in the corresponding wild-type protein expressed in the subject's normal cells, substep f) aims to confirm that the same peptide is not part of another protein expressed in the subject's normal cells. For illustrative purposes, polypeptides identified as mutated are truncated into multiple minimal immunodeterminants (e.g., one-amino acid 9-mer overlaps) and aligned with the host proteome (e.g., the human proteome for a human subject) corresponding to the proteins expressed in the host using blastp (Altschul et al., 1990, J. Mol. Biol. 215(3):403-10). Mutant epitopes that are not shared with the host protein are then selected. In other respects, self-epitopes (i.e., present in the proteome) are eliminated to avoid autoimmune problems and the risk of immune tolerance to self-peptides, which may diminish the immune response. As mentioned above, substep f) may preferably be combined with substep e), thus resulting in the selection of (highly) expressed non-self neopeptides.

[0091] Sub-process g)Alternatively or in combination with substeps e), f), and a combination of e) and f), further screening can be performed by predicting the immunopotency of the selected neopeptides. Such immunopotency-based selection can be performed either by in silico prediction and / or bioanalytical approaches (e.g., immunogenicity biological testing). In silico predictors are constructed using machine learning algorithms, such as artificial neural networks (ANNs) or support vector machines (SVMs), and trained to identify highly immunogenic sequences. This algorithm is then used to classify epitopes with respect to their affinity for HLA class I and / or class II proteins as an indicator of potential presentation. For illustrative purposes, this further and optional step may include HLA typing and prediction of MHC class II binding (truncating the mutant antigen into multiple peptides / epitopes (usually one amino acid, overlapping 8-10 mers for class I binding and 10-20 mers for class II binding) using training algorithms such as seq2HLA (Boegel et al., 2015, Methods Mol Biol 1310: 247-51), NetMHCIIpan (Andreatta et al., 2015, Immunogenetics 67(11-12): 641-50) and IEDB consensus binding prediction). For illustrative purposes, mutant peptides with low scores are preferably selected. Such peptides are otherwise predicted to have higher MHC class binding scores and therefore a higher probability of generating a T cell response.

[0092] Alternatively or in addition to in silico prediction of immune competence, immunogenic neopeptides may be identified by suitable bioanalytical tests known in the art, such as mass spectrometry, cytokine release assays (e.g., IFNγ ELIspot in appropriate animal models), MHC class I or class II binding assays to identify MHC-restricted mutations, flow cytometry, immunofluorescence staining, histochemistry, etc.

[0093] As mentioned above, substep g) may preferably be combined with substep e), substep f) or both substeps e) and f), more preferably both substeps e) and f), thus resulting in the selection of (highly) expressed non-self and immunogenic neopeptides.

[0094] Sub-process h) Alternatively or in combination with substep e), substep f), or substep g), or any combination thereof (e.g., substeps e) and f), or substeps f) and g), or substeps e) and g), or substeps e), f) and g), further screening can be performed by predicting the presence of potential TM segments in the selected neopeptide (intra-peptide TM) or the resulting neopeptide fusion (inter-peptide TM). As mentioned above, the selected neopeptide and the resulting neopeptide fusion preferably do not contain a TM segment to facilitate the generation of recombinant poxviruses and the proper presentation of neoepitopes to the immune system. As mentioned above, several prediction tools can be used to predict the presence of TM segments in a given peptide, such as TMHMM (Krogh et al., 2001, J. Mol. Biol. 305: 567-80) and DAS (Dense Alignment Surface). In the case of predicting inter-peptide TM at the junction of two neopeptides, inverting the neopeptide may be an option to eliminate the presence of a TM. Another option could be to add a suitable linker between the two neopeptides.

[0095] Sub-process i)Alternatively or in combination with substep e), substep f), substep g), or substep h), or any combination thereof (e.g., substeps e) and f), or substeps f) and g), or substeps e) and g), or substeps g) and h), or substeps e), f) and g), or substeps e), f), g), and h), further selection may be added by ranking the neopeptides and / or neopeptide fusions according to their degree of hydrophobicity or hydrophilicity. Indeed, although neopeptide fusions may tolerate the presence of one or more short stretches of hydrophobic amino acid residues, it is preferred that at least 60% of the selected neopeptides are essentially hydrophilic, as described above, to ensure that the resulting neopeptide fusions have a negative global hydrophobicity score and / or a global hydropathy score of less than 0.1. Such predictions may be made by conventional methods and algorithms known in the art, such as those described above (e.g., Kyte and Doolittle, 1982, J. Mol. Biol. 157: 105-32 or the ExPAsy Prot Scale Protein algorithm).

[0096] Generation of recombinant poxviruses In addition to the identification step, the method for preparing a personalized cancer vaccine according to the present invention may comprise a step of generating a recombinant poxvirus. Generally, such a step comprises generating a synthetic nucleic acid encoding a selected neopeptide described herein (e.g., optionally arranged in one or more fusions with a peptide signal, linker, tag, etc.), generating a recombinant poxvirus that expresses the neopeptide, and manufacturing said recombinant poxvirus.

[0097] Creation of synthetic nucleic acid molecules Nucleic acid molecules are designed to encode selected neopeptides, preferably in one or more fusion configurations, optionally with linkers, tags, and peptide signals as described herein, which are preferably produced by chemical synthesis (e.g., assembly from overlapping synthetic oligonucleotides or direct synthesis of synthetic genes) in an automated manner.

[0098] In certain embodiments, it may be advantageous to optimize nucleic acid sequences to provide high-level expression in specific subjects. Briefly, nucleic acid molecules can be designed to improve expression through codon optimization, avoiding the selection of non-optimal codons for a given host organism, and / or eliminating stretches of AT or GC-rich sequences that are expected to negatively affect expression. The sequence TTTTTNT is also avoided because it is a poxvirus early transcription termination signal.

[0099] Expression of synthetic nucleic acid molecules In the present invention, the nucleic acid molecule encoding the neopeptide to be inserted into the genome of the recombinant poxvirus is operably linked to suitable regulatory elements. In the context of the present invention, the nucleic acid molecule encoding the neopeptide can be placed in one or more expression cassettes. Generally, an "expression cassette" comprises one or more nucleic acid molecules encoding the neopeptide under the control of suitable regulatory elements that allow expression in a subject.

[0100] As used herein, the term "regulatory element" refers to any element that enables, contributes to, or regulates the expression of a neopeptide, including replication, duplication, transcription, splicing, translation, stability, and / or transport processes. As used herein, "operably linked" means that the elements that are linked are positioned so that they function in concert for their intended purpose.

[0101] It will be appreciated by those skilled in the art that the choice of regulatory element may depend on factors such as the neopeptide itself, the poxvirus itself, the subject being treated, the desired level of expression, and the like. The promoter is particularly important. In the context of the present invention, it can be constitutively directional in many types of cells, or specific to a particular cell or cell type (e.g., tumor-specific regulatory sequences), or can be regulated in response to specific events or exogenous factors (e.g., temperature, nutrient additives, hormones, etc.) or according to the phase of the viral cycle (e.g., late or early). To optimize viral production and avoid potential toxicity of the expressed polypeptide, a promoter that is repressed during the production process in response to a specific event or exogenous factor may be used. Those skilled in the art will recognize that in addition to the promoter, the regulatory element may further comprise additional elements for proper initiation, regulation, and / or termination of transcription (e.g., polyA transcription termination sequence), transport, processing, and stability of mRNA, and translation (e.g., initiator Met, tripartite leader sequence, IRES ribosome binding site, signal peptide, etc.).

[0102] Vaccinia virus promoters are particularly suitable for expression in recombinant poxviruses. Representative examples include, but are not limited to, vaccinia 7.5K, H5R, 11K7.5 (Erbs et al., 2008, Cancer Gene Ther. 15(1): 18-28), TK, p28, p11, B2R, A35R and K1L promoters, synthetic promoters such as those described in Chakrabarti et al. (1997, Biotechniques 23: 1094-7; Hammond et al., 1997, J. Virol. Methods 66: 135-8; and Kumar and Boyle, 1990, Virology 179: 151-8), and early / late chimeric promoters.

[0103] When a recombinant poxvirus comprises several expression cassettes encoding one or more neopeptides, expression can be controlled by the same or different regulatory elements (e.g., promoters). Preferably, the recombinant poxvirus comprises 1 to 5 expression cassettes, more preferably each for the expression of a neopeptide fusion described herein. A particularly preferred embodiment relates to a recombinant MVA comprising 1 to 3 cassettes for the expression of fusions of 5 to 10 neopeptides, each under the transcriptional control of a vaccinia promoter selected from the group consisting of p11k7.5, pH5R, and pB2R.

[0104] In an alternative embodiment, if the number of expression cassettes complicates the construction of the recombinant poxvirus (e.g., more than three), one may proceed via a system comprising two or more unique recombinant poxviruses, meaning that the personalized cancer vaccine comprises two or more recombinant poxviruses (e.g., two or three) expressing the neopeptide.

[0105] Insertion of a nucleic acid molecule encoding a neopeptide into a recombinant poxvirus Nucleic acid molecules or expression cassettes encoding one or more neopeptides for use herein are inserted into the genome of a parent poxvirus to generate the recombinant poxvirus. The neopeptide-encoding nucleic acid sequence / expression cassette may be inserted at any position within the poxvirus genome, for example, within a viral gene, an intergenic region, a non-essential gene or region, or in place of a viral sequence. Insertion into the poxvirus genome at a non-essential locus is preferred. The thymidine kinase gene is particularly suitable for insertion into Copenhagen and Western Reserve vaccinia viruses, and deletions II and III are particularly suitable for insertion into MVA vectors (WO 97 / 02355; Meyer et al., 1991, J. Gen. Virol. 72: 1031-8). Preferably, when the poxvirus is MVA, the neopeptide-encoding nucleic acid molecule or expression cassette is inserted into deletion III of MVA. When the recombinant poxvirus comprises several of the cassettes described above, they may be inserted into the viral genome at the same or different positions. Insertion of all expression cassettes at the same position is preferred, in particular at the TK locus for recombinant vaccinia viruses and at deletion III for recombinant MVA.

[0106] General conditions for constructing recombinant poxviruses are well known in the art (see, e.g., WO2007 / 147528; WO2010 / 130753; WO03 / 008533; US6,998,252; US5,972,597 and US6,440,422). Generally, the nucleic acid molecule / expression cassette to be inserted is cloned into a transfer plasmid surrounded by two recombination arms corresponding to stretches of poxvirus sequences homologous (e.g., 90-100% identical) to those present in the parent genome on either side of the insertion site. The length of the recombination arms can vary within the transfer plasmid. Desirably, each of the recombination arms comprises at least 150 bp, preferably at least 200 bp, more preferably at least 300 bp, even more preferably 300-600 bp, and particularly preferably 350-500 bp (e.g., approximately 350 bp or 500 bp) or 300-400 bp of homologous poxvirus sequence. The parent poxvirus may be a wild-type poxvirus or may be modified (e.g., attenuated, tumor-specific, etc.), as described above in connection with the term "poxvirus." Insertion then occurs by homologous recombination between stretches of homologous sequence present in both the parent genome and the linearized transfer plasmid, which requires transfection of permissive cells with the linearized transfer plasmid and infection with the parent poxvirus.

[0107] In a specific embodiment of the method of the present invention, the step of generating a recombinant poxvirus involves the use of a parent poxvirus comprising a reporter gene, particularly a fluorescent reporter gene, cloned at the insertion site selected for the neopeptide-encoding nucleic acid molecule / expression cassette. Preferably, the reporter gene is placed under the transcriptional control of a promoter, e.g., a vaccinia promoter, that allows its expression in permissive cells. This embodiment facilitates the selection of recombinant poxviruses relative to the parent poxvirus. Representative examples of fluorescent reporters that can be used in the context of the present invention include, but are not limited to, GFP (green fluorescent protein), eGFP (enhanced green fluorescent protein), AmCyan1 fluorescent protein, and mCherry. For example, when relying on mCherry (a monomeric fluorescent protein derived from Discosoma mushroom with peak absorption / emission at 587 nm and 610 nm), recombinant viruses in which a neopeptide-encoding nucleic acid molecule or expression cassette is inserted in place of the mCherry-encoding sequence produce white plaques, whereas parent viruses with an mCherry expression cassette produce red plaques. Selection of recombinant poxviruses can be by direct visualization (white plaques) or facilitated by selection procedures such as FACS after labeling with APC (allophycocyanin)-tagged anti-vaccinia virus antibodies. Numerous anti-vaccinia antibodies are available from commercial sources. Typically, one recombinant is obtained for every 50–100 parents, and the entire process from insertion to recombinant poxvirus generation takes 5–6 weeks.

