CN118265538A - Transmembrane neoantigenic peptides - Google Patents

Abstract

The present disclosure provides transmembrane chimeric proteins derived from Transposable Element (TE) -exon fusion transcripts, as well as nucleic acids, antibodies, CARs, non-HLA-restricted TCRs, and immune cells targeting such chimeric proteins, useful in the treatment of cancer.

Description

Transmembrane neoantigenic peptides

Technical Field

The present disclosure provides transmembrane chimeric proteins and tumor neoantigen peptides encoded by transposable element (TE)-exon fusion transcripts that can be used for cancer treatment, nucleic acids, vaccines, antibodies and immune cells.

Background technique

The advent of T cell-based immunotherapy has revolutionized high-precision tumor targeting using infusions of activated genetically engineered T cells and other immune cells, or by delivering mono- or bispecific antibodies such as BiTEs. Chimeric antigen receptor (CAR) T cells and BiTEs are the main forms of T cell redirected immunotherapy, using single-chain variable fragments (scFv) to target tumors to induce target cell death. Redirecting immune cell targeting using these approaches has enabled the elimination of malignant cells that were previously "invisible" to the immune system and has provided excellent therapeutic effects for patients with relapsed or refractory tumors. This is particularly effective in the case of CAR T cells, where the fusion of an antibody binding domain to a T cell signaling protein (e.g., CD3) is able to redirect T cells to antigen specificity. The main advantage of CARs is that T cells are activated and can exert effector functions, such as the release of cytotoxic granules and cytokines, without recognizing peptides presented by the major histocompatibility complex (MHC) because CARs interact directly with cell surface molecules.

Despite the ability to engineer, redirect, and influence cellular functions and interactions, challenges remain associated with targeting proteins expressed on tumor cells. The lack of true tumor-specific antigens (TSAs) and the development of intratumoral antigen escape remain major challenges to effective targeted therapy, and approaches such as dual antigen targeting have shown promising early results. These challenges have prompted efforts to discover biomarkers for both hematological and solid malignancies.

TSAs are expressed only in malignant tumors and are generally considered to be mutated proteins presented on the cell surface via MHC. However, the category of TSAs can be expanded to include proteins derived from tumor-specific fusion transcripts, tumor-specific glycosylation, tumor-specific mutations of cell surface proteins, and misfolded proteins that escape refolding in the endoplasmic reticulum.

Therapeutic targeting of tumor-associated antigens (TAAs) has been successful in some cases, but comes with the caveat of potential off-target effects that may be associated with treatment. Such targets are broadly defined as having higher expression levels in malignant tissue compared to matched healthy tissue, such as human epidermal growth factor receptor 2 (HER2), or as having expression that is lineage restricted, such as CD19. Unlike TSAs, TAAs often have on-target side effects, making it more difficult to distinguish direct treatment-related side effects from on-target/off-tumor toxicities. CAR T cell therapy targeting CD19 in CD19-expressing tumors was a breakthrough therapy that demonstrated the clinical efficacy of CAR T cells. Therefore, targeting lineage-specific TAAs is possible but only justified if healthy tissue is deemed dispensable or there is an acceptable level of toxicity.

Identification of shared bona fide TSAs (not present in tissues) or TAAs with minimal off-target risk is a major challenge in the field of immuno-oncology.

Some existing reports on transposable elements (TEs) in tumors include (Helman, E. et al. (2014). Genome Res.)(Schiavetti, F. et al. (2002). Cancer Res., Takahashi, Y. et al. (2008). J. Clin. Invest.).(Chiappinelli, K. B. et al. (2015). Cell, Roulois, D. et al. (2015). Cell). However, the relationship between TEs and the antigenic profile expressed by tumor cells has not been studied in depth.

New tumor neoantigens will gain interest and have the potential to improve or reduce the cost of cancer treatment, especially with cell therapy, targeted therapy via antigen-binding domains, and vaccination strategies.

SUMMARY OF THE INVENTION

The present disclosure provides chimeric polypeptides (or proteins) and nucleotide sequences encoding such polypeptide sequences; antibodies or antigen-binding fragments thereof, T cell receptors (TCRs, particularly non-HLA restricted TCRs) or chimeric antigen receptors (CARs) that specifically bind to such chimeric polypeptides or proteins; methods for producing such antibodies, TCRs or CARs; polynucleotides encoding such new antigenic peptides, antibodies, CARs or TCRs, which are optionally linked to heterologous regulatory sequences; immune cells that specifically bind to such chimeric polynucleotides or proteins; and methods, particularly therapeutic methods using such products.

The present disclosure provides a tumor chimeric polypeptide (or protein) comprising or consisting of any one of SEQ ID NOs: 1-21542 and comprising a neoantigen sequence, wherein the protein is located on a cell membrane.

More specifically, the present disclosure further provides an isolated tumor neoantigen sequence (generally an epitope), wherein the sequence is derived from any one of the chimeric proteins of SEQ ID NO: 1-8202, including fragments thereof, and comprises at least a portion of a TE-derived amino acid sequence, or is derived from any one of SEQ ID NO: 1424-8202, 8203-10163, and 12831-21542, in particular derived from a chimeric fusion transcript sequence in which the donor is TE. In some embodiments, the tumor neoantigen sequence overlaps the breakpoint between the TE-derived amino acid sequence and the exon-derived amino acid sequence. In other embodiments, the tumor neoantigen sequence is derived from a pure TE sequence. In other embodiments, the tumor neoantigen sequence is encoded by a non-classical ORF downstream of the connection between the TE-derived amino acid sequence and the exon-derived amino acid sequence. Typically, the tumor neoantigen sequence is derived from the extracellular portion of the chimeric protein to which it belongs.

Typically, the transmembrane chimeric protein is expressed in more than 1%, particularly more than 5%, and often more than 10% of tumor samples.

Typically, the transmembrane chimeric protein is expressed at higher levels in tumor samples compared to normal samples.

Typically, the expression of the chimeric protein in a normal sample is less than 20%, particularly less than 10%, less than 5% or less than 1%.

In certain embodiments, the portion of the chimeric protein sequence derived from the TE nucleotide sequence is exposed on the cell surface.

The present disclosure further includes an antigen binding domain that binds to a transmembrane chimeric protein as defined herein, in particular a tumor neoantigen sequence (generally an epitope) from any one of the chimeric proteins of SEQ ID 1-8202, with a Kd binding affinity of less than about 10 ^-7 M. An antibody, TCR or CAR that specifically binds to a transmembrane chimeric polypeptide or protein disclosed herein may bind to a tumor neoantigen peptide sequence of at least 4, at least 5, at least 6 or at least 7 amino acids.

In some embodiments, the antigen binding domain binds to a sequence from any of the chimeric proteins disclosed herein, the sequence comprising at least a portion of a TE-derived amino acid sequence or from any of SEQ ID NOs: 1424-8202, 8203-10163, and 12831-21542, particularly derived from a chimeric fusion transcript sequence in which the donor is a TE. In some embodiments, the sequence or fragment of the chimeric protein bound by the antigen binding domain overlaps the breakpoint between the TE-derived amino acid sequence and the exon-derived amino acid sequence. In other embodiments, the new antigenic peptide is derived from a pure TE sequence. In other embodiments, the sequence or fragment of the chimeric protein bound by the antigen binding domain is encoded by a non-classical ORF downstream of the junction between the TE-derived amino acid sequence and the exon-derived amino acid sequence.

In certain embodiments, an antigen binding domain comprises one or more, typically one or two, immunoglobulin regions.

Notably, the antigen binding domain may comprise an antibody heavy chain variable region (VH) and/or an antibody light chain variable region (VL).

The present disclosure also includes antibodies comprising an antigen binding domain as defined herein, wherein the antibody is selected from a complete IgG, a scFv, a BiTE or a multispecific antibody. The antibody may be derived from human, mouse or camelid.

The present disclosure also includes a chimeric antigen receptor (CAR) or a non-HLA restricted recombinant T cell receptor (TCR) comprising an antigen binding domain as defined herein.

The non-HLA restricted recombinant TCRs of the present disclosure generally comprise an extracellular antigen binding domain that can dimerize with a second extracellular antigen binding domain. In some embodiments, the second extracellular antigen binding domain binds to a tumor antigen, preferably wherein the tumor antigen is selected from pHER95, CD19, MUC16, MUC1, CAIX, CEA, CD8, CD7, CD10, CD20, CD22, CD30, CD70, CLL1, CD33, CD34, CD38, CD41, CD44, CD49f, CD56, CD74, CD133, CD138, EGP-2, EGP-40, EpCAM, Erb-B2, Erb-B3, Erb-B4, FBP, fetal acetylcholine receptor, folate receptor-a, GD2, GD3, HER-2, hTERT, IL- 13R-a2, kappa-light chain, KDR, LeY, L1 cell adhesion molecule, MAGE-A1, mesothelin, MAGEA3, p53, MART1, GP100, proteinase 3 (PR1), tyrosinase, survivin, hTERT, EphA2, NKG2D ligand, NY-ESO-1, carcinoembryonic antigen (h5T4), PSCA, PSMA, ROR1, TAG-72, TAG-R2, WT-1, BCMA, CD123, CD44V6, NKCS1, EGF1R, EGFR-VIII, CD99, CD70, ADGRE2, CCR1, LILRB2, LILRB4, PRAME and ERBB.

The present disclosure further includes a CAR comprising:

a) an extracellular domain comprising one or more antigen binding domains, at least one of which is selected from the antigen binding domains described herein,

b) transmembrane domain,

c) optionally one or more co-stimulatory domains, for example selected from CD28, 4-1BB (CD137), ICOS-1, CD27, OX 40 (CD137), DAP10 and GITR (AITR)

d) one or more intracellular signaling domains comprising one or more ITAMs, for example: an intracellular signaling domain from CD3-ζ or a portion thereof, or a variant lacking one or two ITAMs (for example: ITAM3 and/or ITAM2, see above for details and references), FcRγ, FcRβ, CD3γ, CD3δ, CD3ε, CDS, CD22, CD79a, CD79b and/or CD66d, in particular an intracellular signaling domain comprising a modified CD3ζ intracellular signaling domain in which ITAM2 and ITAM3 have been inactivated,

In some embodiments, a CAR disclosed herein comprises a transmembrane domain selected from CD28, CD8, or CD3-ζ.

In some embodiments, the CAR disclosed herein comprises one or more co-stimulatory domains, which can be selected from: CD28, 4-1BB (CD137), ICOS-1, CD27, OX 40 (CD137), DAP10 and GITR (AITR).