[0108] A preferred embodiment of the present invention relates to a process for producing a recombinant poxvirus, which allows the efficiency of homologous recombination to be increased by at least 10 times and the time required for producing a recombinant poxvirus to be reduced by at least 1.5 times. As in the above-described method, the nucleic acid or expression cassette to be inserted into the poxvirus genome is cloned in a transfer plasmid process surrounded by recombination arms, and the parent virus comprises a reporter gene (e.g., a fluorescent reporter) cloned at the insertion site under the control of a suitable promoter. Preferably, the parent virus is an MVA comprising a reporter gene (particularly, an mCherry gene) inserted into deletion III under the control of a poxvirus promoter. However, the process for producing a recombinant poxvirus comprises a further step of cleavage with an endonuclease capable of generating one or more double-strand breaks in the nucleotide sequence of the reporter (e.g., mCherry), but the endonuclease does not cleave the poxvirus genome. Suitable endonucleases are preferably selected from the group consisting of zinc finger nucleases (ZFNs), transcription activator-like effector nucleases (TALENs), clustered regularly interspaced short palindromic repeats (CRISPR) / Cas9 nucleases, and restriction enzymes with unique cleavage within fluorescent reporter genes.The use of the CRISPR / CAS9 system for virus editing has been described in the art (Yuan et al., 2015, J. Virol 89, 5176-9; Yuan et al., 2016, Viruses 8, 72, doi:10.3390), and requires the use of a plasmid encoding Cas9 without a nuclear localization signal and a suitable guide RNA.In this regard, in addition to infection with the parent virus, permissive cells can also be transfected with one or more plasmids encoding a transfer plasmid, a plasmid expressing Cas9, and a guide RNA (for example, an mCherry-targeting guide RNA).Recombinant poxviruses are then selected visually (direct isolation of white plaques corresponding to the recombinant, while colored plaques correspond to the parent, the color depending on the reporter gene) or using conventional sorting means (optionally FACS after a labeling step with an appropriate antibody as described above). This method yields one recombinant for every 1 to 10 parental genes, and the entire process from the insertion step to the generation of recombinant poxvirus takes approximately 3 to 4 weeks. In a preferred embodiment, the process of generating recombinant poxviruses takes up to 4 weeks, preferably up to 3 weeks, and the recombinant to parental ratio is between 2 (2 to 1) and 0.1 (1 to 10), preferably between 2 (2 to 1) and 0.05 (1 to 20). In a specific embodiment, the recombinant / parental ratio is 1 or greater, especially when the two recombinant arms are approximately 350 bp long.

[0109] A particularly preferred embodiment relates to a method that allows for the generation of a recombinant poxvirus (e.g., a vaccinia virus, more preferably, an MVA) comprising one to three expression cassettes inserted into the vaccinia TK locus or MVA deletion III, each encoding a fusion of five to ten neopeptides (preferably, each fusion is under the transcriptional control of a different promoter) placed under the transcriptional control of a vaccinia promoter selected from the group consisting of p11k7.5, p7.5K, pH5R, and pB2R, preferably p11k7.5, pH5R, and pB2R. The present invention also relates to a personalized cancer vaccine comprising such a recombinant poxvirus.

[0110] Production of recombinant poxviruses Once generated, the recombinant poxvirus for use herein may be produced / amplified using conventional techniques. Thus, the method of the present invention may further comprise a step of producing said recombinant poxvirus. In a preferred embodiment, said production step comprises a step of amplifying to a suitable scale in suitable producer cells, a step of recovering the produced recombinant poxvirus from the cell culture, and an optional step of purifying the recovered recombinant poxvirus.

[0111] The amplification step involves culturing producer (e.g., permissive) host cells, infecting the cultured producer host cells, and culturing the infected host cells under suitable conditions to allow the production of recombinant poxvirus (e.g., infectious viral particles).

[0112] The choice of producer cells depends on the type of recombinant poxvirus to be amplified. MVA is strictly host-restricted and is generally amplified in avian cells, either primary avian cells (e.g., chicken embryo fibroblasts (CEFs) prepared from chicken embryos obtained from fertilized eggs) or immortalized avian cell lines. Representative examples of suitable avian cell lines for MVA production include, but are not limited to, Muscovy duck cell lines immortalized with the duck TERT gene (see, e.g., WO2007 / 077256, WO2009 / 004016, WO2010 / 130756 and WO2012 / 001075); avian cell lines immortalized with a combination of viral and / or cellular genes (see, e.g., WO2005 / 042728); naturally immortalized cells (e.g., the chicken DF1 cell line disclosed in US 5,879,924); or immortalized cells derived from embryonic cells by progressive separation from growth factors and feeder layers (e.g., Olivier et al., 2010, mAbs 2(4): and Ebx chicken cell lines disclosed in WO2005 / 007840 and WO2008 / 129058, such as Eb66 described at 405-15.

[0113] For other vaccinia viruses or other poxvirus strains, in addition to avian primary cells (such as CEF) and avian cell lines, many other non-avian cell lines are available for production, including human cell lines such as HeLa (ATCC-CRM-CCL-2™ or ATCC-CCL-2.2™), MRC-5, HEK-293; hamster cell lines such as BHK-21 (ATCC CCL-10), and Vero cells. In a preferred embodiment, non-MVA vaccinia viruses are amplified in HeLa cells (see, e.g., WO2010 / 130753).

[0114] Producer cells are preferably cultured in a synthetic medium free of animal or human products. Growth factors may be present, but they are preferably recombinantly produced and not purified from animal materials. A suitable animal-free medium can be easily selected by those skilled in the art depending on the producer cells selected. Such media are commercially available. In particular, when CEFs are used as producer cells, they can be cultured in VP-SFM cell culture medium (Invitrogen). Producer cells are preferably cultured at a temperature between 30°C and 38°C (more preferably, approximately 37°C) for 1 to 8 days (preferably 1 to 5 days for CEFs and 2 to 7 days for immortalized cells) before infection. If necessary, several 1- to 8-day passages may be performed to increase the total cell number. For example, after isolation of white plaques, amplification of the recombinant neoepitope-expressing poxvirus (e.g., MVA) can be performed in 6- to 12-well cell culture plates and then transferred to flasks to obtain the desired viral load.

[0115] Infection of producer cells with recombinant poxvirus is carried out under appropriate conditions (e.g., using an appropriate multiplicity of infection (MOI)) to allow productive infection of the producer cells. A suitable MOI for use in amplifying poxvirus is generally between 0.001 and 1 (more preferably, about 0.05). The infection step is also preferably carried out using a synthetic medium that is free of animal- or human-derived products, in a medium that is free of animal- or human-derived products (which may be the same or different from the medium used to culture the producer cells). The infection step typically lasts for 1 to 6 days, more preferably 2 to 4 days, and most preferably about 72 hours.

[0116] The infected producer cells are then cultured under appropriate conditions known to those skilled in the art until progeny viral vectors (e.g., infectious viral particles) are produced. Culturing of the infected producer cells is also preferably carried out in a medium (using a synthetic medium free of animal or human products) that is free of animal or human products (which may be the same or different from the medium used for the producer cell culture and infection steps) at a temperature between 30°C and 37°C for 1 to 5 days.

[0117] Poxvirus particles can be recovered from the culture supernatant and / or producer cells. The cell culture supernatant and producer cells may be pooled or recovered separately. Recovery from producer cells (and optionally from the culture supernatant) may require a step that allows disruption of the producer cell membrane to allow vector release. Various techniques are available to those skilled in the art, including, but not limited to, freeze / thaw, hypotonic lysis, sonication, microfluidization, or high-speed homogenization. According to a preferred embodiment, the process for recovering the produced recombinant poxvirus comprises a lysis step in which the producer cell membrane is disrupted, preferably using a high-speed homogenizer. High-speed homogenizers are commercially available from Silverson Machines Inc. (East Longmeadow, USA) or Ika-Labotechnik (Staufen, Germany). According to a particularly preferred embodiment, the high-speed homogenizer is a SILVERSON L4R.

[0118] The poxvirus particles can then be further purified using purification processes well known in the art.Various purification processes can be envisioned, including, for example, clarification, enzyme treatment (e.g., endonuclease, protease, etc.), chromatography, and filtration.Suitable methods have been described in the art (e.g., WO2007 / 147528; WO2008 / 138533, WO2009 / 100521, WO2010 / 130753, WO2013 / 022764).In a preferred embodiment, the purification process comprises a tangential flow filtration (TFF) process, which can be used to concentrate and / or desalt the virus suspension and separate the virus from other biomolecules.Depending on the volume to be filtered, various TFF systems and devices are available in the art, including, but not limited to, Spectrumlabs, Pall Corp, PendoTech, and New Pellicon, among others.

[0119] In a preferred embodiment, the manufacturing process comprises the steps of: (a) preparing a personalized cancer vaccine in a dose suitable for testing and treating a patient (from whom the tumor sample is derived); (b) preparing a personalized cancer vaccine in a dose suitable for testing and treating a patient (from whom the tumor sample is derived); and (c) preparing a personalized cancer vaccine in a dose suitable for testing and treating a patient (from whom the tumor sample is derived). 9 pfu, preferably at least 5 x 10 9 pfu, preferably approximately 10 10 Production reaches more than pfu.

[0120] Armed recombinant poxviruses Certain embodiments of the present invention also encompass recombinant poxviruses comprising, in addition to the nucleic acid molecule encoding the neopeptide, an additional therapeutic gene inserted into the viral genome, for example, to enhance antitumor responses. A vast number of therapeutic genes are envisioned, including in particular those encoding polypeptides capable of enhancing the antitumor efficacy of the virus or strengthening the host's immunity. They may be of human or non-human origin (e.g., bacterial, yeast, or viral origin) and may be native genes, fragments, or functional analogs thereof (obtained from the latter by mutation, deletion, substitution, and / or addition of one or more nucleotides). Such analogs preferably have a nucleotide sequence with a degree of identity of at least 70%, advantageously at least 80%, preferably at least 90%, and most preferably at least 95% with the nucleic acid sequence of the native gene. Preferred therapeutic genes are selected from the group consisting of suicide genes and immunostimulatory genes.

[0121] Immunostimulating therapeutic genes Immunostimulatory therapeutic genes encode polypeptides capable of stimulating the immune system or effector cells in a specific or non-specific manner. Examples of suitable immunostimulatory proteins in the context of the present invention include, but are not limited to, cytokines, particularly preferably interleukins (e.g., IL-2, IL-6, IL-12, IL-15, IL-24), chemokines (e.g., CXCL10, CXCL9, CXCL11), interferons (e.g., IFNγ, IFNα), tumor necrosis factors (TNF), colony-stimulating factors (e.g., GM-CSF, G-CSF, M-CSF, etc.), APC (antigen presenting cell) exposed proteins (e.g., B7.1, B7.2, etc.), These include growth factors (such as transforming growth factor TGF, fibroblast growth factor FGF, vascular endothelial growth factor VEGF, etc.), class I or II major histocompatibility complex (MHC) components, apoptosis inducers or inhibitors (e.g. Bax, Bcl2, BcIX...), cytostatics (p21, p16, Rb...), immunotoxins, blockers of immune escape mechanisms such as immune checkpoint inhibitors (said inhibitors being blocking peptides, antibodies or related biological species such as single domain antibodies or antibody fragments), cGAS, etc.

[0122] Suicide gene The term "suicide gene" refers to a gene encoding a protein capable of converting a drug precursor into a cytotoxic drug. Representative examples of suitable suicide genes include, but are not limited to, thymidine kinase (TK), thymidylate kinase, cytosine deaminase (CDase), and uracil phosphoribosyltransferase (UPRTase).

[0123] CDase and UPRTase are present in prokaryotes and lower eukaryotes (but not in mammals). CDase is involved in the pyrimidine metabolic pathway, in which exogenous cytosine is converted to uracil by hydrolytic deamination. CDase also deaminates an analog of cytosine, i.e., 5-fluorocytosine (5-FC) prodrug, thereby forming 5-fluorouracil (5-FU), a highly cytotoxic compound when converted to 5-fluoro-UMP (5-FUMP) by UPRTase. Gene sequences and the encoded enzymes are publicly available, for example, in specialized data banks (SWISSPROT EMBL, GenBank, Medline, etc.). Particularly preferred are yeast UPRTases and CDases, especially those encoded by Saccharomyces cerevisiae (FUR1 gene; Kern et al., 1990, Gene 88: 149-57). Of particular interest is the FCU1 suicide gene, which encodes a fusion of Saccharomyces cerevisiae CDase with a truncated UPRTase in which the first 35 residues have been deleted (FCY1::FUR1[Delta]105 fusion), the amino acid sequence of which is set forth in sequence identifier SEQ ID NO: 1 of WO2009 / 065546.

[0124] Therapeutic genes for use herein can be independently generated by a number of methods known to those skilled in the art (e.g., cloning, PCR amplification, DNA shuffling, chemical synthesis) and from available resources (e.g., biological material described in the art, such as cDNA, genomic libraries, viral genomes or any prior art vector known to contain same), using sequence data available to those skilled in the art and the sequence information provided herein, and then suitably inserted into a recombinant poxvirus by conventional molecular biology techniques, either in the same position as the nucleic acid molecule encoding the neoepitope (e.g., deletion III of MVA) or in a different position.

[0125] Pharmaceutical Composition In one embodiment, the personalized cancer vaccine of the present invention is in the form of a composition comprising a therapeutically effective amount of a recombinant poxvirus described herein (or obtained according to the methods described herein) and a pharmaceutically acceptable vehicle.

[0126] The term "pharmaceutically acceptable vehicle" is intended to include any and all carriers, solvents, diluents, excipients, adjuvants, dispersion media, coatings, antibacterial and antifungal agents, absorption agents, and the like, which are compatible with administration in mammalian, particularly human, subjects.

[0127] A "therapeutically effective amount" corresponds to the amount of recombinant poxvirus necessary to bring about an observable improvement in a clinical condition, including one or more of those described below, in a subject treated according to the present invention.

[0128] Such a therapeutically effective amount may vary depending on a variety of factors, including, but not limited to, the characteristics of the poxvirus (including the type of virus, bioavailability, and dosage), the severity and course of the disease (e.g., cancer grade), the subject itself (including age, sex, medical history, general health, etc.), the properties of any pharmaceutically acceptable carriers or excipients in the virus formulation, and the treatment modality (administration route, administration frequency, type of concomitant medication, etc.). The appropriate dose of poxvirus can be routinely determined and adapted by a practitioner taking into account the relevant circumstances, for example, by monitoring the subject's response to administration of the virus and adjusting the dose accordingly (for further guidance, see Remington, 2002; The Science and Practice of Pharmacy; Gennaro ed., Pharmaceutical Press, London, UK; e.g., 22nd ed. et seq.).

[0129] For illustrative purposes, suitable therapeutically effective amounts for individual doses range from approximately 10 to 150 mg / kg, depending on the poxvirus and quantification technique used. 5 ~approximately 10 13It can vary in vp (viral particles), iu (infectious units) or pfu (plaque forming units). As a general guidance, approximately 10 6 pfu ~ approx. 10 11 Individual doses of 10 pfu are particularly suitable in the context of the present invention, more preferably approximately 10 7 pfu ~ approximately 5 × 10 9 pfu; even more preferably, approximately 5 x 10 7 pfu ~ approx. 10 9 pfu (e.g., 5 × 10 7 ~6×10 8 , 6×10 7 ~5×10 8 , 7×10 7 ~4×10 8 , 8×10 7 ~3×10 8 , 9×10 7 ~2×10 8 Doses of 10 pfu are convenient for human use, and preferably each dose is approximately 10 8 The individual doses may be reduced 2- to 20-fold for local administration, such as intratumoral injection. The amount of virus present in a sample can be determined by conventional titration techniques, for example, by counting the number of plaques after infection of permissive cells (e.g., BHK-21 or CEF), by immunostaining (e.g., using antiviral antibodies; Caroll et al., 1997, Virology 238: 198-211), by measuring A260 absorbance (vp titer), by quantitative immunofluorescence (iu titer), or by qPCR using specific viral primers and probes.

[0130] Various formulations are contemplated within the context of the present invention, either liquid or lyophilized, to ensure viral stability under conditions of manufacture and long-term storage (i.e., at least 6 months) at frozen (e.g., -70°C, -20°C), refrigerated (e.g., 4°C), or ambient (e.g., 20-25°C) temperatures. The recombinant poxvirus is advantageously placed in a diluent suitable for human or veterinary use. Representative examples of suitable diluents include sterile water, saline (e.g., sodium chloride), Ringer's solution, glucose, trehalose, or sucrose solution, Hank's solution, and other physiologically balanced salt solutions.

[0131] Desirably, the personalized cancer vaccine composition is buffered for human use. Buffers such as TRIS (tris(hydroxymethyl)methylamine), TRIS-HCl (tris(hydroxymethyl)methylamine-HCl), HEPES (4-2-hydroxyethyl-1-piperazineethanesulfonic acid), phosphate buffers (e.g., PBS; a mixture of NaHPO and KHPO; a mixture of NaHPO and NaHPO), TEA (triethanolamine), EPPS (N-(2-hydroxyethyl)-piperazine-N'-3-propanesulfonic acid), TRICINE (N-[tris(hydroxymethyl)-methyl]-glycine), and bicarbonate buffers are particularly suitable for maintaining a physiological or slightly basic pH (e.g., approximately pH 7 to approximately pH 9). Buffers (e.g., TRIS-HCl) are preferably present at a concentration of 10 to 50 mM.

[0132] To ensure an appropriate osmolality, it may also be beneficial to include a monovalent salt, which may in particular be chosen from NaCl and KCl, preferably NaCl, in particular at a concentration of 10 to 500 mM.

[0133] Optionally, the cancer vaccine composition may contain a cryoprotectant to facilitate storage at low temperatures. Suitable cryoprotectants include, but are not limited to, sucrose, trehalose, maltose, lactose, mannitol, sorbitol, and glycerol, at concentrations ranging from 0.5 to 20% (weight (g) / volume (L), referred to as w / v), preferably 5 to 15% (w / v), and preferably about 10%. The presence of high-molecular-weight polymers such as dextran or polyvinylpyrrolidone (PVP) is particularly suitable for freeze-dried formulations to protect recombinant poxviruses during vacuum drying and freeze-drying processes (see, e.g., WO03 / 053463; WO2006 / 0850082; WO2007 / 056847; WO2008 / 114021; WO2014 / 053571).

[0134] For illustrative purposes, buffer formulations containing NaCl and / or sugars are particularly adapted for the storage of poxviruses (e.g., Tris 10 mM pH 8 containing sucrose 5% (W / V), sodium glutamate 10 mM, and NaCl 50 mM; or phosphate buffered saline containing glycerol (10%) and NaCl).

[0135] Therapeutic Uses and Methods of Treatment The personalized cancer vaccines and methods of the present invention are particularly suitable for treating cancer, particularly metastatic cancer and those at high risk of recurrence. Thus, the present invention also provides personalized cancer vaccines (e.g., poxvirus compositions) for use in treating cancer or preventing its recurrence in a subject, as well as the use of personalized cancer vaccines for such therapeutic purposes. The present invention also relates to personalized cancer vaccines or compositions thereof described herein for the manufacture of medicaments for treating cancer or preventing its recurrence.

[0136] In another aspect, the present invention also relates to a method of treatment comprising administering a personalized cancer vaccine composition in a subject in need thereof in an amount sufficient to treat or prevent the recurrence of cancer in said subject.

[0137] Representative examples of cancers that can be treated in the context of the present invention include bone cancer, liver cancer, pancreatic cancer, stomach cancer, colon cancer, esophageal cancer, oropharynx cancer, lung cancer, head or neck cancer, skin cancer, melanoma, uterine cancer, cervical cancer, ovarian cancer, breast cancer, rectal cancer, anal cancer, prostate cancer, lymphoma, endocrine system cancer, thyroid cancer, soft tissue sarcoma, chronic or acute leukemia, bladder cancer, kidney cancer, neoplasms of the central nervous system (CNS), glioma, etc. The present invention is particularly suitable for the treatment of solid tumors. It is also particularly useful for the treatment of advanced cancers, including metastatic solid cancers or cancers associated with a high risk of recurrence. Preferred cancers to be treated according to the methods described herein include brain cancers, such as astrocytoma, embryonal tumor, germ cell tumor, central nervous system atypical teratoma / rhabdoid tumor, craniopharyngioma, ependymoma, glioma and glioblastoma, and head and neck cancer. Other types of cancer to be treated in accordance with the modalities described herein are lung cancer, particularly NSCLC, with adenocarcinoma, squamous cell carcinoma, and large cell carcinoma being particularly preferred. Another preferred cancer to be treated with the personalized cancer vaccines of the present invention is ovarian cancer.

[0138] The beneficial effects provided by the use of the personalized cancer vaccines of the present invention or methods therewith can be evidenced by an observable improvement in clinical status above baseline or above the status expected in the absence of treatment according to the methods described herein. Improvement in clinical status can be readily assessed by any relevant clinical measure commonly used by physicians or other skilled medical professionals. In the context of the present invention, therapeutic benefit can be transient (for one or several months after cessation of treatment) or sustained (for several months or years). Because the natural history of clinical status can vary considerably from subject to subject, therapeutic benefit need not be observed in every subject treated, but rather in a significant number of subjects (e.g., statistical significance between two groups can be determined by any statistical test known in the art, such as Tukey's parametric test, Kruskal-Wallis test, Mann-Whitney U test, Student's t-test, Wilcoxon test, etc.).

[0139] In certain embodiments, such therapeutic benefit may correlate with one or more of the following: inhibition or slowing of tumor growth, prevention or delay of proliferation and metastasis, tumor invasion (spread of tumor cells into adjacent tissues), reduction in tumor lesion count; reduction in tumor size, reduction in number or extent of metastases, prolonged overall survival (OS), increased progression-free survival (PFS), increased length of remission, stabilization of the disease state (i.e., does not worsen), prevention of disease recurrence, providing a good response to another treatment, particularly another immunotherapy, improved quality of life, and / or elicitation of an anti-tumor response (e.g., nonspecific (innate) and / or specific, such as a cytotoxic T cell response) in a subject treated in accordance with the present invention.

[0140] Appropriate measurements, such as blood tests, analyses of biological fluids and biopsies, and medical imaging techniques, are routinely performed in available medical laboratories and hospitals to assess clinical benefit, and many kits are commercially available. They can be performed before administration (baseline), during treatment, and at various time points after treatment has ceased.

[0141] Administration of personalized cancer vaccines or compositions thereof Any conventional route of administration is applicable, preferably parenteral. Parenteral routes involve administration by injection or infusion and include systemic and local routes. Particularly suitable routes of administration include, but are not limited to, intravenous (into veins), intravascular (into blood vessels), intraarterial (into arteries), intradermal (into the dermis), subcutaneous (under the skin), intramuscular (into muscles), intraperitoneal (into the peritoneum), intracerebral (into the brain), and intratumoral (into or near a tumor) routes, as well as scarification. Injection is generally performed intravenously or intratumoral (into a large tumor). Mucosal administration is also contemplated by the present invention, examples of which include, but are not limited to, oral / dietary, intranasal, intratracheal, nasopharyngeal, intrapulmonary, intravaginal, or intrarectal routes. Topical administration involves direct application to the skin or tissue surface (e.g., eye drops; ear drops, etc.). Inhalation may also be envisioned, particularly when the tumor to be treated is in the respiratory tract and lungs. The personalized cancer vaccine or composition thereof is preferably administered to the patient by intravenous, subcutaneous, intramuscular or intratumoral injection.

[0142] Administration may be performed using a standard needle and syringe or any device available in the art that can facilitate or improve delivery, such as a catheter, an electric syringe, a Quadrafuse injection needle, a needle-free injection device (e.g., a Biojector™ device), an infusion pump, a spray, etc. Electroporation may also be performed to facilitate intramuscular administration. Local administration may also be performed using transdermal means (e.g., a patch, a microneedle, etc.).

[0143] The personalized cancer vaccine can be administered in a single dose, or more desirably, in multiple doses over an extended period of time. In the context of the present invention, it is possible to proceed through sequential cycles of administration, repeated after a rest period. The interval between each viral administration can be from a few hours to 6 months (e.g., 24 hours, 48 ​​hours, 72 hours, 1 week, 2 weeks, 3 weeks, 1 month, 2 months, etc.). The intervals can be regular or irregular. The dose can vary with each administration within the above range. For illustrative purposes, a preferred treatment scheme is 10 doses at intervals of approximately 1 or 2 weeks until a clinical benefit is observed, and then every 1 to 6 months. 7 ~5×10 9 It involves 1 to 10 (e.g., 2, 3, 4, 5, 6, 7, 8, 9, 10) administrations of pfu recombinant MVA.

[0144] Combination therapy In a further embodiment of the methods and therapeutic uses described herein, personalized cancer vaccine can be administered in combination with one or more additional anti-cancer therapies that are useful in treating the aforementioned cancers.In particular, the additional anti-cancer therapies are selected from the group consisting of surgery, radiotherapy, chemotherapy, cryotherapy, hormone therapy, toxin therapy, immunotherapy and cytokine therapy.Such additional anti-cancer therapies can be administered to subjects according to standard techniques before, after, essentially simultaneously with, or intermittently with the personalized cancer vaccine of the present invention.

[0145] In certain embodiments, the method or use according to the present invention can be performed in conjunction with surgery. For example, the personalized cancer vaccine can be administered after partial or complete surgical resection of the tumor (e.g., by local application within the resection area).

[0146] In other embodiments, the methods of the present invention can be used in conjunction with radiation therapy. Those skilled in the art can readily formulate appropriate radiation therapy protocols and parameters (see, e.g., Perez and Brady, 1992, Principles and Practice of Radiation Oncology, 2nd ed., J.B. Lippincott Co., with appropriate adaptations and modifications readily apparent to those skilled in the art). Types of radiation that can be used in cancer treatment are well known in the art and include electron beams, high-energy photons from linear accelerators or radioactive sources such as cobalt or cesium, protons, and neutrons. Dose ranges for radioisotopes vary widely and depend on the half-life of the isotope, the strength and type of radiation emitted, and uptake by neoplastic cells. Conventional X-ray doses over a prolonged period (3-6 weeks), or high single doses, are contemplated by the present invention.

[0147] In certain embodiments of the present invention, personalized cancer vaccines can be used in combination with currently available chemotherapy for the treatment of cancer.Representative examples of suitable chemotherapeutic agents include, but are not limited to, alkylating agents, topoisomerase I inhibitors, topoisomerase II inhibitors, PARP inhibitors, platinum derivatives, tyrosine kinase receptor inhibitors, cyclophosphamide, antimetabolites, DNA damaging agents and antimitotic agents.

[0148] In further embodiments, personalized cancer vaccines can be used in combination with immunotherapy, such as anti-neoplastic antibodies and siRNA and antisense polynucleotides.Representative examples include, inter alia, monoclonal antibodies that block specific immune checkpoints, such as anti-PD-1, anti-PD-L1, anti-CTLA-4, anti-LAG3 (for example, ipilimumab, tremelimumab, pembrolizumab, nivolumab, pidilizumab, AMP-224MEDI4736, MPDL3280A, BMS-936559, etc.), monoclonal antibodies that block epidermal growth factor receptors (particularly, cetuximab, panitumumab, zalutumumab, nimotuzumab, matuzumab, trastuzumab (Herceptin™)), and monoclonal antibodies that block vascular endothelial growth factor (particularly, bevacizumab and ranibizumab).