In some embodiments, a CAR disclosed herein comprises an intracellular signaling domain comprising an intracellular signaling domain of a CD3-zeta polypeptide, or a fragment thereof, optionally comprising a CD3-zeta polypeptide in which the immunoreceptor tyrosine-based activation motif 2 (ITAM2) and the immunoreceptor tyrosine-based activation motif 3 (ITAM3) are inactivated.

This disclosure also covers:

-An antibody or antigen-binding fragment thereof, a T cell receptor (TCR), or a chimeric antigen receptor (CAR) selected for its binding affinity to a chimeric protein from any one of SEQ ID NOs: 1-21542, including a portion thereof, such as at least 4, 5, 6, 7 or 8 amino acids in length, or a composition comprising such an antibody, antigen-binding fragment thereof, TCR or CAR.

- a polynucleotide encoding a neoantigenic peptide, antibody, CAR or TCR as defined herein, which is usually operably linked to a heterologous regulatory nucleotide sequence, and a vector encoding such a polynucleotide;

- An immune cell or immune cell population targeted to one or more chimeric proteins from any one of SEQ ID NOs: 1-21542, including a portion thereof, such as at least 4, 5, 6, 7 or 8 amino acids in length, wherein the immune cell population is preferably targeted to a plurality of different chimeric proteins or fragments thereof as disclosed herein, or a composition comprising such an immune cell or immune cell population, optionally in combination with a physiologically or pharmacologically acceptable buffer, carrier, excipient, immunostimulant and/or adjuvant.

Typically, the antibody or antigen-binding fragment thereof, TCR or CAR binds to the chimeric protein or fragment thereof expressed on the surface of a cell with a Kd affinity of about 10 ^-6 M or less.

The present disclosure further provides a method for producing an antibody, a non-HLA restricted TCR or a CAR as defined herein, comprising the following steps: selecting an antibody, a non-HLA restricted TCR or a CAR that binds to a chimeric protein or a fragment thereof of the present disclosure, generally any one of SEQ ID NOs: 1-21542 transmembrane chimeric polypeptide sequences, wherein the Kd affinity constant is about 10-7 M or less. The present disclosure also includes antibodies, CARs and TCRs produced by this method.

The present disclosure further encompasses polynucleotides encoding chimeric proteins or polypeptides, antibodies, CARs and/or non-HLA restricted TCRs as defined herein, optionally connected to heterologous regulatory sequences, and vectors comprising them. The present disclosure also includes immune cells comprising CAR and/or TCR, particularly non-HLA restricted TCRs as defined herein. The immune cells may be allogeneic or autologous. They are generally selected from T cells, natural killer T cells, CD4+/CD8+T cells, TIL/tumor-derived CD8T cells, central memory CD8+T cells, Treg, MAIT, YδT cells, human embryonic stem cells, and pluripotent stem cells that can differentiate into lymphocytes. In certain embodiments, immune cells are Suv39h1 defective, particularly in the immune cells, the Suv39h1 gene is destroyed by deletion of the entire gene, exon or region, replacement with an exogenous sequence, and/or by frameshift or missense mutations within the suv39h1 gene. The Suv39h1 gene is typically a human Suv39h1 gene encoding the O43463 human Suv39h1 protein mentioned in UniProt. Methods of preparing such immune cells are also contemplated, for example, by delivering a nucleic acid or vector encoding any antibody, TCR or CAR as described herein to a cell in vivo or ex vivo.

The present disclosure further encompasses a pharmaceutical composition comprising an effective amount of an immune cell as defined herein and a pharmaceutically acceptable excipient.

The present disclosure also encompasses therapeutic uses, in particular, the therapeutic use of chimeric proteins or polypeptides, antibodies, non-HLA restricted TCRs, CARs, polynucleotides, vectors, immune cells or compositions comprising them as defined herein to inhibit cancer cell proliferation or cancer treatment in subjects in need of treatment. Typically, such compositions further comprise a pharmaceutical excipient. The treatment used herein includes preventive and therapeutic treatments.

The present disclosure also encompasses the use of chimeric proteins or polypeptides, antibodies, non-HLA restricted TCRs, CARs, polynucleotides, vectors, immune cells, or compositions comprising the same as defined herein in cell therapy for cancer in subjects in need thereof. Typically, such compositions further comprise a pharmaceutical excipient.

Also contemplated are pharmaceutical compositions comprising any of the foregoing, optionally together with sterile pharmaceutically acceptable excipients, carriers, and/or buffers, and methods of using such pharmaceutical compositions.

In any embodiment as described herein, the cancer treatment product defined above (i.e., a transmembrane chimeric protein or polypeptide, an antibody, a non-HLA restricted TCR, a CAR, a polynucleotide, a carrier, an immune cell, or a composition comprising the same as defined herein) can be administered in combination with at least one other therapeutic agent. This further therapeutic agent can generally be a chemotherapeutic agent or an immunotherapeutic agent, optionally a checkpoint inhibitor.

For example, according to the present disclosure, any cancer treatment product can be administered in combination with an anti-immunosuppressive/immunostimulatory agent. For example, one or more checkpoint inhibitors are further administered to the subject, typically selected from PD-1 inhibitors, PD-L1 inhibitors, Lag-3 inhibitors, Tim-3 inhibitors, TIGIT inhibitors, BTLA inhibitors, V-domain Ig inhibitors of T cell activation (VISTA) inhibitors, and CTLA-4 inhibitors or IDO inhibitors.

Various embodiments of the methods, chimeric proteins or polypeptides, and cancer therapeutic products are described in detail below. Except for the alternatives explicitly mentioned, the present application encompasses combinations of these embodiments.

DETAILED DESCRIPTION OF THE INVENTION

The expression of transposable elements in normal tissues is silenced by DNA methylation established early in embryonic development. An additional layer of inhibition is provided by histone modification. TEs can be reactivated in tumor cells. The inventors have found and provided clear evidence that non-classical alternative splicing events between exons and TEs may be the source of tumor antigens, especially tumor-specific antigens.

The inventors have developed methods for identifying tumor antigens, particularly tumor-specific antigens. Specifically, the inventors have determined methods for identifying tumor antigens derived from junctions (JETs) between TEs and exons. Therefore, in some embodiments, the present invention relates to methods for identifying and selecting tumor neoantigenic peptides encoded by fusion transcript (i.e., chimeric, also referred to herein as junction exon TE-JET) sequences, which fusion transcript sequences comprise a portion of a TE sequence and a portion of an exon sequence.

The present inventors further provide a group of transmembrane tumor chimeric proteins or polypeptides that represent excellent cell surface tumor neoantigen target candidates.

The new antigenic tumor-specific peptides identified by the method according to the present disclosure are highly immunogenic. In fact, since they are derived from fusion transcripts (composed of transposable elements -TEs- and exon sequences) that are absent or expressed at low levels in normal cells, the peptides of the present disclosure are expected to exhibit very low immune tolerance.

The present disclosure also allows the selection of peptides with shared tumor neo-epitopes in a patient population. Such shared tumor peptides have high therapeutic value because they can be used for immunotherapy in large patient populations.

definition

According to the present disclosure, the term "disease" refers to any pathological condition, including cancer diseases, particularly the forms of cancer diseases described herein.

The term "normal" refers to the healthy state or condition of a healthy subject or tissue, ie, a non-pathological condition, wherein "healthy" preferably refers to non-cancerous.

Cancer (medical term: malignancy) is a class of diseases in which a group of cells exhibit uncontrolled growth (dividing beyond normal limits), invasion (invading and destroying neighboring tissues), and sometimes metastasis (spreading to other parts of the body through lymph fluid or blood). These three malignant characteristics of cancer distinguish it from benign tumors, which are self-limited and do not invade or metastasize. Most cancers form tumors, but some, such as leukemia, do not.

Malignant neoplasm is essentially a synonym for cancer. Malignant neoplasm, malignant growth, and malignant tumor are essentially synonyms for cancer.

As used herein, the term "tumor" or "neoplastic disease" refers to an abnormal growth of cells (called neoplastic cells, tumorigenic cells or tumor cells), preferably forming a swelling or lesion. "Tumor cell" refers to an abnormal cell that grows by rapid, uncontrolled cell proliferation and continues to grow after the stimulus that initiated new growth has ceased. Tumors display a partial or complete lack of structural organization and functional coordination with normal tissues, and usually form a distinct mass of tissue, which can be benign, pre-malignant or malignant.

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

A neoplasm is an abnormal mass of tissue caused by a neoplasm. Neoplasm (Greek for new growth) is an abnormal proliferation of cells. The growth of cells exceeds and is out of harmony with the growth of the normal tissue around it. Even after the stimulus stops, the growth continues in the same excessive manner. It usually results in a lump or tumor. Neoplasms can be benign, pre-malignant, or malignant.

"Growth of a tumor" or "tumor growth" according to the present disclosure relates to the tendency of a tumor to increase its size and/or the tendency of tumor cells to proliferate.

For the purposes of this disclosure, the terms "cancer" and "cancer disease" are used interchangeably with the terms "tumor" and "tumor disease."

Cancers are classified based on the type of cells that resemble the tumor and therefore based on the tissue from which the tumor is presumed to have originated. These are histology and location, respectively.

According to the present application, cancer may affect any of the following tissues or organs: breast; liver; kidney; heart, mediastinum, pleura; floor of mouth; lip; salivary glands; tongue; gums; oral cavity; palate; tonsils; larynx; trachea; bronchi, lungs; pharynx, hypopharynx, oropharynx, nasopharynx; esophagus; digestive organs such as stomach, intrahepatic bile duct, bile duct, pancreas, small intestine, colon; rectum; urinary organs such as bladder, gallbladder, ureters; rectosigmoid colon; anus, anal canal; skin; bones; joints, limbs Articular cartilage of the body; eyes and appendages; brain; peripheral nerves, autonomic nervous system; spinal cord, cranial nerves, meninges; and various parts of the central nervous system; connective tissue, subcutaneous tissue and other soft tissues; retroperitoneum, peritoneum; adrenal glands; thyroid gland; endocrine glands and related structures; female reproductive organs, such as ovaries, uterus, cervix; uterine body, vagina, vulva; male reproductive organs, such as penis, testes and prostate; hematopoietic system and reticuloendothelial system; blood; lymph nodes; thymus.

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

The main types of lung cancer are small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC). NSCLC is divided into three main subtypes: squamous cell lung cancer, lung adenocarcinoma (LUAD), and large cell lung cancer. Adenocarcinoma accounts for about 10% of lung cancers. This cancer is usually found in the periphery of the lungs, in contrast to small cell lung cancer and squamous cell lung cancer, which tend to be more centrally located.