[0149] In a further embodiment, the personalized cancer vaccine may be used in combination with an adjuvant. Representative examples of suitable adjuvants include, but are not limited to, TLR3 ligands (Claudepierre et al., 2014, J. Virol. 88(10): 5242-55), TLR9 ligands (e.g., ODN1826 (Fend et al., 2014, Cancer Immunol. Res. 2, 1163-74) and Litenimod (Li28) (Carpentier et al., 2003, Frontiers in Bioscience 8, e115-127; Carpentier et al., 2006, Neuro-Oncology 8(1): 60-6; ​​EP1162982; US7,700,569 and US7,108,844) and PDE5 inhibitors such as sildenafil (US5,250,534, US6,469,012 and EP463756).

[0150] In a preferred embodiment, the methods of the invention increase survival time in treated subjects, e.g., by at least 3 months, compared to untreated subjects. Alternatively, the methods of the invention generate a T cell response (CD4+ and / or CD8+ T cell response) against the tumor.

[0151] The administration of the personalized cancer vaccine and one or more additional anti-cancer therapies may be spaced apart by minutes to weeks. For example, a subject may receive the recombinant poxvirus and the additional anti-cancer therapies sequentially or intermittently, although simultaneous administration of both therapies within the same period is also contemplated. The course of treatment can be routinely determined by practitioners, but various protocols are encompassed by the present invention. For example, 1 to 10 administrations of a personalized cancer vaccine can be administered after surgery and chemotherapy / radiotherapy. Furthermore, it is contemplated that after the course of treatment, there will be a period during which no anti-cancer therapy is administered before the treatment cycle is repeated.

[0152] All patent disclosures, publications, and database entries cited above are specifically incorporated herein in their entirety by reference. Other features, objects, and advantages of the present invention will be apparent from the specification and drawings, and from the claims. The following examples are included to demonstrate preferred embodiments of the present invention. However, in light of this disclosure, those skilled in the art will recognize that changes can be made to the specific embodiments disclosed without departing from the spirit and scope of the present invention. [Example]

[0153] Materials and Methods The constructions described below are carried out according to general genetic engineering and molecular cloning techniques detailed in Maniatis et al. (1989, Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, or later editions), or, if a commercially available kit is used, according to the manufacturer's recommendations. PCR amplification techniques are known to those skilled in the art (e.g., PCR protocols - A guide to methods and applications, 1990, Innis, Gelfand, Sninsky, and White, Academic Press).

[0154] Production of recombinant MVA Synthetic genes encoding different fusions of vaccinia promoters and neoepitopes were synthesized by Geneart (Regensburg, Germany). Sequences were optimized for human codon usage, and a Kozak sequence (ACC) was added before the ATG start codon. In addition, some motifs were excluded: TTTTTNT, GGGGG, and CCCCC, which are detrimental to expression in poxvirus vectors.

[0155] Generation of multi-epitope constructs: MVATG19022 and MVATG19023 MVATG19022 contains an expression cassette encoding five peptide fusions derived from five different antigens: FCU1, b-gal, MUC1, HPV-16 E7, and HPV-16 E1. All peptides are 27-mers comprising C57bl / 6 T cell epitopes. More specifically, the pentatope fusions comprise, from 5' to 3', the FCU1 peptide (SEQ ID NO: 1), the b-Gal peptide (SEQ ID NO: 2), the HPV-16 E1 peptide (SEQ ID NO: 3), the HPV-16 E7 peptide (SEQ ID NO: 4), and the MUC1 peptide (SEQ ID NO: 5). For optimal processing of the fusion polypeptide and epitope presentation, 10-mer GS linkers are present at the N-terminus of the fusion, between the peptides, and after the last peptide. A tag sequence, Flag tag (DYKDDDDK; SEQ ID NO: 6), was added to the C-terminus of the fusion to facilitate detection. The pentatope fusion (SEQ ID NO: 7) was placed under the control of the p11K7.5 promoter (SEQ ID NO: 8).

[0156] MVATG19023 contains an expression cassette encoding the same pentatope fusion with the N-terminal addition of a signal peptide (SEQ ID NO: 9) derived from the rabies glycoprotein. The modified pentatope fusion (SEQ ID NO: 10) was placed under the control of the p11K7.5 promoter (SEQ ID NO: 8).

[0157] DNA fragments corresponding to each of the pentatope fusions and surrounded by approximately 30 bp of sequence homology to vaccinia transfer plasmids were generated synthetically. The GeneArt® Strings™ DNA fragments were inserted into the vaccinia transfer plasmid pTG18626 digested with NotI and BglIII by In-Fusion cloning (In-Fusion HD Cloning Kit, Clontech), resulting in pTG19022 and pTG19023, respectively.

[0158] The MVA transfer plasmid, pTG18626, is designed to allow insertion of nucleotide sequences introduced by homologous recombination into deletion III of the MVA genome. It is derived from the plasmid pUC18 into which the flanking sequences (BRG3 and BRD3) surrounding the MVA deletion III were cloned (Sutter and Moss, 1992, Proc. Natl. Acad. Sci. USA 89:10847).

[0159] Homologous recombination was performed using a parental MVA (MVA-mCherry) containing the gene encoding the mCherry fluorescent protein within its deletion III. The advantage of MVA-mCherry is that it distinguishes cells infected with recombinant viruses that have successfully integrated the expression cassette from cells infected with the original starting MVA-mCherry virus (parental virus). Indeed, if the expression cassette successfully recombines within deletion III, the mCherry gene is removed and viral plaques appear white.

[0160] Although recombination is a relatively frequent event, only 1–5% of recombinant plaques in vaccinia virus contain inserted DNA. Therefore, to increase the efficiency of homologous recombination, an additional step of endonuclease cleavage was added. For example, the CRISPR-CAS9 system may be used to specifically generate double-strand breaks in the mCherry gene in MVA, thereby increasing the efficiency of recombinant MVA selection (e.g., typically 10–50% of viral plaques contain the expression cassette). Considering that MVA replicates in the cytoplasm, it is recommended to provide a Cas9 enzyme (pCas9 plasmid) without an NLS (nuclear localization signal) as described by Yuan et al. (2015, J. Virol. doi:10.1128) along with a pgRNA containing a gRNA specific for the mCherry sequence.

[0161] MVATG19022 was generated by homologous recombination in primary chicken embryo fibroblasts (CEFs). CEFs were isolated from 11-12-day-old embryonated SPF eggs (Charles River) and cultured in culture medium containing 40 mg / L gentamicin at a final concentration, supplemented with 4 mM glutamine. CEFs were cultured at 37°C and 5% CO2 for 24 or 48 hours before MVA infection. 10 8 CEF cells were infected with parental mCherry-encoding MVA at an MOI of 0.02 for 2 h in separate culture medium containing gentamicin at a final concentration of 40 mg / L supplemented with 2 mM L-glutamine. Infected cells were trypsinized and 10 7 Cells were transfected by nucleofection with 2 μg of pTG19022 (according to Amaxa Nucelofector technology) and, where appropriate, pCAS9 and pgRNA. Transfected cells were transferred to 6-well plates and incubated at 37°C for 2 days. Infected and transfected cells were harvested and used for plaque isolation. Serial dilutions were used to infect CEF monolayers for 30 minutes at room temperature. 2 × 10 cells seeded in 60 mm culture dishes were used. 7 CEF cells were infected with 500 μL of each dilution. After 30 minutes, 5 mL of culture medium containing 1% agarose was layered on the monolayer. The culture dish was subsequently incubated at 37°C in a 5% CO2 atmosphere for 96 hours. A new layer (5 mL) of culture medium containing 1% agarose and 0.008% neutral red was added to facilitate plaque detection by microscopy. Selected white plaques were picked, placed in 500 μL of PBS, and frozen at -20°C. Virus amplification was performed by infecting CEF monolayer cells in 6-well cell culture plates for 30 minutes at room temperature using the frozen plaques. After 30 minutes, 2 mL of culture medium was added, and the infected cells were incubated at 37°C in a 5% CO2 atmosphere. After 72 hours, the entire contents of the dish were harvested (amplification 1). Optionally, a second step of amplification can be performed in an F175 flask. In this case, 100 μL of amplified plaques were diluted in 10 mL of PBS and used to inoculate two culture flasks (approximately 7 × 10 7Fresh CEFs were infected with the 200-kDa ribosomal RNA (F175 containing 1000 kDa) for 30 minutes at room temperature. 20 ml of culture medium was added, and the infected cells were incubated at 37°C in a 5% CO2 atmosphere for 72 hours. Amplification 2 was then harvested. The presence of the expression cassette and the absence of contamination with the parental MVA were confirmed by PCR. This entire process could be performed in just 3 weeks.

[0162] MVATG19022 may be produced on a larger scale for preclinical experiments using several F500 flasks. CEFs were first cultured in appropriate medium containing 40 mg / L gentamicin at a final concentration, supplemented with 4 mM glutamine, and incubated at 37°C and 5% CO2 for 48 hours prior to infection. The cells were then infected with the above-described bulk amplificate 2. Viral amplification was carried out at 37°C or below (e.g., 34-35°C) and 5% CO2 for 72-96 hours. The infected cells and medium were then pelleted and frozen. The crude harvest was disrupted using a high-speed homogenizer (SILVERSON L4R) and sent for purification (e.g., as described in WO2007 / 147528). Briefly, the lysed virus preparation can be clarified by filtration and purified by a tangential flow filtration (TFF) step. Purified virus was resuspended in a suitable virus formulation buffer (eg, 5% (w / v) sucrose, 50 mM NaCl, 10 mM Tris / HCl, 10 mM sodium glutamate, pH 8).

[0163] MVATG19023 virus was generated in CEF by homologous recombination and produced as described above.

[0164] Generation of CT26 neopeptide constructs: MVATG19030 and MVATG19038MVATG19030 contains two expression cassettes encoding fusions of five CT26 neopeptides (pentatopes), the sequences of which are described in Kreiter et al. (2015, Nature, 520: 692-6). To facilitate detection of the pentatopes, a tag sequence was added to the C-terminus of each pentatope: a Flag tag (DYKDDDDK; SEQ ID NO: 6) for pentatope 1 and a c-myc tag (EQKLISEEDL; SEQ ID NO: 11) for pentatope 2. Tagged pentatope 1 (SEQ ID NO: 12) was under the control of the p11K7.5 promoter (SEQ ID NO: 8), while tagged pentatope 2 (SEQ ID NO: 13) was under the control of the pH5R promoter (SEQ ID NO: 14).

[0165] MVATG19038 contains two expression cassettes encoding the same pentatope fusions, with the signal peptide of the rabies glycoprotein (SEQ ID NO:9) appended to the N-terminus of pentatope 1 and the signal peptide of the measles F glycoprotein (SEQ ID NO:15) appended to the N-terminus of pentatope 2. Modified pentatope 1 (SEQ ID NO:16) was placed under the control of the p11K7.5 promoter (SEQ ID NO:8), while modified pentatope 2 (SEQ ID NO:17) was placed under the control of the pH5R promoter (SEQ ID NO:14).

[0166] A DNA fragment corresponding to the two expression cassettes separated by suitable restriction sites and surrounded by approximately 30 bp of sequences homologous to vaccinia transfer plasmids was generated by synthetic methods. After restriction with SnaB1, the fragment was inserted by In-Fusion cloning (In-Fusion HD Cloning Kit, Clontech) into the vaccinia transfer plasmid pTG18626 digested with NotI and BglII, yielding pTG19030 and pTG19038, respectively.

[0167] Generation of MVATG19030 and MVATG19038 viruses was performed in CEFs by homologous recombination as described above.

[0168] Generation of a human neopeptide construct: MVATG19111 MVATG19111 contains three expression cassettes simultaneously encoding 18 human neopeptides distributed across three fusions: the first fusion containing seven neopeptides (heptatope), the second fusion containing six neopeptides (hexatope), and the third fusion containing five neopeptides (pentatope). Most of the 18 encoded neopeptides consist of 29 amino acids with mutations at central positions, with four exceptions: a 19-amino acid neopeptide (NP1) with a mutation located at its N-terminus; a 21-amino acid neopeptide (NP18) with a mutation located at its C-terminus; a 26-amino acid neopeptide (NP10) with a mutation at the central position as defined herein but with a three-amino acid deletion in the C-terminal flanking sequence; and a 76-amino acid neopeptide (NP15) resulting from a frameshift mutation. In each fusion, the neopeptides were separated by a 5-amino acid linker for optimal processing and epitope presentation of the fusion polypeptide. Four different linkers were used: GSGSG, SGSGS, GSTSG, and SGTGS. The linker sequence codons were degenerated to avoid repetition and limit recombination events during poxvirus production.

[0169] To facilitate detection of the peptide fusions, a tag sequence was added to the C-terminus of each fusion: an HA tag (YPYDVPDYA; SEQ ID NO: 18) for the heptatope, an HSV tag (QPELAPEDPED; SEQ ID NO: 19) for the hexatope, and a VSV tag (YTDIEMNRLGK; SEQ ID NO: 20) for the pentatope. To enhance ER processing, a signal peptide was added to the N-terminus of each neopeptide fusion. A signal peptide from the rabies glycoprotein (SEQ ID NO: 9) was added to the N-terminus of the heptatope and pentatope, and a signal peptide from the measles F glycoprotein (SEQ ID NO: 15) was added to the N-terminus of the hexatope.

[0170] To control the expression of the neopeptide fusions, different promoters were used: the pH5R promoter (SEQ ID NO: 14) driving expression of the heptatope, the pB2R promoter (SEQ ID NO: 21) for expression of the hexatope, and the p11K7.5 promoter (SEQ ID NO: 8) for the pentatope, respectively.

[0171] A DNA fragment corresponding to the three expression cassettes separated by suitable restriction sites and surrounded by approximately 350 bp of sequences homologous to the vaccinia transfer plasmid was generated by synthetic methods, resulting in pTG19111.

[0172] Generation of MVATG19111 virus was performed in CEF by homologous recombination as described above.

[0173] Construction of the human neopeptide construct: pTG19247 A plasmid construct, pTG19247, containing three expression cassettes each encoding 10 neopeptides was generated. A pool of 30 neopeptides was selected based on their predicted immunogenicity from a publicly available patient dataset (study PRJEB3132, exome sequencing of human lung adenocarcinoma samples and their normal counterparts). For illustrative purposes, the amino acid sequences are given in SEQ ID NOS: 60, 61, and 62. Each cassette comprises a signal sequence (obtained from rabies and measles F glycoproteins) and is placed under the control of the pH5R, pB2R, and p11k7.5 promoters, respectively. To facilitate fusion construction, no linkers were included between the selected neopeptides. The corresponding virus was generated in CEFs by homologous recombination according to the method disclosed above. However, a low ratio of white plaques (less than 5%) was obtained, and PCR analysis of the white plaques revealed no generation of recombinant MVA with the neopeptide fusions (the white plaques are likely the result of mutations in the mCherry sequence).

[0174] Neopeptide expression and Western blot analysis 4×10 6CEF cells were infected in 6-well plates with various MVAs encoding neopeptide fusions at an MOI of 0.2. MVATGN33.1 empty vector was used as a negative control. After 24 hours, the medium was discarded, and the cells were lysed with 300 μL / dish of Tris-glycine-SDS 2x buffer (reference: LC2676; Novex) containing β-mercaptoethanol (5% V:V). The lysate was then sonicated and heated at 95°C for 5 minutes. Twenty microliters of cell lysate was subjected to electrophoresis in a precast 4-15% Criterion gel using the Criterion Precast Gel System (Biorad). After electrophoresis, proteins were transferred to a PVDF membrane (Trans-blot® Turbo™ Transfer System (#170-4155, Biorad)). Immunodetection was performed using an HRP-anti-TAG antibody. A 1 / 2000 dilution of each of the following antibodies was applied to the membrane: monoclonal anti-FLAG M2, HRP antibody (Sigma, A8592), HA tag monoclonal antibody, HRP (Thermofisher, 26183-HRP), VSV-G tag polyclonal antibody, HRP (Thermofisher, PA1-26564), goat polyclonal anti-HSV tag antibody, HRP (Abcam, 19392), while a 1 / 5000 dilution of monoclonal anti-cMyc, HRP antibody (Novex 46-0709) was used. Immune complexes were revealed using Amersham ECL prime Western blot reagents (GE Healthcare, RPN2236) and analyzed using a Molecular Imager ChemiDOC™ XRS (BIO-RAD).

[0175] ELISpot assay Plate preparation Plates (Millipore, MSIPS4W10) were pretreated with 35% ethanol (15 μl / well) for 1 minute and washed five times with sterile water (200 μl / well). Wells were coated with 100 μl of 15 μg / ml anti-mouse IFNγ antibody (Mabtech, AN18, 3321-3-1000; diluted in PBS) and incubated overnight or over the weekend at +4°C. On the day of the experiment, plates were washed five times with sterile PBS (200 μL / well) and saturated with 200 μL / well of complete medium (CM) at 37°C for at least 1 hour.

[0176] Sample preparation In vitro Elispot was performed using fresh splenic lymphocytes. Spleens from five animals in each group were collected and pooled for each experiment. Spleens were collected in 4 mL of complete medium (CM: RPMI 10% FCS or X-Vivo if the mouse had a tumor) and then crushed with a syringe plunger through a 70 μm cell strainer in a 6-well plate. The resulting splenocyte suspension was diluted two-fold with CM, placed in 4 mL of Lympholyte®-M Cell Separation Medium (Cedarlane, Reference: CL5035), and centrifuged at 1500 g for 20 minutes at room temperature. The lymphocyte-containing interface was collected and washed twice with 10 mL of RPMI 1640 medium (centrifuged at 400 g for 5 minutes at room temperature). The lymphocytes were resuspended in 2 mL of RBC lysis buffer (BD PharmLyse, BD BioScience, Reference 555899) and incubated at room temperature for 5–15 minutes to lyse red blood cells (RBCs). After one washing step with CM (centrifugation at 400 g for 5 minutes at room temperature), the lymphocytes were resuspended in 1 mL of CM and counted using a Beckman Coulter Z2 Coulter Particle Number and Size Analyzer. The cell concentration was adjusted to 1 × 10 using CM. 7 The concentration was adjusted to 1 / mL.

[0177] Assay First, the saturation medium was removed by emptying the plate, and 50 μL of CM was added to all wells. Each peptide or pool of peptides at 4 μg / mL in 50 μL of CM (i.e., for pools, the concentration of each peptide was 0.4 μg / mL), or 50 μL of CM (negative control) was added to the relevant wells (according to the pipetting scheme previously defined; each condition was tested four times). As a positive control, 50 μL of 20 μg / mL concanavalin A (ConA) was added. Second, 100 μL of each lymphocyte suspension (1 × 10 6 ) into the wells (except for T8V wells, where 3 × 10 5 Only one cell was added), and the plate was incubated at 37°C in 5% CO2 for 18-20 hours.

[0178] The cells were removed by emptying the plate, washed five times with PBS (200 μL / well), and 100 μL of biotinylated anti-mouse IFNγ monoclonal antibody (Mabtech, R4-6A2, 3321-6-1000; final concentration of 1 μg / mL in PBS 0.5% FCS) was dispensed into all wells. The plate was incubated at room temperature for 2 hours, then washed five times with PBS. Extravidin-alkaline phosphatase (SIGMA, E236, 1 / 5000e in PBS 0.5% FCS) was added to each well (100 μL / well). The plate was incubated at room temperature for 1 hour, washed five times with PBS, and 100 μL of BCIP / NBT substrate solution (BCIP / NBT tablets, SIGMA, B5655; 0.45 μM filtered) was dispensed into each well. Plates were incubated in the dark at room temperature until clear spots were observed in the positive wells (approximately 5-10 minutes). Color development was stopped by emptying the plates and thoroughly washing with tap water. Plates were left uncovered in the dark at room temperature until dry (at least 1 hour).

[0179] Data collection The spots were counted with an ELISpot reader (CTL Immunospot reader, S5UV). Quality control was performed for each well to ensure that the number indicated by the ELISpot reader was consistent with the reality of the image. The result was 1 × 10 6 The mean number of spot-forming units (sfu) per splenic lymphocyte was expressed for each of the four replicates. Positivity was determined as described in Moody et al. (2006, J. Immunol Methods 315: 121-32) and Moody et al. (2010, Cancer Immunol Immunother doi 10.1007 / s00262-010-0875-4).

[0180] Example 1: Immunization with multi-peptide constructs Vector construction and production As illustrated in Figure 1, two MVA constructs, MVATG19022 and MVATG19023, were generated for the expression of fusions of five peptides derived from various antigens, separated from each other by a 10-amino acid GS linker for optimal processing of the fusion polypeptide and epitope presentation. Each peptide is a 27-mer and comprises a validated C57bl / 6 T cell epitope. The pentatope-encoding sequence is placed under the control of the p11k7.5 promoter (SEQ ID NO: 8) and a cassette inserted into deletion III of the MVA genome. More specifically, the pentatope fusion comprises, from 5' to 3', an FCU1-derived peptide (SEQ ID NO: 1), a b-Gal-derived peptide (SEQ ID NO: 2), an HPV-16 E1-derived peptide (SEQ ID NO: 3), an HPV-16 E7-derived peptide (SEQ ID NO: 4), and a MUC1-derived peptide (SEQ ID NO: 5), with 10-mer GS linkers before, between, and after the FCU1 peptide. A Flag tag (SEQ ID NO: 6) is present at the C-terminus to facilitate expression assessment. Furthermore, MVATG19023 is distinguished from MVATG19022 by the presence of a signal peptide at its N-terminus. For illustrative purposes, the pentatope fusion comprises the amino acid sequence set forth in SEQ ID NO: 7 for MVATG19022 and SEQ ID NO: 10 for MVATG19023 (with signal peptide).

[0181] Assessment of peptide expression Expression of the peptide fusions was assessed by Western blot in CEFs infected with each vector using an antibody against the tag Flag peptide included in the expression cassette. Briefly, CEFs were infected with MVATG19022 or MVATG19023, or empty (i.e., non-recombinant) MVA (TGN33.1 as a negative control) at an MOI of 0.2, or were not infected (mock control). Cells were harvested 24 h later, and cell lysates were analyzed by SDS-PAGE. Immunodetection of the expression products was performed with an anti-Flag antibody. Results show that bands corresponding to the expected sizes for the pentatope fusions targeted by MVATG19022 (19.7 kDa) and MVATG19023 (22.2 kDa) were detected. As expected, no bands were detected in cell lysates obtained from cells infected with empty MVA or mock buffer.

[0182] Furthermore, higher expression levels were observed when the cassette was equipped with a peptide signal, as evidenced by the stronger band obtained with MVATG19023 compared to that detected in cells infected with MVATG19022.

[0183] Immunogenicity assessment The ability of vectors expressing these two pentatopes to generate T cell responses was tested in C57bl / 6 mice by ELIspot IFNγ and compared with responses generated by MVA vectors encoding full-length antigens, MVATG9931 (MUC-1), MVATG15637 (FCU-1), MVATG18124 (β-galactosidase), MVATG8042 (HPV-16 E7), and MVATG17409 (HPV-16 E1), respectively.

[0184] More specifically, MVATG9931 (clinical name TG4010) is a recombinant MVA encoding the MUC1 tumor-associated antigen and human interleukin-2 (IL-2). TG4010 demonstrated efficacy in two separate randomized controlled phase 2b clinical trials in combination with first-line standard-of-care chemotherapy in advanced metastatic non-small cell lung cancer (NSCLC) (Quoix et al., 2011, The Lancet Oncol 12(12): 1125-33). MVATG15637 is a recombinant MVA encoding yeast-derived FCU1, which expresses the yeast CDase and UPRTase enzymes that convert the prodrug fluorocytosine (5-FC) to the cytotoxic 5-fluorouracil (5-FU) and 5-fluorouridine-5′-monophosphate (5-FUMP), respectively (Husseini et al., 2017, Ann Oncol 28(1): 169-74). MVATG18124 contains the bacterial LacZ gene encoding the β-galactosidase antigen under the control of the poxvirus promoter pH5R. MVATG8042 (clinical name TG4001) encodes a mutant inactivated human papillomavirus (HPV) 16 E6 and E7 oncoproteins and human interleukin-2 (WO99 / 03885). MVATG17409 encodes an HPV-16 E1 antigen modified to eliminate the replication function of the native polypeptide (WO2008 / 092854).

[0185] Briefly, five C57bl / 6 mice per group were subcutaneously injected twice with the corresponding MVA at a weekly interval (days 0 and 7). Each injection contained 2.5 × 10 7Animals were administered pfu of MVA (100 μL in the following buffer: Tris 10 mM [tris(hydroxymethyl)aminomethane / HCl], sucrose 5% (W / V), sodium glutamate 10 mM, sodium chloride 50 mM, pH 8) into the flank. One week after the last injection (day 14), all mice were sacrificed by cervical dislocation. Spleens were collected for lymphocyte isolation, and IFNγ-producing T cells were quantified by Elispot (cytokine-specific enzyme-linked immunospot) assay as described in Materials and Methods. Stimulation was performed with the following peptides specific for each expressed antigen / peptide: I8L (HPV16 E1): IAYKYAQL (SEQ ID NO: 50) I8V (β-galactosidase): ICPMYARV (SEQ ID NO: 51) R9F(E7 HPV16): RAHYNIVTF (SEQ ID NO: 52) L15L-3 (MUC1): LSYTNPAVAATSANL (SEQ ID NO: 53), and EG-15 (FCU-1): EKYHAAFPEVRIVTG (SEQ ID NO: 54) And as a control: T8V (specific for MVA): TSYKFESV (positive control) (SEQ ID NO: 55) K9i-3 (RMA, unrelated): KNGENAQAI (SEQ ID NO: 56) G15H (PyMT, unrelated): GICLMLFILIKRSRH (SEQ ID NO: 57)

[0186] As illustrated in Figure 2A-E, strong cellular responses were detected for four of the five epitopes expressed by MVATG19022 and MVATG19023 (although technical background issues hindered interpretation of the MUC-1 response). With the exception of the anti-E1 response, MVATG19023 tended to show a better response than MVATG19022. Addressing the peptide fusion to the endoplasmic reticulum likely protects the expressed fusion from degradation by the cellular proteasome and promotes its accumulation in infected cells where it is phagocytosed by professional antigen-presenting cells such as DCs.