"Metastasis" refers to the spread of cancer cells from their original site to another part of the body. The formation of metastasis is a very complex process that depends on the separation of malignant cells from the primary tumor, the invasion of the extracellular matrix, the penetration of the endothelial basement membrane into the body cavity and blood vessels, and then infiltrates into the target organ after being transported by the blood. Finally, the growth of new tumors (i.e., secondary tumors or metastatic tumors) at the target site depends on angiogenesis. Even after the primary tumor is resected, tumor metastasis often occurs because tumor cells or components may remain and develop into metastatic potential. In one embodiment, the term "metastasis" according to the present disclosure relates to "distant metastasis" and refers to metastasis away from the primary tumor and the regional lymph node system.

The cells of a secondary or metastatic tumor are similar to those of the primary tumor. This means, for example, that if ovarian cancer spreads to the liver, the secondary tumor is made up of abnormal ovarian cells rather than abnormal liver cells. The tumor in the liver is called metastatic ovarian cancer, not liver cancer.

Recurrence or regeneration occurs when a person is again affected by a condition that affected them in the past. For example, if a patient suffers from a neoplastic disease, receives successful treatment for the disease and develops the disease again, the newly developed disease may be considered a recurrence or regeneration. However, according to the present disclosure, the recurrence or regeneration of a neoplastic disease may, but does not necessarily, occur at the site of the original neoplastic disease. Thus, for example, if a patient suffers from an ovarian tumor and receives successful treatment, the recurrence or regeneration may be the occurrence of an ovarian tumor or the occurrence of a tumor at a site other than the ovary. The recurrence or regeneration of a tumor also includes situations where the tumor occurs at a site different from the original tumor site as well as the original tumor site. Preferably, the original tumor for which the patient has been treated is a primary tumor, and the tumor located at a site different from the original tumor site is a secondary or metastatic tumor.

"Treatment" refers to administering a compound or composition described herein to a subject to prevent or eliminate a disease, including reducing the size or number of a tumor in a subject; stopping or slowing a disease in a subject; inhibiting or slowing the development of a new disease in a subject; reducing the frequency or severity of symptoms and/or recurrences in a subject currently or previously suffering from a disease; and/or prolonging, i.e., increasing, the life span of a subject. Specifically, the term "treatment of a disease" includes curing, shortening the duration, ameliorating, preventing, slowing or inhibiting the progression or worsening of a disease, or preventing or delaying the onset of a disease or its symptoms.

"At risk" refers to a subject, i.e., a patient, who is identified as having a higher than normal chance of developing a disease, especially cancer, compared to the general population. In addition, a subject who has had or currently has a disease, especially cancer, is a subject at increased risk of developing the disease because the subject may go on to develop the disease. A subject who currently has or has had cancer also has a higher risk of cancer metastasis.

The therapeutically active agents, vaccines and compositions described herein may be administered by any conventional route, including injection or infusion.

The agents described herein are administered in an effective amount. An "effective amount" refers to an amount that, alone or in combination with other doses or other therapeutic agents, achieves a desired response or desired effect. In the case of treating a specific disease or a specific condition, the desired response preferably involves inhibiting the progression of the disease. This includes slowing the progression of the disease, particularly interrupting or reversing the progression of the disease. In the treatment of a disease or condition, the desired response may also be delaying the onset of the disease or condition or preventing the onset of the disease or condition.

The effective amount of an agent described herein will depend on the condition to be treated, the severity of the disease, the individual parameters of the patient (including age, physical condition, size and weight), the duration of treatment, the type of concomitant treatment (if any), the specific route of administration, and similar factors. Thus, the dosage of an agent described herein administered may depend on a variety of such parameters. In the event that a patient does not respond adequately to an initial dose, a higher dose may be used (or a higher effective dose obtained by a different, more localized route of administration).

The pharmaceutical compositions described herein are preferably sterile and contain an effective amount of the therapeutically active substance to produce the desired response or effect.

The pharmaceutical compositions described herein are generally administered in pharmaceutically compatible amounts and pharmaceutically compatible formulations. The term "pharmaceutically compatible" refers to non-toxic materials that do not interact with the effects of the active ingredients of the pharmaceutical composition. Such formulations may generally include salts, buffer substances, preservatives, carriers, supplementary immune enhancing substances (e.g., adjuvants, such as CpG oligonucleotides, cytokines, chemokines, saponins, GM-CSF and/or RNA), and other therapeutically active compounds (if applicable). When used in medicine, salts should be pharmaceutically compatible.

As used herein, the term "nucleic acid molecule" includes any nucleic acid molecule encoding a target polypeptide or a fragment thereof. Such a nucleic acid molecule need not be 100% homologous or identical to an endogenous nucleic acid sequence, but may exhibit substantial identity. A polynucleotide having "substantial identity" or "substantial homology" to an endogenous sequence is generally capable of hybridizing to at least one strand of a double-stranded nucleic acid molecule. "Hybridization" refers to the formation of a pair of double-stranded molecules between complementary polynucleotide sequences (e.g., genes described herein) or portions thereof under various stringent conditions. (See, e.g., Wahl, G.M. and S.L. Berger (1987) Methods Enzymol. 152: 399; Kimmel, A.R. (1987) Methods Enzymol. 152: 507). For example, stringent salt concentrations are typically less than about 750 mM NaCl and 75 mM trisodium citrate, such as less than about 500 mM NaCl and 50 mM trisodium citrate, or less than about 250 mM NaCl and 25 mM trisodium citrate. Low stringency hybridization can be obtained in the absence of an organic solvent (e.g., formamide), while high stringency hybridization can be obtained in the presence of at least about 35% formamide (e.g., at least about 50% formamide). Stringent temperature conditions typically include a temperature of at least about 30° C., at least about 37° C., or at least about 42° C. Those skilled in the art are familiar with different additional parameters, such as hybridization time, the concentration of a detergent (e.g., sodium dodecyl sulfate (SDS)), and the inclusion or exclusion of carrier DNA. Different stringencies are achieved by combining these different conditions as needed. In certain embodiments, hybridization will be carried out at 30° C. in 750 mM NaCl, 75 mM trisodium citrate, and 1% SDS. In certain embodiments, hybridization will be carried out at 37°C in 500 mM NaCl, 50 mM trisodium citrate, 1% SDS, 35% formamide, and 100 pg/ml denatured salmon sperm DNA (ssDNA). In certain embodiments, hybridization will be carried out at 42°C in 250 mM NaCl, 25 mM trisodium citrate, 1% SDS, 50% formamide, and 200 pg/ml ssDNA. Effective variations of these conditions are apparent to those skilled in the art. For most applications, the post-hybridization washing steps also vary in stringency. Washing stringency conditions can be defined by salt concentration and temperature. As described above, washing stringency can be increased by reducing salt concentration or increasing temperature. For example, the stringent salt concentration of the washing step can be less than about 30 mM NaCl and 3 mM trisodium citrate, such as less than about 15 mM NaCl and 1.5 mM trisodium citrate. Stringent temperature conditions for the washing steps typically include a temperature of at least about 25° C., a temperature of at least about 42° C., or a temperature of at least about 68° C. In certain embodiments, the washing steps will be carried out at 25° C. in 30 mM NaCl, 3 mM trisodium citrate, and 0.1% SDS. In certain embodiments, the washing steps will be carried out at 42° C. in 15 mM NaCl, 1.5 mM trisodium citrate, and 0.1% SDS. In certain embodiments, the washing steps will be carried out at 68° C. in 15 mM NaCl, 1.5 mM trisodium citrate, and 0.1% SDS. Hybridization techniques are well known to those skilled in the art and are described, for example, in Benton and Davis (Science 196:180, 1977); Grunstein and Rogness (Proc. Natl. Acad. Sci., USA 72:3961, 1975); Ausubel et al. (Current Protocols in Molecular Biology, Wiley Interscience, New York, 2001); Berger and Kimmel (Guide to Molecular Cloning Techniques, 1987, Academic Press, New York); and Sambrook et al., Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, New York.

"Substantially identical" or "substantially homologous" refers to a polypeptide or nucleic acid molecule that is at least about 50% homologous or identical to a reference amino acid sequence (e.g., any of the amino acid sequences described herein) or nucleic acid sequence (e.g., any of the nucleic acid sequences described herein). In certain embodiments, such a sequence is at least about 60%, at least about 65%, at least about 70%, at least about 75%, at least about 80%, at least about 85%, at least about 90%, at least about 95%, at least about 99%, or at least about 100% homologous or identical to the amino acid or nucleic acid sequence used for comparison.

Sequence identity can be measured using sequence analysis software (e.g., Sequence Analysis Software Package of the Genetics Computer Group, University of Wisconsin Biotechnology Center, 1710 University Avenue, Madison, Wis. 53705, BLAST, BESTFIT, GAP or PILEUP/PRETTYBOX programs). Such software matches identical or similar sequences by assigning degrees of homology to various substitutions, deletions and/or other modifications. Conservative substitutions typically include substitutions within the following groups: glycine, alanine; valine, isoleucine, leucine; aspartic acid, glutamic acid, asparagine, glutamine; serine, threonine; lysine, arginine; phenylalanine, tyrosine. In an exemplary method of determining the degree of identity, the BLAST program can be used, with closely related sequences represented by probability scores between e-3 and e-100.

An "analog" refers to a structurally related polypeptide or nucleic acid molecule that has the function of a reference polypeptide or nucleic acid molecule.

Unless expressly stated or apparent in the context, the term "about" as used herein is to be understood as being within the normal tolerance range of the art, such as within 2 standard deviations of the mean. Approximately can be understood as being within 50%, 45%, 40%, 35%, 30%, 25%, 20%, 15%, 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, 1%, 0.5%, 0.1%, 0.05% or 0.01% of the specified value. Unless expressly stated otherwise in the context, all numerical values provided herein are modified by the term "about".

"Transposable element" as used herein is a repetitive DNA sequence that can be moved from one position to another in the genome by RNA copies produced by reverse transcriptase (class I TE or retrotransposon) or by excision of itself from its original position (class II TE or DNA transposon). Therefore, it includes class I (retrotransposon, including those containing LTR, LINE and SINE) in the endogenous part of the genome (i.e. not from infection) and class II (DNA transposon) in the endogenous part of the genome (i.e. not from infection). This includes autonomous and non-autonomous elements from both classes. According to the present disclosure, TE sequences can, for example, be selected from class I TE, such as retrotransposons including endogenous retroviruses (ERVs), long interspersed nuclear elements (LINEs) and short interspersed nuclear elements (SINEs) and mammalian long terminal repeat transposons (MaLRs), and class II TE, such as DNA transposons in the endogenous part of the genome.

To date, there are more retrotransposons, which have characteristics similar to those of retroviruses such as HIV. Retrotransposons work by reverse transcription of an RNA intermediate replication mechanism. They are generally grouped into three main categories: retrotransposons with long terminal repeats (LTRs) flanking the main body of the retrovirus, which encode reverse transcriptases similar to retroviruses; retrotransposons with long interspersed nuclear elements (LINE, LINE-1, or L1), which encode reverse transcriptases but lack LTRs and are transcribed by RNA polymerase II; and retrotransposons with short interspersed nuclear elements (SINEs), which do not encode reverse transcriptases and are transcribed by RNA polymerase III. DNA transposons have a transposition mechanism that does not involve RNA intermediates. Transposition is catalyzed by several transposases. LTRs include endogenous retroviruses (ERVs), while non-LTR TEs are further subdivided into long interspersed (LINEs) and short interspersed elements (SINEs), non-autonomous transposons mobilized by the LINE integration mechanism. These lineages consist of phylogenetically related families that further branch into multiple subfamilies, each derived from a precursor copy. Over time, the accumulation of mutations leads to divergence in the consensus sequence of each subfamily member. For a review of TE retrotransposons, see Richardson, Sandra R et al. "The Influence of LINE-1 and SINE Retrotransposons on Mammalian Genomes." Microbiology spectrum vol. 3, 2 (2015): MDNA3-0061-2014.

The classical L1 element is approximately 6000 base pairs (bp) in length and consists of two non-overlapping open reading frames (ORFs) flanked by untranslated regions (UTRs) and target site repeats. LINE-1 retrotransposons have been amplified in mammalian genomes for more than 160 million years. In humans, the vast majority of LINE-1 sequences have been amplified since the separation of the ancestral mouse and human lineages approximately 65 to 75 million years ago. Sequence comparisons between individual genomic LINE-1 sequences and consensus sequences derived from modern active LINE-1 can be used to estimate the age of genomic LINE-1 (Khan H, Smit A, Boissinot S; Genome Res. 2006 Jan; 16(1): 78-87). L1 subfamilies are generally divided into old (L1M, AluJ), intermediate (L1P, L1PB, AluS), young (L1HS, L1PA, AluY) and related (HAL, FAM) subfamilies. In humans, the only autonomously active family is long interspersed element-1 (LINE-1 or L1), but a few L1 copies still have retrotransposition capacity, all of which belong to the youngest, human-specific L1HS subfamily.

SVA elements comprise an evolutionarily young, non-autonomous family of retrotransposons that emerged in the primate lineage approximately 25 million years ago (Hancks DC, Kazazian HH Jr, Semin Cancer Biol. 2010 Aug; 20(4): 234-45). The classical SVA element is approximately 2000 bp and has a complex structure consisting of: 1) hexameric CCCTCT repeats; 2) inverted Alu-like repeats; 3) a set of GC-rich variable nucleotide tandem repeats (VNTRs); 4) a SINE-R sequence that shares homology with HERVK-10 (an inactive LTR retrotransposon); and 5) a classical polyadenylation cleavage specificity factor (CPSF) binding site followed by a poly(A) tail. The youngest SVA subfamilies include the SVA-D, SVA-E, SVA-F, and SVA-F1 subfamilies.

"Messenger RNA (mRNA)" is a single-stranded RNA molecule that corresponds to the genetic sequence of a gene and is read by the ribosome in the process of producing proteins. mRNA is produced during transcription, during which RNA polymerase converts the gene into an initial transcript mRNA (also called pre-timRNA). This pre-mRNA usually still contains introns, which are regions that do not go on to encode the final amino acid sequence. They are removed during the RNA splicing process, leaving only the exons, those regions that encode proteins. This exon sequence constitutes the mature mRNA. The mature mRNA is then read by the ribosome, and using the amino acids carried by the transfer RNA (tRNA), the ribosome produces a peptide sequence. This process is called translation.

As used herein, a "transcript" is a messenger RNA (or mRNA) or a portion of an mRNA that is expressed by an organism, particularly in a specific tissue or even in a specific tissue. The expression of a transcript varies depending on a variety of factors. The expression of a transcript in cancer cells may be altered compared to normal healthy cells.

" Transcriptome " referred to herein is the messenger RNA or mRNA molecule of the complete group expressed or transcribed by cell gene.In some embodiments, the term " transcriptome " can also be used to describe a series of mRNA transcripts produced in a specific cell (or tissue type).Compared with the genome characterized by its stability, the transcriptome changes actively.In fact, the transcriptome of an organism depends on a variety of factors and varies, including developmental stage, environment and physiological conditions.Usually, compared with corresponding normal healthy cells, the transcriptome in cancer cells has also changed.Usually, the transcriptome referred to herein is the human transcriptome.The terms " transcriptome pattern " and " transcriptome " are used as synonyms in this article.

A reading frame is the way in which the nucleotide sequence in a nucleic acid (DNA or RNA) molecule is divided into a set of consecutive, non-overlapping triplets.

An open reading frame (ORF) is the portion of a reading frame that can be translated into a peptide. An ORF is a continuous string of codons that includes a start codon (e.g., AUG) and a stop codon (e.g., UAA, UAG, or UGA) located at the start site of transcription. The ATG codon in the ORF (not necessarily the first) can indicate the start position of translation. The transcription termination site is located after the ORF, outside the translation stop codon. In eukaryotic genes with multiple exons, the ORF spans the intron/exon region, which can be spliced together after the ORF is transcribed to produce the final mRNA for protein translation.

"Classic ORF" referred to herein is a protein coding sequence with a specific reading frame in an mRNA sequence described or annotated in a database (e.g., Ensembl genome/transcriptome/proteome database collection, usually HG19). Typically, a classic ORF is identical to one of the exons of a normal healthy cell.

"Non-classical ORF" as referred to herein is a protein coding sequence that is not described (i.e., not annotated) in a genome database (e.g., the Ensembl genome/transcriptome/proteome database) and has a specific reading frame in an mRNA sequence. Therefore, in general, non-classical ORF means that the reading frame is frameshifted compared to the conventional reading frame of the exon in a normal healthy cell. However, in some embodiments, non-classical can be described in a genome database (e.g., the Ensembl database), but the mRNA sequence represents a minor type in a normal cell. With respect to minor types, it is generally less than 5%, particularly less than 2%, or preferably less than 1% in a normal cell.

Exon is any part of a gene that will encode a portion of the final mature RNA produced by the gene after the introns are removed by RNA splicing. The term "exon" refers to a DNA sequence within a gene and the corresponding sequence in the RNA transcript. In RNA splicing, introns are removed and exons are covalently bound to each other as part of producing mature messenger RNA. An exon sequence according to the present application comprises at least a portion of one or more exons. Typically, an exon sequence comprises at least a portion of one or two exons.

The untranslated sequences at the 3' and 5' ends (3'UTR and 5'UTR) present in the mature RNA after splicing are exonic sequences, but they are non-coding sequences because these sequences are located upstream of the translation start codon (5'UTR) or downstream of the translation stop codon (3'UTR).

In the present application, the terms "fusion transcript", "chimeric transcript", "TE-exon transcript" or "joined exon TE" (JET) are used indiscriminately as synonyms. According to the present disclosure, a "fusion or chimeric""transcript or sequence" is defined as a transcript that is partially aligned with an exon sequence and partially aligned with a transposable element (TE) sequence. Fusion or chimeric transcripts are also referred to herein as JETs (junctions between exons and TEs). Typically, fusion transcripts according to the present specification have a normalized number of reads greater than ^2.10-6 . The normalized number of reads is defined as the number of reads covering the fusion divided by the sample library size.

The term "polypeptide" is used in this specification to refer to a series of residues, usually L-amino acids, which are usually connected to each other by peptide bonds between the α-amino and carboxyl groups of adjacent amino acids. Polypeptides or peptides can be of various lengths, in neutral (uncharged) form or in salt form, and are free of modifications such as glycosylation, side chain oxidation or phosphorylation, or contain these modifications, provided that such modifications do not destroy the biological activity of the polypeptides described herein. Unless the term "polypeptide" is explicitly mentioned, peptides and proteins refer interchangeably to JET-derived neoantigenic peptides, polypeptides or proteins as described herein.

As used herein, pJET is a peptide or polypeptide derived from (ie, encoded by) a chimeric/fusion transcript or JET. pJET is also referred to herein as translated JET.

A "reference genome" or "representative genome" is a digital nucleic acid sequence database assembled by scientists that serves as a representative example of a collection of genotypes. Because a reference genome is usually assembled by sequencing DNA from multiple donors, it cannot accurately represent the collection of genes of any single individual (animal or human). Instead, the reference provides a haploid mosaic of different DNA sequences from each donor.

RNA-Seq (short for RNA sequencing) is a sequencing technology that uses next-generation sequencing (NGS) to reveal the presence and quantity of RNA (usually messenger RNA, mRNA) in biological samples and produces a large number of raw sequencing reads (usually at least tens of millions). Single-cell RNA sequencing (scRNA-Seq) provides expression profiles of individual cells. Reads refer to the RNA sequence of RNA fragments from a biological sample or a single cell. The sequenced RNA sample is called an RNA library. Therefore, RNA sequencing data is often referred to as RNA reads.

In the present application, "MHC molecule" or "HLA molecule" refers to at least one MHC/HLA class I molecule or at least one MHC/HLA class II molecule. MHC class I proteins form functional receptors on most nucleated cells in the body. There are three major MHC class I genes in HLA: HLA-A, HLA-B, HLA-C, and three minor genes HLA-E, HLA-F and HLA-G. 32-microglobulin binds to the major and minor gene subunits to produce heterodimers. Class I MHC molecules are composed of heavy chains and light chains, and can bind peptides of about 8-11 amino acids, but if the peptide has a suitable binding motif, it is usually 8 or 9 amino acids, and presents it to cytotoxic T-lymphocytes. The binding of the peptide is stabilized at both ends by contact between atoms in the peptide backbone and constant sites in the peptide binding groove of all MHC class I molecules. There are constant sites at both ends of the groove that binds the amino and carboxyl ends of the peptide. Variations in peptide length are accommodated by kinking of the peptide backbone, typically at proline or glycine residues that allow the required flexibility. Peptides that bind to class I MHC molecules are typically derived from endogenous protein antigens. For example, in humans, the heavy chains of class I MHC molecules are typically HLA-A, HLA-B, or HLA-C monomers, and the light chains are beta-2-microglobulin. There are three major and two minor MHC class II proteins encoded by HLA. Class II genes bind to form heterodimeric (αβ) protein receptors that are typically expressed on the surface of antigen presenting cells. Peptides that are bound by class II MHC molecules are typically derived from extracellular or exogenous protein antigens. For example, in humans, the α and β chains are particularly HLA-DR, HLA-DQ, and HLA-DP monomers. MHC class II molecules are able to bind peptides of about 8-20 amino acids, particularly 10-25 amino acids or 13-25 amino acids (if the peptide has a suitable binding motif), and present them to helper T cells. The peptide is in an extended conformation along the MHC class I peptide binding groove, which (unlike the MHC class I peptide binding groove) is open at both ends. It is primarily held in place by contacts between backbone atoms and conserved residues lining the peptide binding groove.