[0187] Importantly, the IFNγ responses induced after immunization with MVATG19023 were comparable to those induced by the full-length antigen and even superior to the anti-E7 response (Figure 2C). Thus, the proximity of the peptides in the MVA backbone allows for the generation of diverse immune responses without the observed immunodominance phenomenon that shortens these responses.

[0188] Example 2: Immunization with CT26 neopeptide constructs Selection of CT26 neopeptides CT26 mutant neopeptides were selected for expression in MVA constructs (amino acid sequences shown in SEQ ID NOs: 22-31) based on various criteria described in the art, including low scores for MHC class I binding, the presence of CD8+ T cell epitopes (Kreiter et al., 2015, Nature 520 (7549): 692-6), and the presence of single residue mutations in the genome of the respective CT26 tumor cell line. Each neopeptide is a 27-mer comprising a mutation at the central position (position 14) and is separated from the following by a 10 amino acid GS linker for optimal processing and epitope presentation.

[0189] Vector construction and production Two MVA constructs, MVATG19030 and MVATG19038, were generated for expression of the above-described CT26 neopeptide arranged in two fusions (pentatopes). The first pentatope fusion is under the transcriptional control of the p11K7.5 promoter (SEQ ID NO: 8), and the second is under the control of pH5R (SEQ ID NO: 14). A Flag tag (SEQ ID NO: 6) is present at the C-terminus of the first pentatope fusion, and a c-MYC tag (SEQ ID NO: 11) is present at the end of the second fusion. MVATG19038 is distinguished from MVATG19030 by the presence of signal peptides derived from the rabies glycoprotein (SEQ ID NO:9) at the N-terminus of the first pentatope fusion and the measles F glycoprotein (SEQ ID NO:15) at the N-terminus of the second pentatope fusion. Two neopeptide fusion cassettes were inserted into deletion III of the MVA genome. In MVATG19030, the first and second pentatope fusions comprise the amino acid sequences set forth in SEQ ID NO:12 and SEQ ID NO:13, respectively (also illustrated in Figure 3), while MVATG19038 comprises the amino acid sequences set forth in SEQ ID NO:16 and SEQ ID NO:17.

[0190] Assessment of CT26 neopeptide expression Expression of the CT26 neopeptide fusion was assessed by Western blot in CEFs infected with MVATG19030 and MVATG19038. Briefly, CEFs were infected with MVATG19030 or MVATG19038 or empty (i.e., non-recombinant) MVA (TGN33.1 as a negative control) at an MOI of 0.2 or were not infected (mock control). Cells were harvested 24 hours later, and cell lysates were analyzed by SDS-PAGE. Immunodetection of the expression products was performed using antibodies against the Flag and c-MYC tags included in the expression cassette. Nonspecific bands were highlighted with the anti-MYC antibody but were also present in the negative control. Regardless of the background, pentatopes were detected with both antibodies corresponding to the expected sizes for the neopeptide fusions targeted by MVATG19030 (20.6 kDa for both pentatopes) and MVATG19038 (23.1 kDa for the first pentatope and 23.5 kDa for the second pentatope). As expected, no specific bands were detected in cell lysates from empty MVA-infected cells or mock buffer (although background bands were present in samples treated with anti-cMYC, as described above).

[0191] Thus, this study confirms the expression of the two neopeptide fusions in infected cells. As in Example 1, the presence of the peptide signal positively affected expression, and higher levels were observed for the neopeptide fusion with the peptide signal, as evidenced by the stronger band obtained with MVATG19038 compared to that detected in cells infected with MVATG19030.

[0192] Immunogenicity assessment The ability of MVATG19030 and MVATG19038 to generate a T cell response against the encoded CT26 neopeptide was tested by ELIspot IFNγ in a CT26 tumor model in Balb / c mice. Five animals were treated per group. Half of the Balb / c mice were injected with tumor CT-26MUC1 cells (2x10 in 100 μl) on day 0. 5 1 × 10 cells / mL of ... 7 MVA injections were administered intravenously on days 2 and 9 at pfu. Mice were sacrificed 1 week after the last MVA immunization (day 16). Spleens were harvested for lymphocyte isolation and processed as defined in Materials and Methods. Stimulation was performed with either mutant CT26 neopeptides (SEQ ID NOS: 22-31) or their non-mutated homologs (wild-type peptides listed in SEQ ID NOS: 32-41), or with either mutant (listed in SEQ ID NOS: 42-45) or wild-type (SEQ ID NOS: 46-49) CD8+ CT26 peptides. Wild-type CT26 peptides were pooled to reduce the number of samples, while mutant CT26 peptides were used either individually or in peptide pools.

[0193] As illustrated in Figure 4, after immunization with the MVATG19030 and MVATG19038 vectors and stimulation with a pool of 10 mutant CT26 peptides, high IFNγ responses were generated, regardless of the injected vector and the animal model (primed or not with CT26-MUC1 cells). In contrast, stimulation with the pool of 10 wild-type CT26 peptides did not result in any IFNγ production, nor did the medium-negative control.

[0194] Figure 5 illustrates the IFNγ response generated against one of the mutant peptides (EH27m; SEQ ID NO: 30) after immunization with MVATG19030 and MVATG19038 in a mouse model, both primed and unprimed with CT26-MUC1 cells. In both cases (priming and unpriming), responses ranging from 30 to 60 spots are generated. In contrast, stimulation with the wild-type peptide (PE27WT, SEQ ID NO: 37) does not result in any IFNγ production (fewer than 5 spots). Note that the response generated by the MVA vector with a signal peptide (MVATG19038) is slightly higher than that obtained with MVATG19030, whose fusion does not incorporate any signal peptide. Again, no response (fewer than 5 spots) is detected in the medium-negative control.

[0195] Furthermore, the response to the mutant peptide (EH27m) was inhibited after blocking with anti-mouse MHCII antibodies, as shown in Figure 6, which may indicate that the response was CD4+ type.

[0196] In conclusion, these results support the feasibility of personalized cancer vaccines for human anticancer therapy.

[0197] Example 3: Human Neopeptide Constructs Neopeptide identification and selection Whole-exome sequencing (WES) of tumor and germline samples from patients with NSCLC was performed in paired-end mode (2 × 150 bp) to identify tumor-specific somatic mutations. For information, the sequencing facility processed the raw data into a standard file format called FASTQ. Each sequencing run results in the generation of a FASTQ file, which provides short-read sequences and a per-base quality score, or Phred score, based on the estimated error probability during the base-calling process (Cock et al., 2010, Nucl. Acid Res. 38(6): 1767-71). Paired-end sequencing generates two FASTQ files, each corresponding to the end of a library fragment. Therefore, the resulting files must remain paired during the quality selection process; that is, both reads of a pair must pass selection. The criteria evaluated during this selection process are primarily based on Phred scores, particularly at the 3' end of the sequence, which often requires trimming to maintain accurate information for the variant calling process (Edgar and Flyvbjerg, 2015, Bioinformatics 31(21): 3476-82).

[0198] The short reads are then aligned to a reference genome (Li et al., 2009, Bioinformatics 25(16): 2078-79). After this mapping step, variant calling algorithms are used to detect various structural variations, such as single nucleotide variants (SNVs, including silent and non-silent missense mutations) or small insertions and deletions (indels) between the reference genome and the patient sequence (Danecek P. et al., 2011, Bioinformatics 27(15): 2156-58; Nielsen et al., 2011, Nature Reviews Genetics 12(6): 443-51). Samples from tumor and control tissues are processed simultaneously to detect tumor-specific mutations.

[0199] Mutations are then evaluated at the protein level, and peptides are designed based on the location of the mutations, with 17-mer mutations centered where possible. These candidate peptides are then evaluated at the expression level with data from the tumor transcriptome (Hundal et al., 2016, Genome Medicine 8(1): 11). The tumor transcriptome was measured by RNA-Seq in paired-end mode (2 × 150 bp) to assess neoantigen expression. Due to the sensitivity of RNA-Seq, genes were considered expressed if a read coverage depth of more than 3x was achieved. The quantification process begins after mapping by counting the number of reads mapped to a specific transcript, adjusted for library size, in total mapped reads per million (Wang et al., 2009, Nature Reviews Genetics 10(1): 57-63).

[0200] Thus, a total of 20 transcripts (corresponding to 18 unique genes) with one or more tumor-specific mutations were identified. Based on this, 18 tumor-specific mutations were selected and 18 neopeptides were designed.

[0201] The 18 neopeptides have the characteristics described in Table 2 below.

[0202] [Table 2]

[0203] Of these 18 selected neopeptides: All have sizes from 19 to 76. 17 / 18 (94%) are based on missense mutations resulting in a single amino acid change in the wild-type peptide, and 1 / 18 (6%) have frameshift mutations. Among the missense mutant neopeptides: They have a size of 19 to 29 amino acids, 15 / 17 (88%) had a central mutation as defined herein, and 14 / 17 (83%) have a size and central mutation of 29 amino acids.

[0204] Vector construction and production A search for transmembrane domains and signal peptides was performed for each peptide, using the Uniprot description of the corresponding gene as the source of information. None of the neopeptides were known to be part of either the transmembrane or signal peptide in their respective proteins. Uniprot was used to search for known functions associated with these peptides. The peptides were either outside of known domains or too short to form functional domains. Three expression cassettes containing 7 neopeptides (heptatopes), 6 neopeptides (hexatopes), and 5 neopeptides (pentatopes) were designed using the following rules: The 19-mer neopeptide NP1, comprising a missense mutation located at the neoantigen N-terminus, was placed at the N-terminus of the heptatope expression cassette. Similarly, the neopeptide NP18, which is naturally located at the C-terminus of the neoantigen, was placed at the C-terminus of an expression cassette encoding a pentatope. The neopeptide NP10 was truncated by its last three residues due to their possible involvement in disulfide bond formation and placed in the middle of a hexatope cassette to avoid any folding. Finally, the neopeptide NP13 was placed at the end of the hexatope-encoding cassette because it was predicted to form a potential transmembrane helix by the Geneious software.

[0205] The neopeptides were separated by a five-amino acid spacer for optimal processing and epitope presentation of the fusion polypeptide. Four different spacers were used: GSGSG, SGSGS, GSTSG, and SGTGS. The amino acid and codon sequences were varied to reduce the percentage of nucleotide identity between the different linkers and thus the risk of homologous recombination between the 17 linkers used in this construction. A signal peptide was fused to the N-terminus of each neopeptide fusion to direct the polypeptide to the RE. A signal peptide from the rabies glycoprotein was added to the N-terminus of the heptatope and pentatope, while a signal peptide from the measles F glycoprotein was added to the N-terminus of the hexatope.

[0206] To detect the expression of these three cassettes, virus-derived tags (i.e., HA (SEQ ID NO: 18), HSV (SEQ ID NO: 19), and VSV (SEQ ID NO: 20) tags for the heptatope, hexatope, and pentatope, respectively) were added to the C-terminus of each construct.

[0207] Three neopeptide fusion cassettes were inserted into deletion III of the MVA genome, resulting in MVATG19111. More specifically, the first cassette encoded a 255-residue heptatope fusion, the second a 239-residue hexatope fusion, and the third a 243-residue pentatope fusion. MVATG19111 was produced by homologous recombination in CEFs according to the methods described in Materials and Methods. This optimized protocol allowed for the generation of recombinant MVA in just 3 weeks and the production of purified bulk MVA in an additional week.

[0208] Expression assessment Expression of patient neopeptide fusions was assessed by Western blot in CEFs infected with MVATG19111. Briefly, CEFs were infected with MVATG19111 or empty (i.e., non-recombinant) MVA (TGN33.1 as a negative control) at an MOI of 0.2 or were not infected (mock control). Cells were harvested 24 hours later, and cell lysates were analyzed by SDS-PAGE. Immunodetection of expression products was performed using antibodies against the HA, HSV, and VSV tags contained in the expression cassette. Results show that specific bands were detected for each neopeptide fusion targeted by MVATG19111. The apparent sizes of the bands were slightly higher than expected for each fusion: approximately 38 kDa for the heptatope (expected size: 28.2 kDa), approximately 30 kDa for the hexatope (expected size: 25.5 kDa), and approximately 36 kDa for the pentatope (expected size: 25.9 kDa). This difference could be due to post-translational modifications. As expected, no specific bands were detected in cell lysates obtained from cells infected with empty MVA or mock buffer.

[0209] Immunogenicity assessment By design, MVATG19111 targets human antigens derived from tumor-specific mutations. Therefore, to evaluate the immunogenicity of this vaccine, it is necessary to use a humanized mouse model. The human leukocyte antigen (HLA)-A2.1 transgenic mouse (HHD) mouse model offers the possibility of generating an immune response specific to the human HLA A02*01 haplotype (Firat et al. 1999, European Journal of Immunology 29(10): 3112-21). Furthermore, the expression of mouse HLA molecules is disrupted in these mice, preventing the development of mouse-specific immune responses. The immunogenicity of the vaccine was evaluated using HHD mice. Mice were injected with 1.10 mAb in 200 μl of phosphate-buffered saline (PBS). 7pfu of MVATG19111 vaccine or MVA TGN33.1 was administered intravenously on days 1 and 7. Animals were sacrificed on day 14, and splenocytes were harvested and frozen until analysis. The number of IFN-γ-producing cells was determined by IFN-γ ELISpot.