The term "peptide group" refers to the complete set of peptides expressed by a specific genome or present in a specific organism or cell type (e.g., cancer cells). Thus, proteomic analysis (proteomics) refers to the separation, identification, and quantification of the complete set of peptides or proteins expressed by a genome, cell, or tissue at a specific time point.

Proteomic analysis is usually based on two main techniques, namely two-dimensional gel electrophoresis (2-DGE) (Harper S et al., In: Coligan JE, Dunn BM, Speicher DW, Wing-field PT, editors. Current Protocols in Protein Science. John Wiley &Sons; Hoboken, N.J.: 1998. pp. 10.4.1-10.4.36.) and mass spectrometry (MS) (Aebersold & Mann, 2003), which are powerful methods for analyzing complex mixtures of proteins. HPLC is an alternative separation technique for proteomic research, especially in the separation and identification of low molecular weight proteins and peptides (Garbis et al., 2005). MS allows the molecular mass of a protein or peptide to be determined based on the mass-to-charge ratio (m/z) of ions in the gas phase. The terms "gel-based" or "gel-free" proteomics are used in relation to the separation technique applied, 2-DGE or HPLC; proteomics approaches may also be "bottom-up" or "top-down," essentially identifying proteins as a whole from their protease (e.g., trypsin) digests or by mass spectrometry, respectively.

Bottom-up proteomics is a common approach to identify proteins from biological samples (tissues or cells) and characterize their amino acid sequence and post-translational modifications by proteolytic digestion of proteins prior to mass spectrometric analysis. Crude protein extracts are enzymatically digested and then separated into peptides in one or more dimensions, usually by liquid chromatography coupled to mass spectrometry (a technique known as shotgun proteomics). Peptides can be identified by comparing the masses of proteolytic peptides or their tandem mass spectra with the masses predicted from annotated peptide spectra from sequence databases or peptide spectral libraries, and multiple peptide identifications can be assembled into a protein identification.

In top-down proteomics, intact proteins are purified in a mass spectrometer or by 2D electrophoresis prior to digestion and/or fragmentation. Top-down proteomics uses an ion trap mass spectrometer to store isolated protein ions for mass measurement and tandem mass spectrometry (MS/MS) analysis, or other protein purification methods such as two-dimensional gel electrophoresis coupled to MS/MS.

Based on the data generated by MS, the protein is sequenced de novo by manual mass spectrometry analysis or automatically processed by sequence search engines such as SEQUEST, Mascot, Phenyx, X! Tandem and OMSSA. These algorithms are developed based on the correlation between experimental and theoretical MS/MS data; the latter are generated by computer-simulated enzyme digestion of protein databases such as UniProt/Swiss-Prot (Deutsch, Lam, & Aebersold, 2008).

The term "immunopeptidome", also often referred to as "immunopeptidome pattern", "pMHC repertoire" or "MHC-ligandome" or "HLA ligandome", refers to a complete set of peptides within a specific cell type that bind to at least one MHC/HLA molecule on the cell surface. Accordingly, "immunopeptidomics" has become a term to describe the analysis of MHC/HLA-ligandomes. The most common immunopeptidomics methods rely on mass spectrometry (MS). Immunopeptidomics samples are typically prepared from MHC isolated from lysed cells or tissues, for example by using allele-specific antibodies, pan-specific antibodies, or engineered affinity tag systems. The separated complexes are acid eluted, and the peptides are purified from the MHC molecules using molecular weight cutoff (MWCO), solid phase extraction or other techniques, and subsequently analyzed by MS (see, for example, review L.E.Stopfer et al., Immuno-Oncology and Technology, Volume 11, 2021, 100042).

As used herein, the term "antibody" refers not only to complete antibody molecules, but also to antibody molecule fragments that retain the ability to bind to an immunogen. Such fragments are also well known in the art and are often used in vitro and in vivo. Therefore, as used herein, the term "antibody" refers not only to complete immunoglobulin molecules, but also to the well-known active fragments F(ab') ₂ and Fab.

F(ab') ₂ and Fab fragments lacking the Fc fragment of an intact antibody are cleared from the circulation more rapidly and may have less nonspecific tissue binding than intact antibodies (Wahl et ah, J. Nucl. Med. 24:316-325 (1983)).

As used herein, antibodies include all-natural antibodies, bispecific antibodies; chimeric antibodies; Fab, Fab', single-chain V region fragment (scFv), fusion polypeptides and unconventional antibodies.

In certain embodiments, antibodies are glycoproteins comprising at least two heavy chains (H) and two light chains (L) interconnected by disulfide bonds. Each heavy chain consists of a heavy chain variable region (abbreviated herein as _VH ) and a heavy chain constant region ( _CH ). The heavy chain constant region consists of three domains, CH1, CH2, and CH3. Each light chain consists of a light chain variable region (abbreviated herein as _VL ) and a light chain constant region _CL . The light chain constant region consists of one domain _CL . The _VH and _VL regions can be further subdivided into hypervariable regions, called complementarity determining regions (CDRs), which intersect with more conservative regions called framework regions (FRs). Each _VH and _VL consists of three CDRs and four FRs, arranged in the following order from amino terminus to carboxyl terminus: FR1, CDR1, FR2, CDR2, FR3, CDR3, FR4.

The variable regions of the heavy and light chains contain the binding domain that interacts with the antigen. The constant regions of the antibodies mediate the binding of the immunoglobulin to host tissues or factors, including various cells of the immune system (e.g., effector cells) and the first component (C1q) of the classical complement system.

As used herein, "CDR" is defined as the complementary determining region amino acid sequence of an antibody, which is a hypervariable region of an immunoglobulin heavy chain and light chain (see, e.g., Rabat et al., Sequences of Proteins of Immunological Interest, 4th U.S. Department of Health and Human Services, National Institutes of Health (1987)). Typically, an antibody comprises three heavy chain and three light chain CDRs or CDR regions in the variable region. The CDRs provide most of the contact residues for the binding of an antibody to an antigen or epitope. In certain embodiments, the CDR regions are described using the Rabat system (Rabat, E.A., et al. (1991) Sequences of Proteins of Immunological Interest, Fifth Edition, ET.S. Department of Health and Human Services, NIH Publication no. 91-3242).

As used herein, the term "single-chain variable fragment" or "scFv" is a fusion protein of the variable regions of the heavy chain ( _VH ) and light chain ( _VL ) of an immunoglobulin, which are covalently linked to form a _VH :: _VL heterodimer. _VH and _VL are linked directly or through a linker (e.g., 10, 15, 20, 25 amino acids) encoding a peptide that connects the N-terminus of _VH to the C-terminus of _VL , or connects the C-terminus of _VH to the N-terminus of _VL . The linker is usually rich in glycine for flexibility, and rich in serine or threonine for solubility. Despite the removal of the constant region and the introduction of the linker, the scFv protein retains the specificity of the original immunoglobulin. Single-chain Fv polypeptide antibodies can be expressed from nucleic acids comprising sequences encoding _VH and _VL , such as Huston, et al. (Proc. Nat. Acad. Sci. USA, 85: 5879-5883, 1988). See also U.S. Patent Nos. 5,091,513; 5,132,405; and 4,956,778; and U.S. Patent Publication Nos. 20050196754 and 20050196754.

As used herein, the term "affinity" refers to a measure of binding strength. Affinity depends on the closeness of the stereochemical matching between the antibody binding site and the antigenic determinant, the size of the contact area between them and/or the distribution of charged and hydrophobic groups. As used herein, the term "affinity" also includes "avidity", which refers to the strength of the antigen-antibody bond after forming a reversible complex. The method for calculating the affinity of an antibody to an antigen is known in the art, including but not limited to various antigen binding experiments, such as functional assays (such as flow cytometry assays), surface plasmon resonance assays (such as BIACORE assays) and kinetic exclusion assays (such as KINEXA assays).

The term "chimeric antigen receptor" or "CAR" as used herein refers to a molecule comprising an extracellular antigen binding domain and a transmembrane domain fused to an intracellular signaling domain capable of activating or stimulating an immune response cell. In certain embodiments, the extracellular antigen binding domain of CAR comprises scFv. ScFv may be derived from fusing the heavy chain variable region and the light chain variable region of an antibody. As an alternative or in addition, scFv may be derived from Fab's (rather than antibodies, such as obtained from a Fab library). In certain embodiments, scFv is fused to a transmembrane domain and then fused to an intracellular signaling domain. In certain embodiments, CAR has a high binding affinity or avidity for an antigen.

As used herein, the term "antigen binding domain" refers to a domain that is capable of specifically binding to a particular antigenic determinant or a group of antigenic determinants present on a cell.

The term "immune cells" herein generally includes T cells, natural killer T cells, CD4+/CD8+T cells, TIL/tumor-derived CD8T cells, central memory CD8+T cells, Treg, MAIT, YδT cells, human embryonic stem cells and pluripotent stem cells that can differentiate into lymphocytes.

An "isolated cell" refers to a cell that is separated from molecules and/or cellular components that are naturally associated with the cell.

The terms "isolated", "purified" or "biologically pure" refer to materials that are free to varying degrees from components that normally accompany them in their native state. "Isolated" means a degree of separation from the original source or environment. "Purified" means a degree of separation greater than separation. A "purified" or "biologically pure" protein is sufficiently free of other substances so that any impurities do not seriously affect the biological properties of the protein or cause other adverse consequences. That is, if a nucleic acid or peptide produced by recombinant DNA technology is substantially free of cellular material, viral material or culture medium, or a nucleic acid or peptide synthesized by chemical synthesis is substantially free of chemical precursors or other chemicals, the nucleic acid or peptide is purified. Purity and homogeneity are generally determined using analytical chemistry techniques, such as polyacrylamide gel electrophoresis or high performance liquid chromatography. The term "purified" can mean that a nucleic acid or protein produces essentially one band in an electrophoretic gel. For proteins that can be modified (e.g., phosphorylated or glycosylated), different modifications can produce different isolated proteins, which can be purified separately.

Method for selecting tumor neoantigen peptides

The method for selecting tumor neoantigen peptides according to the present disclosure comprises

- a step of identifying a fusion transcript (or JET) sequence from mRNA sequences in a cancer cell sample of a subject, wherein the fusion transcript (JET) sequence comprises a transposable element (TE) sequence and an exon sequence and includes an open reading frame (ORF), and

- a step of selecting a tumor neoantigenic peptide of at least 8 amino acids encoded by a portion of said ORF of the fusion transcript sequence,

wherein the ORF overlaps with the junction between the TE and exon sequence, is pure TE and/or non-canonical, and

Wherein the tumor neoantigen peptide binds to at least one major histocompatibility complex (MHC) molecule of the subject.

Typically, peptides translated from a portion of a non-canonical ORF in an exonic sequence are recognized by the immune system as non-self.

In some embodiments, the exonic sequence is from an oncogene and/or a tumor suppressor gene and/or from one of its mutant variants.

Conceptually, cancer is the result of the accumulation of continuous somatic mutations. Many studies have shown that gain of function of oncogenes and loss of function of tumor suppressor genes are required for normal cells to develop cancer. For diploid organisms, gain-of-function mutations are usually dominant or semi-dominant, while loss-of-function mutations are usually recessive. The two-mutation hypothesis of tumorigenesis holds that the development of cancer is caused by the loss of both alleles of tumor suppressor genes.

Oncogenes (also called cancer genes) are genes whose action actively promotes cell proliferation or growth. Normal non-mutated types are called proto-oncogenes. The mutant type has excessive or inappropriate activity, leading to tumor growth. Oncogenes can be identified in the Cancer Gene Marker Database (CGMD) (Pradeepkiran, J., Sainath, S., Kramthi Kumar, K. et al. CGMD:. Sci Rep 5, 12035 (2015) "An integrated database of cancer genes and markers"). Oncogenes (ONC) can also be downloaded from the Cancer Gene Network Database (NCG 5.0) (An O, Dall'OlioGM, Mourikis TP, Ciccarelli FD, Nucleic Acids Res. 2016 Jan 4; 44 (D1): D992-9; "NCG5.0: updates of a manually curated repository of cancer genes and associated properties from cancer mutational screenings"). Non-limiting examples of oncogenes include: L-MYC, LYL-1, LYT-10, LYT-10/Cα1, MAS, MDM-2, MLL, MOS, MTG8/AML1, MYB, MYH11/CBFB, NEU, N-MYC, OST, PAX-5, PBX1/E2A, PIM-1, PRAD-1, RAF, RAR/PML, RAS-H, RAS-K, RAS-N, REL/NRG, RET, RHOM1, RHOM2, ROS, SKI, SIS, SET/CAN, SRC, TAL1, TAL2, TAN-1, TIAMI, TSC2, and TRK.

Tumor suppressor genes (also called tumor suppressor genes) represent the other side of cell growth control, usually acting to inhibit cell proliferation and tumor development. Therefore, tumor suppressor genes are generally genes that inhibit cell division or growth. Loss of TSG function promotes uncontrolled cell division and tumor growth. Rb (a tumor suppressor gene), identified by genetic analysis of retinoblastoma encoding a transcriptional regulatory protein, serves as a prototype for identifying other tumor suppressor genes that lead to the development of a variety of different human cancers. Tumor suppressor genes are described in particular in "Cooper GM. The Cell: A Molecular Approach. 2nd edition. Sunderland (MA): Sinauer Associates; 2000. Tumor Suppressor Genes". Tumor suppressor genes (TSGs) can also be downloaded from the tumor suppressor gene database (TSGene 2.0) (see reference Zhao M, Kim P, Mitra R, Zhao J, Zhao Z; Nucleic Acids Res. 2016 Jan 4; 44(D1): D1023-31; "TSGene 2.0: an updated literature-based knowledgebase for tumor suppressor genes"). In the present context, non-limiting examples of tumor suppressor genes include: APC, BRCA1, BRCA2, DPC4, INK4, MADR2, NF1, NF2, p53, PTC, PTEN, Rb, RB1, VHL, WT1, BUB1, bub 1, TGF-βRII, Axin, DPC4, p300, PPARγ, p16, DPC4, PTEN and hSNF5.

Oncogenes, tumor suppressor genes or "double-factor" genes (having both oncogenic and tumor suppressor functions) can be systematically identified by database searching and text mining. In fact, information about oncogenes or tumor suppressor genes can usually be found in the Ensembl database (but see also Shen L, Shi Q, Wang W. Double agents: genes with both oncogenic and tumor-suppressor functions. Oncogenesis. 2018; 7(3): 25. Published 2018 Mar 13). Double-factor genes can be identified as genes that overlap between the above two databases (see also Shen et al., Oncogenesis 2018, above).

Without being bound by any theory, the inventors believe that choosing fusions in which the exon sequences are from oncogenes and/or tumor suppressor genes is highly relevant for the following reasons:

TE insertions in oncogenes can alter their oncogenic activity. Therefore, insertion of TE sequences into oncogene active domains may result in constitutive activity of oncogenes, similar to driver mutations. Therefore, chimeric oncogenes generated by these fusions may represent a new family of oncoproteins. If this is the case, targeting the activity of these new “fusion oncogenes” with small molecule antagonists may represent a potential therapeutic approach for cancers expressing these chimeric oncogenes.

Insertion of TEs into tumor suppressors can inactivate their suppressive function, often leading to loss of function (e.g., by introduction of stop codons, ORF changes, or disruptive amino acid stretches), thereby contributing to the oncogenic process.

Fusions involving cancer driver genes would be excellent targets for adoptive cell therapy, antibodies, ADCs, T-cell binders, etc. If the fusion oncogenes are involved in tumorigenesis, they would be expected to be more specific to cancer cells, thereby reducing the development of resistance (due to the oncogenic activity of the target).

In some embodiments, the TE sequence is located at the 5' end of the fusion transcript sequence (TE sequence is also called donor sequence), and the exon sequence is located at the 3' end of the fusion transcript sequence relative to the junction (exon sequence is therefore called acceptor sequence). The expression "located at the 5' end of the fusion transcript sequence" means that the element is located upstream of the junction of the fusion transcript sequence. The expression "located at the 3' end of the fusion transcript sequence" means that the element is located downstream of the junction of the fusion transcript sequence.

In a more specific embodiment, the TE sequence is located at the 5' end of the fusion transcript sequence, and the exon sequence is located at the 3' end of the fusion transcript sequence, and a portion of the ORF of the fusion transcript sequence encoding the neoantigenic peptide overlaps with the junction. In this case, the ORF can be classical or non-classical. It should be understood that the ORF may include a junction, but the neoantigenic peptide sequence may not be derived from the junction. In some embodiments, when the neoantigenic peptide sequence includes a sequence derived from a junction, the peptide obtained is therefore encoded by the TE sequence and the exon sequence.

The expression "a portion of the ORF is overlapping or overlaps with the junction between the TE sequence and the exon sequence" means that the junction is contained in a portion of the ORF of the fusion transcript sequence, which encodes the neo-antigenic peptide.

In embodiments where (i) a portion of the ORF encoding the neo-antigenic peptide overlaps with the junction between the TE sequence and the exon sequence, and (ii) the TE sequence and the exon sequence are located at the 5' end and 3' end of the fusion transcript sequence, respectively, the portion of the ORF generally encodes a neo-antigenic peptide of at least 8 amino acids, including at least 1-6 amino acids, particularly 2-6, from the TE sequence, and at least 1-6, particularly 2-6 amino acids, from the exon sequence.

In another embodiment in which the TE sequence is located at the 5' end of the fusion transcript sequence and the exon sequence is located at the 3' end of the fusion transcript sequence, a portion of the ORF encoding the neoantigenic peptide is located downstream of the junction, so that the ORF is non-classical.

The expression “a portion of the ORF is located downstream of the connection” means that a portion of the ORF encoding the neo-antigenic peptide does not overlap with the connection, but is included in the 3’ end of the fusion transcript sequence relative to the connection. In the present embodiment, since the 3’ end portion relative to the connection is an exon sequence, a portion of the ORF encoding the neo-antigenic peptide is contained in the exon sequence. Therefore, since a portion of the ORF is located only in the exon sequence, the obtained peptide is encoded by the exon sequence in the non-classical ORF. Therefore, in a specific embodiment in which the exon sequence is located at the 3’ end of the fusion transcript sequence relative to the connection, and in which a portion of the ORF encoding the neo-antigenic peptide is located downstream of the connection having a non-classical reading frame, a portion of the ORF of the fusion transcript sequence encodes a neo-antigenic peptide comprising 0 amino acids from the TE sequence and at least 8 amino acids from the exon sequence.

In another embodiment, the TE sequence is located 3' to the fusion transcript sequence and the exon sequence is located 5' to the fusion transcript sequence relative to the junction.

In some embodiments, the TE sequence is located at the 3' end of the fusion transcript sequence, and the exon sequence is located at the 5' end of the fusion transcript sequence, and a portion of the ORF of the fusion transcript sequence encoding the new antigenic peptide overlaps with the connection between the TE sequence and the exon sequence. In this case, the ORF can also be classical or non-classical. The obtained peptide is encoded by the TE sequence and the exon sequence.

In a specific embodiment in which a portion of the ORF encoding the neo-antigenic peptide overlaps with the junction between the exon sequence and the TE sequence, and in which the exon sequence and the TE sequence are located at the 5' end and 3' end of the fusion transcript sequence, respectively, the portion of the ORF encodes a neo-antigenic peptide of at least 8 amino acids, including at least 1 to 6, particularly 2 to 6 amino acids from the TE sequence and at least 1 to 6, particularly 2 to 6 amino acids from the exon sequence.

In another embodiment, the TE sequence is located at the 3' end of the fusion transcript sequence, the exon sequence is located at the 5' end of the fusion transcript sequence, and a portion of the ORF encoding the new antigenic peptide is located downstream of the connection between the exon sequence and the TE sequence. Optionally, the peptide sequence encoded by the pure TE sequence is non-standard.

In this embodiment, since the 3' end relative to the connection is divided into a TE sequence, a portion of the ORF encoding the new antigenic peptide is encoded by the TE sequence. Therefore, the new antigenic peptide encoded by the portion of the ORF does not include any amino acids from the exon sequence and includes at least 8 amino acids from the TE sequence. In a specific embodiment in which the TE sequence is located at the 3' end of the fusion transcript sequence relative to the connection, and a portion of the ORF encoding the new antigenic peptide is located downstream of the connection, the new antigenic peptide encoded by the portion of the ORF of the fusion transcript sequence includes 0 amino acids from the exon sequence and at least 8 amino acids from the TE sequence.

Tumor neoantigen peptides are peptides generated by somatic alterations (classical mutations within the DNA sequence) that are recognized as different from self and presented by antigen presenting cells (APCs) such as dendritic cells (DCs) and tumor cells themselves. Cross-presentation plays an important role because APCs are able to transfer exogenous antigens from phagosomes to the cytosol for proteolytic cleavage into major histocompatibility complex I (MHC I) epitopes by the proteasome.