[0210] Briefly, single cell suspensions of spleens from mice vaccinated with MVATG19111 or the control virus MVATGN33.1 were cultured in 200 μl complete medium at 5 × 10 cells per well under different stimulation conditions. 5 Cells were cultured in ELISPOT plates at 1000 cells / well for 48 hours. Stimulation conditions included a positive control (a mix of MVA peptides), a negative control (5 μg / ml of an irrelevant peptide), individual peptides, or one of 18 peptide pools, each related to one of the antigens expressed in the vaccine (5 μg / ml of overlapping 20-mer peptides), or overlapping peptides corresponding to non-mutated human proteins. After incubation, plates were washed, and IFN-γ-producing cells were detected using a colorimetric method. Spots were counted using commercially available automated software. An immune response was considered positive if the number of spots was statistically different from the analytical background.

[0211] Seven of 18 conditions resulted in positive immune responses in animals administered the MVATG19111 vaccine. More specifically, four conditions were highly immunogenic, inducing detectable T cell responses in all animals, whereas the other three conditions were immunogenic only in a subset of animals. The difference in immunogenicity may be related to assay sensitivity or immunophenotypic variability. Responses were restricted to the mutant protein; the vaccine did not induce immunization against the non-mutated protein. No responses were observed in control animals administered the control virus MVATGN33.1.

[0212] To further characterize the phenotype of responding T cells, ELISPOT assays were performed against highly immunogenic peptides in the presence of anti-class II MHC II or after depletion of CD8 cells using anti-CD8 depletion beads as previously described (Puliaev et al., 2004, J. Immunol. 173(2):910-9). Responses were observed both after CD8 depletion and MHC II blockade, indicating that MVATG19111 was able to induce both class I (CD8)- and class II (CD4)-mediated responses.

[0213] Production and therapeutic use The produced MVATG19111 bulk is stored at -80 or -20°C for approximately 10 days before being administered to patients with tumors. 8 It can be distributed in individual doses of pfu.

[0214] Example 4: Role of the linker in neoepitope fusions 4.1 Effect of linker on immunogenicity The effect of the linker was first tested, starting with the MVATG19023 vector described in Example 1, by varying the linker length and sequence. MVATG19023 encodes a fusion of five antigenic peptides derived from FCU1, β-galactosidase, HPV-16 E1, HPV-16 E7, and MUC1 (from the N- to C-terminus of the fusion). Each peptide was a 27-mer comprising a validated C57bl / 6 (b haplotype) T cell epitope located in the center of the 27-mer, except for Muc1, whose epitope corresponds to the last 15 C-terminal residues of the antigen (see Table 3). Ten-residue (5xGS) linkers were present at the N- and C-termini of each peptide, as well as a rabies signal peptide at the N-terminus of the fusion and a Flag tag at its C-terminus. The fusions were placed under the control of the p11k7.5 promoter and inserted into deletion III of MVA.

[0215] [Table 3]

[0216] In MVATG19023, the same 10-residue linker preceded and followed each epitope. To avoid potential homologous recombination events that could result in partial or complete cassette deletion, it is advisable to exploit the degeneracy of the genetic code to degenerate the DNA sequence of each linker. However, this sequence degeneracy is limited by the number, nature, and length of the linkers. When the number of epitopes to be vectorized is large (e.g., 10 or more), the genetic code does not allow for sufficient DNA diversity to avoid homologous recombination with the 5x GS linker. Therefore, we explored the possibility of using a smaller linker (GSG), a residue other than serine (GAS or GTS), and ultimately no linker at all. Viruses were produced by CEP. After clarification, the viruses were purified by tangential flow filtration (TFF). This resulted in the production of MVATG190158, MVATG190159, and MVATG190157, as exemplified in Table 4.

[0217] [Table 4]

[0218] The immunogenicity induced by these three constructs was assessed by ELISPOT and compared with that induced by MVATG19023 under the same immunization conditions described above.

[0219] ELISPOT results show that immune responses were broadly similar for MVATG19023, MVATG19158, MVATG19159 and MVATG19157 for four of the five epitopes (L15L3, I8L, I8V and R9F), but responses to EG15 tended to be slightly higher for MVATG19023, which contained a 10-residue linker.

[0220] These results indicate that the presence, nature and length of linkers at the N- and C-termini of the various expressed antigenic peptides do not have a significant effect on the immunogenicity of such peptides.

[0221] 4.2 Effect of linkers on the generation of recombinant poxviruses The ability of the constructs to generate recombinant poxvirus in the context of the pTG19111 vector (described in Example 3) was also tested with or without a linker. An additional plasmid was generated that contained the same neopeptide fusion as pTG19111, but without any linker, resulting in pTG19264. The yield of recombinant MVA virus was compared between the linker-free (pTG19264) and 5aa linker (pTG19111) constructs in transfected CEF cells described herein, with white plaques indicative of potential recombinant MVA and red plaques indicative of parental (non-recombinant) virus.

[0222] As shown in Table 5, both constructs resulted in the acquisition of white plaques, however, the proportion of these was higher with the linker-containing pTG19111 vector.

[0223] [Table 5]

[0224] Taken together, these results indicate that the absence of a linker negatively alters the number of white plaques produced in CEF cells and thus the production of recombinant MVA, but does not significantly affect immunogenicity. Therefore, to facilitate the design of vector constructs and therefore reduce the risk of deleterious homologous recombination events, especially in constructs where expression of more than 10 neopeptides is contemplated, but at the risk of reducing the percentage of recombinant poxviruses produced, the linker may be omitted if desired.

[0225] Example 5: Hydrophobicity analysis 5.1 TM Segment Forecast The presence of intra- or inter-peptide TM segments was tested in the context of the pTG19247 plasmid. As previously described, pTG19247 contains three expression cassettes, each encoding 10 neopeptides (the amino acid sequences of which are set forth in SEQ ID NOS: 60, 61, and 62) derived from a human lung adenocarcinoma (study PRJEB3132), with a signal sequence at the N-terminus of each fusion. To facilitate assembly, no linkers were included between the selected neopeptides. The generation of the corresponding MVA was assayed in CEFs by homologous recombination according to the method disclosed above, and the number of white and red plaques was counted. A low ratio of white plaques (less than 5%) was obtained, and PCR analysis of the white plaques revealed no recombinant MVAs with neopeptide fusions.

[0226] TM sequences in each neopeptide fusion were predicted using the TMHMM prediction algorithm. No TM sequence could be predicted in fusion 1. However, two potential interpeptide TM segments were identified in fusion 2, at the junctions of neopeptides 1 and 2 and neopeptides 7 and 8, respectively. On the other hand, three intrapeptide TM segments were identified in fusion 3, specifically within neopeptides 2, 4, and 5.

[0227] Next, a non-TM construct, designated pTG19258, was generated, comprising the same neopeptide fusion 1 as pTG19247; fusion 2, exemplified in SEQ ID NO: 63, in which neopeptides 1 and 2 and neopeptides 7 and 8 are inverted, allowing for suppression of these interpeptide TM segments (this new fusion 2 thus comprises, from N- to C-terminus, neopeptide 2, neopeptide 1, neopeptide 3, neopeptide 4, neopeptide 6, neopeptide 8, neopeptide 7, neopeptide 9, and neopeptide 10); and fusion 3, exemplified in SEQ ID NO: 64, in which neopeptides 2, 4, and 5 were omitted. Additionally, the promoter p11K7.5 used to drive fusion 3 in pTG19247 was replaced with the p7.5K promoter in pTG19258. Upon transfection of CEF cells, a higher percentage of white plaques was obtained. However, after isolating and amplifying 20 white plaques, only 6 white plaques remained, and PCR analysis revealed that all corresponded to the parental virus mutated in mCherry (none of them corresponded to the recombinant virus).

[0228] The effect of hydrophobicity was further explored by determining the hydrophobicity and hydropathy scores for each neopeptide and fusion.

[0229] 5.2 Hydrophobicity analysis Hydrophobicity analysis was performed to determine the hydrophobicity and hydropathy scores of all neopeptides encoded by constructs MVATG19111, pTG19247, and pTG19258 using the Kyte-Doolittle method. The results for the three fusions encoded by MVATG19111 are reported in Table 6 below.

[0230] [Table 6]

[0231] These results show that 17 of the 18 neopeptides encoded by MVATG19111 exhibited favorable hydrophobicity and hydropathy scores (negative values ​​and less than 0.1, respectively). Only neopeptide 12, contained in the second fusion, exhibited a positive score, reflecting the hydrophobic nature of this particular peptide. Fusion 2, however, maintained its hydrophilic nature, as reflected by its negative hydrophobicity (-74) and hydropathy (-0.42) scores.

[0232] The same prediction test was performed on the neopeptides and fusions encoded by pTG19247 and pTG19258, and the scores are reported in Tables 7 and 8, respectively.

[0233] [Table 7]

[0234] The results highlight the highly hydrophobic nature of fusion 3 as evidenced by the positive hydrophobicity (133.5) and hydropathy (0.494) scores.

[0235] The same study was performed in the context of pTG19258, which corresponds to the TM-deleted version of pTG19247 described above. The hydrophobicity and hydropathy scores for fusions 1 and 2 are the same as those reported in Table 7 (the same fusion carried by both vectors, and fusion 2, which comprises the same neopeptide with the order of neopeptides 1 and 2 and 7 and 8 in pTG19258 reversed to eliminate the interpeptide TM segment). Table 8 shows the hydrophobicity and hydropathy scores for fusion 3 of pTG19258 (a fusion of seven neopeptides with the TM-containing neopeptides 2, 4, and 5 deleted relative to its pTG19247 counterpart).

[0236] [Table 8]

[0237] Despite the exclusion of the highly hydrophobic and TM-containing neopeptides 2, 4, and 5, the global score for pTG19258 fusion 3 is in a hydrophobic nature.

[0238] Therefore, further constructs were made, which were respectively as follows: A version of pTG19258 lacking fusion 3 (referred to as pTG19288) A version of pTG19258 (pTG19267) containing a three amino acid linker at the N-terminus of the first neopeptide and between each neopeptide in the three fusions. Elimination of the third fusion in pTG19267 (pTG19293) Substitution of a third fusion in pTG19258 with another containing 10 other neopeptides selected from the public lung adenocarcinoma databank and predicted to be hydrophilic (negative hydrophobicity score), without linkers (pTG19290; SEQ ID NO: 65), or with a three amino acid linker at the N- and C-termini of each neopeptide (pTG19291). Elimination of fusion 1 in pTG19291 (pTG19298).

[0239] The yield of recombinant MVA virus was assayed in CEF cells transfected with these novel plasmid constructs described herein and compared to the parental plasmid pTG19258. Red plaques correlate with parental (non-recombinant) virus, while white plaques are indicative of potential recombinant MVA. However, further PCR analysis of the panel of white and red plaques is required to distinguish between non-recombinant (i.e., parental) and recombinant viruses (mCherry-negative or mCherry-positive, the latter reflecting contamination of recombinant plaques with the parental). The results are summarized in Figure 8.

[0240] As shown in Figure 8, pTG19258 did not result in the generation of any recombinant MVA viruses. Although white plaques were produced, viruses with fusions could not be detected by PCR analysis, and no TM segment could be identified in this construct (however, fusion 3 showed a positive score, reflecting its hydrophobic nature). The white plaques were due to the parent virus mutated in the mCherry marker gene.

[0241] Addition of a linker between the neopeptides in each fusion cassette reduced the hydrophobic character of fusion 3 (the hydropathy score decreased from 0.151 to 0.08, a value less than 0.1). This modification resulted in a high percentage of white plaques (13.3%) in CEF cells, and seven recombinant viruses were obtained, as confirmed by PCR analysis.

[0242] Replacing hydrophobic fusion 3 in pTG19258 with another consisting of 10 hydrophilic neopeptides (hydropathy score −0.52) was also beneficial for generating recombinant MVA viruses, as 7 and 6 recombinant viruses were generated, respectively, regardless of the presence (pTG19291) or absence (pTG19290) of a linker in the neopeptide fusion.

[0243] Elimination of fusions leaving only fusions with two of the ten neopeptides, each scoring negatively for hydrophobicity, increased the percentage of white plaques to 16% and the recovery of recombinant virus (10 and 11 recombinants were identified by PCR).

[0244] Taken together, these results point to optimal designs of neopeptide fusions, as summarized below: elimination of potential TM segments (intra- or inter-peptides); Selecting somewhat hydrophilic neopeptides (negative scores for hydrophobicity) or reducing the number of hydrophobic neopeptides (negative scores for hydrophobicity) to, for example, less than 40%; placement of neopeptides in fusions; The presence of a signal peptide at the N-terminus of each fusion is recommended The presence of a three amino acid linker between the neopeptides and at the N-terminus of the first neopeptide (e.g., GTS, GSG, or GAS, as illustrated in Figure 7) Incorporation of one to three fusions of six to ten neopeptides into the viral genome.