In the present disclosure, the change corresponds to the transcription of a fusion mRNA sequence comprising a transposable element (TE) sequence and an exon sequence. This may be caused by somatic transposition (i.e., particularly in tumor clones). It may also not be caused by de novo transposition, but by tumor-specific transcriptional derepression, so that TE and nearby genes are co-transcribed.

The new antigenic peptides according to the present disclosure may be completely absent in normal healthy samples (i.e., not expressed in normal healthy samples), and therefore specific to tumor samples. Alternatively, they may be expressed at low levels in normal cells and/or expressed disproportionately in tumor samples compared to normal (healthy) samples.

It may also be selectively expressed by the cell lineages from which cancer develops.

Cancer or tumor samples according to the present disclosure can be isolated from any solid tumor or non-solid tumor of any tissue or organ defined above, such as breast cancer, lung cancer and/or melanoma. In some embodiments, the cancer sample is from acute myeloid leukemia, adrenocortical carcinoma, bladder urothelial carcinoma, breast ductal carcinoma, breast lobular carcinoma, cervical cancer, bile duct carcinoma, colorectal adenocarcinoma, esophageal cancer, gastric adenocarcinoma, glioblastoma multiforme, head and neck squamous cell carcinoma, hepatocellular carcinoma, renal chromophobe cell carcinoma, renal clear cell carcinoma, renal papillary cell carcinoma, low-grade glioma, lung adenocarcinoma, lung squamous cell carcinoma, mesothelioma, ovarian serous adenocarcinoma, pancreatic ductal adenocarcinoma, paraganglioma and pheochromocytoma, prostate cancer, sarcoma, skin melanoma, testicular germ cell carcinoma, thymoma, thyroid papillary carcinoma, uterine carcinosarcoma, uterine body endometrioid carcinoma or uveal melanoma sample. In a specific embodiment, the cancer sample is from a lung cancer sample, particularly a LUAD sample.

Generally, according to the present disclosure, the step of identifying the fusion transcript sequence is performed by aligning the mRNA sequence from the cancer sample with a reference genome and then distinguishing between normal and abnormal (non-annotated or non-canonical based on database information) junctions.

According to the present disclosure, normal ligation generally corresponds to ligation in which the donor and acceptor are on the same strand and not too far apart (eg, not on different chromosomes).

According to the present disclosure, aberrant junctions typically correspond to junctions between donor and acceptor sequences on different chromosomes, or junctions in cis (same chromosome) but on different strands (regardless of order and 5'-3' sense).

According to the present disclosure, the mRNA sequence that can be generally used is RNA seq data (as shown in the results herein). RNA seq data is generally obtained from purified RNA obtained from a cell or tissue sample, fragmented and reversely transcribed into cDNA. The obtained cDNA is then amplified and sequenced (next generation sequencing-NGS) on a high-throughput platform (e.g., Illumina GA/HiSeq-see http://www.illumina.com, SOLiD or Roche454). This process generates millions of short reads from one end of the cDNA fragment. A common variant of this process is to generate reads from both ends of each cDNA fragment, referred to as "double-ended" reads.

In some embodiments, adaptive software can be used, such as: alignment of spliced transcripts to a reference (i.e., STA-see RDobin, Alexander et al. "STAR: ultrafast universal RNA-seq aligner." Bioinformatics (Oxford, England) vol. 29, 1 (2013): 15-21), TopHat2 (Kim, Daehwan et al. "TopHat2: accurate alignment of transcriptomes in the presence of insertions, deletions and gene fusions." Genome biology vol. 14, 4 R36. 25Apr. 2013, doi: 10.1186/gb-2013-14-4-r36) or HISAT (Kim, Daehwan et al. "HISAT: a fast spliced aligner with low memory requirements." Nature methods vol.12, 4(2015):357-60.doi:10.1038/nmeth.3317) aligns the mRNA sequence against the corresponding reference genome (e.g., human reference genome Hg19 ENSEMBL (RNA sequence, GRCh37)). STAR is an independent software that uses a sequential maximum alignable seed search, followed by seed clustering and splicing to align RNA sequence reads. It can detect classical connections, non-classical splicing, and fusion/chimeric transcripts. Generally, connection detection can be performed according to the detailed instructions in the results, based on the definitions obtained from the ENSEMBL and RepeatMasker databases (downloaded from the UCSC genome browser), respectively. Therefore, in some embodiments, a dedicated database (e.g., Ensembl and Repeatmasker databases) is used to determine normal and abnormal connections by computer simulation, and fusion transcripts with connections between TE and exon sequences are extracted by computer simulation.

More specifically, in some embodiments, the RNAseq reads of the target sample (or cell) are aligned to a reference genome (e.g., a classic hg19 genome) using the classic STAR two-pass mode 27 to identify unannotated connections. As previously described, JETs are identified as connections between exons (especially coding DNA sequences-CD-exons) and TEs (or repeat elements REs). According to the present disclosure, TEs (or REs) can be identified (i.e., filtered) according to definitions of commonly used databases in the art (e.g., ENSEMBL (GRCh37) and RepeatMasker).

According to the present invention, mRNA sequences can be from all types of cancer cells or tumor cell samples. Tumors can be solid or non-solid tumors. Specifically, mRNA gene sequences are from any tissue or organ affected by previously defined cancer or tumors, such as breast cancer, lung cancer and/or melanoma. In a specific embodiment, mRNA sequences are from LUAD samples. Tumor samples can be obtained from, for example, the Cancer Genome Atlas (TCGA). In some embodiments, the mRNA sequence is obtained from a cell line, such as a tumor cell line from the Cancer Cell Line Encyclopedia (CCLE).

In some embodiments, the number of spliced reads can be normalized by the number of uniquely aligned reads.JETs with expression levels exceeding ^2.10-7 are typically selected.

In some embodiments of the present disclosure, the fusion transcript sequence is shared by more than 1%; notably, more than 5%, more than 10%, more than 15%, more than 20% or even more than 25% of cancer samples (typically obtained from various patients, such as cancer samples collected for a given cancer type in TCGA) and/or cell lines. In some embodiments, the fusion transcript sequence according to the present disclosure is shared in cancer samples by more than 1%; in particular, more than 2%, more than 5%, more than 10%, more than 15%, more than 20% or even more than 25% of cancer patients. Therefore, the fusion transcript sequence can be specific to a cancer type shared between several cancers.

According to the present disclosure, compared with normal healthy cells, the fusion transcript sequence is expressed at a higher level in tumor cells. In some embodiments, the fusion transcript sequence is expressed in cancer cells (obtained from one or more cancer samples or one or more cell lines), but not in healthy cells (from one or more tissue samples or one or more cell lines), especially not in thymic healthy cells. In some embodiments, when the expression level of JET is lower than ^2.10-7 , especially lower than ^2.10-8 and usually undetectable, JET is considered not to be expressed in cells. According to the present disclosure, such fusion transcripts may be referred to as tumor-specific fusions. According to the present invention, fusion transcripts expressed at higher levels in tumor cells compared with normal cells as defined above, fusion transcripts that are usually expressed disproportionately in cancer cells compared with normal cells may be referred to as tumor-associated fusion transcripts (TAF). According to the present application, if the tumor-associated fusion transcript is present in more than 1%, especially more than 2%, more than 5%, especially more than 10% of the tumor samples (from the same or different tumor types, especially obtained from the TCGA database, preferably for the same cancer type), and is present in less than 20% of the normal samples, then the tumor-associated fusion transcript may be selected. Alternatively or additionally, the fusion transcript sequence may be expressed in at least 1; 2; 3; 4; 5; 6; 7; 8; 9; 10; 11; 12; 13; 14; 15; 16; 17; 18; 19; 20 cell lines.

In some embodiments, the method further comprises the step of determining the binding affinity of the tumor neoantigenic peptide to at least one MHC molecule of the cancer subject, optionally by computer simulation or using in vitro techniques (see the Examples for illustration).

When the method is performed on a human sample, the method may include a step of determining the class I or class II major histocompatibility complex (MHC, also known as human leukocyte antigen (HLA) alleles) of the patient. It should be noted that since the MHC alleles of laboratory mice are well known, this step may not be required in this particular case. In this application, "MHC molecule" refers to at least one MHC class I molecule or at least one MHC class II molecule.

An MHC allele database was established by analyzing the known sequences of MHC I and MHC II and determining the allelic variation of each domain. This can usually be determined in silico using appropriate software algorithms well known in the art. A variety of tools have been developed to obtain HLA allele information from whole genome sequencing data (whole exome, whole genome, and RNA sequencing data), including OptiType, Polysolver, PHLAT, HLAreporter, HLAforest, HLAminer, and seq2HLA (Kiyotani K et al., Immunopharmacogenomics towards personalized cancer immunotherapy targeting neoantigens; Cancer Science 2018; 109: 542-549). For example, the seq2hla tool (see Boegel S, Lower M, Schafer M, et al. HLA typingfrom RNA-Seq sequence reads. Genome Med. 2012; 4: 102) is well designed to perform the method disclosed in this article. It is a computer simulation method written in python and R, which takes standard RNA-Seq sequence reads in fastq format as input, uses a bowtie index containing all HLA alleles (Langmead B, et al., Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome Biol. 2009, 10: R25-10.1186/gb-2009-10-3-r25), and outputs the most likely HLA class I and class II genotypes (4-bit resolution), p-values for each call, and expression of each class.

Typically, sequences with connections between TE and exon sequences are extracted by computer simulation. The affinity of all possible peptides encoded by each sequence to each MHC allele from a patient (or mouse) can be determined in silico, for example, using computational methods to predict the binding affinity of the peptide to HLA molecules. In fact, accurate prediction methods are based on artificial neural networks and predicted _IC50 . For example, the NetMHCpan software, which has been modified to predict peptides that bind to alleles of unreported ligands, is well suited for implementing the methods disclosed herein ((Lundegaard C et al., NetMHC-3.0: accurate web accessible predictions of human, mouse and monkey MHC class I affinities for peptides of length 8-11; Nucleic Acids Res. 2008; 36: W509-W512; Nielsen M et al. NetMHCpan, a method for quantitative predictions of peptide binding to any HLA-A and -B locus protein of known sequence. PLoS One. 2007; 2: e796, but see also Kiyotani K et al., Immunopharmacogenomics towards personalized cancer immunotherapy targeting neoantigens; Cancer Science 2018; 109: 542-549 and Yarchoan M et al., Nat rev. cancer 2017;17(4):209-222). NetMHCpan software uses an artificial neural network (ANN) to predict peptide binding to any MHC molecule of known sequence. The method is trained on a combination of over 180,000 quantitative binding data and MS-derived MHC-eluted ligands. Binding affinity data cover 172 MHC molecules from human (HLA-A, B, C, E), mouse (H-2), bovine (BoLA), primate (Patr, Mamu, Gogo), and porcine (SLA). MS-eluted ligand data cover 55 HLA and mouse alleles.