[0245] References Acres and Bonnefoy, 2008, Expert Review of Vaccines 7, 889-93; Altschul et al., 1990, J. Mol. Biol. 215(3):403-10; Anderson and Schrijver, 2010, Genes 1(1): 38-69; Andreatta et al., 2015, Immunogenetics 67(11-12): 641-50; Antoine et al., 1998, Virol. 244: 365-96; Bainbridge et al., 2010, Genome Biol.11:R62; Boegel et al., 2015, Methods Mol Biol 1310: 247-51; Caroll et al., 1997, Virology 238: 198-211; Carpentier et al., 2003, Frontiers in Bioscience 8, e115-127; Carpentier et al., 2006, Neuro-Oncology 8(1): 60-6; Chakrabarti et al. 1997, Biotechniques 23: 1094-7; Chu and Corey, 2012, Nucleic Acid Ther. 22: 271-4; Claudepierre et al., 2014, J. Virol. 88(10): 5242-55 ; , Nucl. Acid Res. 38(6): 1767-71; Danecek P. et al., 2011, Bioinformatics 27(15): 2156-58; , Bioinformatics 31(21): 3476-82; EP 1 162 982; EP 463 756; Erbs et al., 2008, Cancer Gene Ther. 15(1): 18-28; Farsaci et al., 2011, In Cancer Vaccines: From Research to Clinical Practice, Ed Bot; CRC Press, pp56-77; Fend et al., 2014, Cancer Immunol. Res. 2, 1163-74; Firat et al. 1999, European Journal of Immunology 29(10): 3112-21; Gulley et al., 2008, Clin Cancer Res 14(10): 3060-9; Guse et al., 2011, Expert Opinion Biol. Ther.11(5):595-608; Hammond et al, 1997, J. Virol Methods 66: 135-8; Hundal et al., 2016, Genome Medecine 8(1): 11; Husseini et al., 2017, Ann Oncol 28(1): 169-74 ; Kallol et al., 2003, J. Chromat. 1000: 637-55; Kern et al., 1990, Gene 88: 149-57; Kreiter et al. 2015, Nature, 520: 692-6 ; Krogh et al., 2001, J. Mol. Biol. 305: 567-80; Kumar and Boyle, 1990, Virology 179: 151-8; Kyte and Doolittle, 1982, J. Mol. Biol. 157: 105-32; Li et al., 2009, Bioinformatics 25(16): 2078-79; Maniatis et al. 1989, Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor NY; Mayr et al., 1975, Infection 3: 6-14; Meyer et al., 1991, J. Gen. Virol. 72: 1031-8; Moodie et al. 2006, J. Immunol Methods 315: 121-32; Moodie et al. 2010, Cancer Immunol Immunother doi 10.1007 / s00262-010-0875-4; Ng S.B. et al, 2009, Nature 461: 272-6 ; Nielsen et al., 2010, Immunology 130(3): 319-28; Nielsen et al., 2011, Nature Reviews Genetics 12(6): 443-51; Olivier et al., 2010, mAbs 2(4): 405-15; Plotkin, 2008, Clin Infect Dis.47(3):401-9; Quoix et al., 2011, The Lancet Oncology 12(12): 1125-33; Relman, 2008, J Infect Dis. 198(1):4-5; Rock et al., 2010, J. Immunol. 184(1): 9-15; Rose et al., 1993, Ann. Rev. Biomol. Struc. 22: 381-415; Sutter and Moss, 1992, Proc. Natl. Acad. Sci. USA 89: 10847-51; Sweet et al., 1983, J. Mol. Biol. 171: 479-88; US 5,250,534; US 5,972,597; US 6,440,422; US 6,469,012; US 6,998,252; US 7,108,844; US 7,700,569; US 5,879,924; Wang et al., 2009, Nature Reviews Genetics 10(1): 57-63; WO03 / 008533; WO03 / 053463; WO2005 / 007840; WO2005 / 042728; WO2006 / 0850082; WO2007 / 056847; WO2007 / 077256; WO2007 / 147528; WO2008 / 092854; WO2008 / 114021; WO2008 / 129058; WO2008 / 138533, WO2009 / 004016; WO2009 / 065546; WO2009 / 065547; WO2009 / 100521; WO2010 / 130753; WO2010 / 130756; WO2012 / 001075; WO2012 / 159754; WO2013 / 022764; WO2014 / 053571; WO2015 / 175334; WO2015 / 175340; WO2016 / 187508; WO2016 / 191545; WO2016 / 207859; WO97 / 02355; WO99 / 03885; Yuan et al., 2015, J. Virol 89, 5176-9, doi:10.1128; Yuan et al., 2016, Viruses 8, 72, doi:10.3390 ;

Claims

1. A personalized cancer vaccine comprising a recombinant poxvirus encoding one or more neopeptides.

2. The personalized cancer vaccine of claim 1 , wherein the recombinant poxvirus belongs to the genus Orthopoxvirus, and is preferably a vaccinia virus.

3. 3. The personalized cancer vaccine of claim 2, wherein the recombinant poxvirus is a replication-deficient vaccinia virus, such as MVA.

4. The personalized cancer vaccine of any one of claims 1 to 3, wherein each of the one or more neopeptides encoded by the recombinant poxvirus comprises one or more tumor-specific mutations.

5. 5. The personalized cancer vaccine of claim 4, wherein at least 60% of the neopeptides comprise missense or frameshift mutations.

6. 6. The personalized cancer vaccine of claim 5, wherein at least 60% of the neopeptides comprise a missense mutation.

7. 7. The personalized cancer vaccine of claim 1, wherein the one or more neopeptides have a length of 16 to 90 amino acid residues, preferably 17 to 85 amino acid residues, more preferably 18 to 80 amino acid residues.

8. 8. The personalized cancer vaccine of claim 5, wherein the neopeptide comprising a missense mutation has a length of 18 to 29 residues and the neopeptide comprising a frameshift mutation has a length of 30 to 80 amino acid residues.

9. 9. The personalized cancer vaccine of any one of claims 5 to 8, wherein at least 80%, preferably at least 85%, more preferably at least 90% of the neopeptides with missense mutations have a substituted amino acid at the central position.

10. 10. The personalized cancer vaccine of claim 1, wherein some, preferably all, neopeptides are expressed by the recombinant poxvirus in the form of one or more fusions, preferably one or more fusions of 2 to 15 neopeptides.

11. The personalized cancer vaccine of claim 10 , wherein the fusion does not include any TM segments.

12. 12. The personalized cancer vaccine of claim 10 or 11, wherein the fusion scores negatively for hydrophobicity.

13. The personalized cancer vaccine of any one of claims 10 to 12, wherein the fusion exhibits a hydropathy score of 0.1 or less.

14. The recombinant poxvirus encodes two or three fusions, wherein: each fusion comprises 5-10 neopeptides; each fusion comprising a linker, preferably a linker three amino acids in length, at the N-terminus of the first neopeptide and between each neopeptide; each fusion comprising a signal peptide at the N-terminus of the fusion; and each fusion optionally comprising a tag at the C-terminus of the fusion; and Each fusion is exhibit a negative global hydrophobicity score and / or a global hydropathy score of 0.1 or less, and The personalized cancer vaccine of any one of claims 10 to 13, which does not contain any potential TM segments.

15. 15. The personalized cancer vaccine of any one of claims 1 to 14, wherein the recombinant virus is a recombinant MVA comprising 1 to 3 cassettes, each cassette for expression of a fusion of 5 to 10 neopeptides under the transcriptional control of a vaccinia promoter selected from the group consisting of p11k7.5, pH5R, p7.5K, and pB2R.

16. The personalized cancer vaccine according to any one of claims 1 to 15, wherein the recombinant poxvirus further encodes one or more therapeutic genes, preferably one or more therapeutic genes selected from the group consisting of suicide genes and immunostimulatory genes.

17. 17. A method for preparing a personalized cancer vaccine according to any one of claims 1 to 16, comprising the step of identifying one or more neopeptides suitable to be encoded by a recombinant poxvirus, wherein the one or more neopeptides comprise one or more tumor-specific mutations.

18. The method comprises an identifying step, The identifying step comprises the following substeps a) to d): a) extracting DNA from tumor and non-tumor samples; b) selecting a target region, preferably the entire coding region of the genome (exome); c) sequencing the target region (e.g., exome) from the extracted DNA; and d) identifying one or more tumor-specific mutations by comparing DNA sequences obtained from said tumor and non-tumor samples; 18. The method of claim 17, comprising:

19. 19. The method of claim 18, wherein the tumor sample is preferably a tumor biopsy comprising more than 20% tumor cells, and the non-tumor sample is a biological fluid, cytological material or a biopsy.

20. said identifying step comprising the following further sub-steps e) to i): e) ranking potential neopeptides by their expression level in the tumor, either at the mRNA transcription level or at the protein translation level; f) selecting non-self-expressing tumor-specific mutations; g) predicting the immunopotency of neoepitopes contained in said neopeptides; h) predicting the presence of potential TM segments in the neoepitope itself and / or in the neopeptide fusion comprising several neopeptides; i) ranking potential neopeptides and / or neopeptide fusions by degree of hydrophobicity, and in particular selecting neopeptides and / or neopeptide fusions that exhibit a negative global hydrophobicity score and / or a global hydropathy score of 0.1 or less. The method of any one of claims 17 to 19, further comprising one or more of:

21. 21. The method of claim 20, wherein the method further comprises the step of producing the recombinant poxvirus.

22. 22. The method of claim 21, wherein the nucleic acid molecules encoding the one or more neopeptides to be inserted into the genome of the recombinant poxvirus are placed within one or more expression cassettes under the control of suitable regulatory elements allowing expression in the subject.

23. 23. The method of claim 21 or 22, wherein the nucleic acid molecule or expression cassette encoding the one or more neopeptides is inserted into the genome of a parent poxvirus to generate the recombinant poxvirus, preferably wherein the parent poxvirus comprises a fluorescent reporter gene cloned at the insertion site selected for the nucleic acid molecule or expression cassette encoding the neopeptide.

24. 24. The method of claim 23, wherein the fluorescent reporter is selected from the group consisting of GFP (green fluorescent protein), eGFP (enhanced green fluorescent protein), AmCyan1 fluorescent protein, and mCherry.

25. 25. The method of claim 24, wherein the step of producing the recombinant poxvirus comprises a further step of cleavage with an endonuclease capable of generating at least one double-stranded break in the nucleotide sequence of the fluorescent reporter, wherein the endonuclease does not cleave the genome of the poxvirus.

26. 26. The method of claim 25, wherein the endonuclease is selected from the group consisting of zinc finger nucleases (ZFNs), transcription activator-like effector nucleases (TALENs), clustered regularly interspaced short palindromic repeats (CRISPR) / Cas9 nucleases, and restriction enzymes with unique cleavage sites within fluorescent reporter genes.

27. 27. The method according to any one of claims 21 to 26, wherein said step of generating the recombinant poxvirus allows generating the recombinant poxvirus in up to 4 weeks, preferably up to 3 weeks, at a recombinant:parental ratio comprised between 2 and 0.

05.

28. 28. The method according to any one of claims 17 to 27, wherein said method further comprises a production step, said production step comprising a step of amplifying to a suitable scale in suitable producer cells, a step of recovering the produced recombinant poxvirus from the cell culture, and an optional step of purifying the recovered recombinant poxvirus.

29. 29. The method of claim 28, wherein the producer cells are chicken embryo fibroblasts (CEF).

30. 30. The method according to claim 28 or 29, wherein the step of recovering the produced recombinant poxvirus comprises a lysis step in which the producer cell membranes are disrupted, preferably by using a high speed homogenizer.

31. 31. The method of any one of claims 28 to 30, wherein the purification step comprises a tangential flow filtration (TFF) step.

32. The manufacturing process comprises at least 10 ９ pfu, preferably at least 5 x 10 ９ pfu, preferably approximately 10 １０ 32. The method according to any one of claims 28 to 31, wherein the production of pfu or more is reached.

33. 33. The personalized cancer vaccine according to any one of claims 1 to 16, or obtained according to the method of any one of claims 17 to 32, wherein the personalized cancer vaccine is a composition comprising a therapeutically effective amount of the recombinant poxvirus and a pharmaceutically acceptable vehicle.

34. The therapeutically effective amount for each individual dose is approximately 5 x 10 ７ pfu ~ approximately 10 ９ 34. The personalized cancer vaccine of claim 33, wherein the individualized cancer vaccine is pfu.

35. 35. The personalized cancer vaccine of any one of claims 33 or 34, for use in a subject in need thereof to treat or prevent the recurrence of cancer in the subject.

36. 36. The personalized cancer vaccine for use according to claim 35, wherein the cancer is a solid tumor.

37. 37. The personalized cancer vaccine for use according to claim 36, wherein the cancer is brain cancer, such as astrocytoma, embryonal tumor, germ cell tumor, central nervous system atypical teratoid / rhabdoid tumor, craniopharyngioma, ependymoma, glioma and glioblastoma, and head and neck cancer.

38. 37. The personalized cancer vaccine for use according to claim 36, wherein the cancer is lung cancer, in particular NSCLC, particularly preferably adenocarcinoma, squamous cell carcinoma and large cell carcinoma.

39. 37. The personalized cancer vaccine for use according to claim 36, wherein the cancer is ovarian cancer.

40. 40. The personalized cancer vaccine for use according to any one of claims 35 to 39, wherein said personalized cancer vaccine is administered to a patient by intravenous, subcutaneous, intramuscular or intratumoral injection.

41. 41. The personalized cancer vaccine for use according to any one of claims 35 to 40, wherein said personalized cancer vaccine is administered in combination with one or more additional anti-cancer therapies having utility in the treatment of cancer.

Images