In an exemplary embodiment, a neoantigenic peptide encoded by the fusion transcript as described above and having a Kd affinity for an MHC allele of less than 10 ⁻⁴ , 10 ⁻⁵ , 10 ⁻⁶ , 10 ⁻⁷ M or less than 500 nM, particularly less than 50 nM, is selected as a tumor neoantigenic peptide.

As described above, the affinity of the selected peptide to the MHC allele can be determined by computer simulation using appropriate software (e.g., netMHCpan). Thus, in some embodiments, the neoantigenic peptide binds to MHC class I with a binding affinity that is lower than the 2% percentile outlier score predicted by NetMHCpan 4.0. In other embodiments, the neoantigenic peptide binds to class II with a binding affinity that is lower than the 10% percentile outlier score predicted by NetMHCpanII 3.2.

Affinity can also (alternatively or additionally) be estimated in vitro, for example by using the MHC tetramer formation assay described in the results (see Example 2, points 2.1 and 2.2.2). The skilled person can usually use the data from commercial assays (according to their training guide, specifically using of Kit). Typically, the binding affinity is determined as a percentage of binding to a positive control. In general, a peptide is selected whose binding percentage is at least 30%, particularly at least 40%, or even at least 50% of that of the positive control. Typically, according to the present disclosure, the neoantigenic peptides that can be obtained according to the present method typically bind to at least one HLA/MHC molecule with an affinity sufficient to present the peptide as an antigen on the cell surface. Typically, the neoantigenic peptides have an IC50 affinity of less than ^10-4 , or ^10-5 , or ^10-6 , or ^10-7 , or less than 500nM, at least less than 250nM, at least less than 200nM, at least less than 150nM, at least less than 100nM, at least less than 50nM or less for at least one HLA/MHC molecule (typically a molecule of the cancer patient) (lower numbers indicate stronger binding affinity).

Therefore, further optional steps according to the present method may independently include:

- A step of excluding fusion transcripts or predicted peptides that are expressed at high levels or frequencies on healthy cells. Alignment of fusion transcript sequences to RNAseq data of healthy cells generally allows the relative quantity of fusion transcript sequences present in healthy cells to be determined; in one embodiment, fusion transcripts or predicted peptides expressed on healthy cells are discarded.

- A step of confirming that the tumor neoantigen peptide is not expressed in the healthy cells of the subject. This step can usually be performed using a basic local alignment search tool (BLAST) and comparing the neoantigen peptide sequence against the proteome of healthy cells; preferably, peptides compared against the proteome of normal healthy cells (e.g., using BLAST) are discarded.

- A step of confirming that the fusion transcript or predicted peptide is expressed in a cancer cell of the subject. The presence of the selected fusion transcript sequence in cancer cells can usually be examined by RT-PCR in mRNA extracted from a cancer cell sample.

In some embodiments, the method may also include a step of performing computer simulation translation of the identified fusion (JET) transcript to generate a JET-derived protein database (JET-db). Typically, the chain index JET containing genes as donors can use classical ORF translation, from the first stop codon after the involved gene until the breakpoint. In JET with TE as donor, 3 possible ORFs can be translated, and usually only the sequence closer to the breakpoint and between the two stop codons is retained. The JET db (usually also combined with the human proteome) can then query the mass spectrometry-based proteome data set obtained from tumor samples and/or tumor cell lines, which typically contains proteome data obtained from tumor samples and/or tumor cell lines. In some embodiments, a public mass spectrometry data set can be used. This embodiment is well described in the results provided in the present application. This analysis is also used to identify JET-derived peptides or proteins.

In a more specific embodiment, JETdb (typically bound to the human proteome) can be interrogated against immunopeptidomics-based mass spectrometry datasets, as detailed in the Examples included herein. This embodiment allows identification of JET-derived peptides or proteins (pJET) presented to MHC molecules.

To ensure that JET-derived peptides do not match classical proteins or JET-derived peptides found in normal samples, the identified peptides can be filtered using the UniProt/TrEMBL database and/or in silico translations of JET from normal (including, e.g., near-tumor) samples or cells (e.g., from public databases such as TCGA and/or CCLE).

New antigen peptide

The present disclosure also relates to an isolated tumor neoantigen peptide comprising at least 8, 9, 10, 11 or 12 amino acids, encoded by a portion of an open reading frame (ORF) from a fusion transcript, which is a human mRNA sequence comprising a transposable element (TE) sequence and an exon sequence. The length of the peptide can be 8-9, 8-10, 8-11, 12-25, 13-25, 12-20 or 13-20 amino acids. Although the connection between the ORF and the TE sequence and the exon sequence overlaps, it is understood that the tumor neoantigen peptide itself may not contain the connection.

More specifically, the present disclosure also encompasses an isolated tumor neoantigenic peptide encoded by a portion of a human fusion mRNA sequence from a cancer cell, the fusion mRNA comprising a TE sequence and an exon sequence.

The length of the peptide may be 8-9, 8-10, 8-11, 12-25, 13-25, 12-20 or 13-20 amino acids and meet one or more of the above-mentioned new antigenic peptide characteristics. The N-terminus of the peptide of at least 8 amino acids may be encoded by a triplet codon starting at any nucleotide position of 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18 or higher (it should be understood that the present disclosure contemplates any integer from 1 to 8000 starting position without listing every number from 1 to 8000).

The peptides defined above are generally obtainable according to the methods of the present disclosure and thus include one or more of the aforementioned characteristics. Specifically, the new antigenic peptides according to the present disclosure may exhibit one or a combination of the following further characteristics:

- It binds or specifically binds to the subject's MHC class I and is 8 to 11 amino acids, in particular 8, 9, 10 or 11 amino acids. Typically, the neoantigenic peptide is 8 or 9 amino acids in length and binds to at least one MHC class I molecule of the subject; or, it binds to at least one MHC class II molecule of the subject and contains 12 to 25 amino acids, in particular 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24 or 25 amino acids.

- It binds to at least one HLA/MHC molecule of the cancer patient with an affinity sufficient to present the peptide as an antigen on the cell surface. Typically, the neoantigenic peptide has an IC50 of less than ^10-4 , or ^10-5 , or ^10-6 , or ^10-7 , or less than 500 nM, at least less than 250 nM, at least less than 200 nM, at least less than 150 nM, at least less than 100 nM, at least less than 50 nM or less (lower numbers indicate stronger binding affinity).

- When administered to a subject, it does not induce a significant autoimmune response and/or induce immune tolerance.

-Its expression level in tumor samples is higher than that in normal healthy samples. Generally, according to the present disclosure, a fusion transcript can be selected if it is present in more than 1%, particularly more than 2%, more than 5%, or more than 10% of tumor samples (from the same or different tumor types, typically from one or more subjects, typically from TCGA tumor samples) and less than 20% of normal samples. Alternatively or additionally, the transcript can be identified in one or more (at least 2, 5, 10, 20, 50, 100 cell lines, such as cell lines from CCLE). In some embodiments, the new antigen is more specifically a tumor-specific antigen (TSA), that is, it is only expressed in cancer samples, not in normal samples, or is expressed at a relatively low level in normal samples (for example, the expressed mRNA sequence represents a minor type in normal cells from normal samples).

- it contains a junction between a TE sequence and an exon sequence, in other words, it is encoded by a portion of a TE sequence and a portion of an exon sequence, the ORF being canonical or non-canonical, or

- it is encoded by a non-canonical ORF in an exonic sequence, or

- which is encoded by a TE sequence, optionally in a non-classical ORF.

The tumor neoantigen peptide can first be verified by RT transcription analysis of the fusion transcript sequence in the subject's tumor cells. Likewise, immunization with the tumor neoantigen peptide according to the present invention will generally induce a T cell response.

In certain embodiments, the present disclosure encompasses NSCLC neoantigenic peptides comprising at least 8 amino acids of any one of SEQ ID NOs: 1-117. Typically, the neoantigenic peptides of SEQ ID NOs: 1-117 bind to HLA-A02 with an affinity sufficient to present the peptide as an antigen on the cell surface. Affinity for MHC alleles can be determined by techniques known in the art, particularly computer simulations or in vitro techniques as described above;

In certain embodiments, the tumor neoantigen peptides according to the present disclosure bind to at least 1%, 5%, 10%, 15%, 20%, 25% or more of the MHC molecules present in a subject. Of note, in a population of subjects with cancer, at least 1%, 5%, 10%, 15%, 20%, 25% of the subjects express the tumor neoantigen peptides disclosed herein.

More specifically, the tumor neoantigen peptides disclosed herein are capable of eliciting an immune response against tumors present in at least 1%, 5%, 10%, 15%, 20% or 25% of the subjects in a cancer patient population.

As defined above, cancer may affect any of the following tissues or organs: breast; liver; kidney; heart, mediastinum, pleura; floor of mouth; lips, salivary glands; tongue; gums; oral cavity; palate; tonsils; larynx; trachea; bronchi, lungs; pharynx, hypopharynx, oropharynx, nasopharynx; esophagus; digestive organs such as stomach, intrahepatic bile ducts, bile ducts, pancreas, small intestine, colon; rectum; urinary organs such as bladder, gallbladder, ureters; rectosigmoid colon; anus, anal canal; skin; bones; joints, limbs Articular cartilage of the body; eyes and appendages; brain; peripheral nerves, autonomic nervous system; spinal cord, cranial nerves, meninges; and various parts of the central nervous system; connective tissue, subcutaneous tissue, and other soft tissues; retroperitoneum, peritoneum; adrenal glands; thyroid gland; endocrine glands and related structures; female reproductive organs, such as ovaries, uterus, cervix; uterine body, vagina, vulva; male reproductive organs, such as penis, testes, and prostate; hematopoietic system and reticuloendothelial system; blood; lymph nodes; thymus. For example, tumors or cancers according to the present application include leukemia, seminoma, melanoma, teratoma, lymphoma, neuroblastoma, glioma, rectal cancer, endometrial cancer, kidney cancer, adrenal cancer, thyroid cancer, blood cancer, skin cancer, brain cancer, cervical cancer, intestinal cancer, liver cancer, colon cancer, stomach cancer, intestinal cancer, head and neck cancer, gastrointestinal cancer, lymph node cancer, esophageal cancer, colorectal cancer, pancreatic cancer, ear, nose, throat (ENT) cancer, breast cancer, prostate cancer, uterine cancer, ovarian cancer and examples thereof are lung cancer, breast cancer, prostate cancer, colon cancer, renal cell carcinoma, uterine cancer or metastasis of the above cancer types or tumors. The term "cancer" according to the present disclosure also includes cancer metastasis and cancer recurrence.

Typically, when administered to a subject, the neoantigenic peptides according to the present disclosure do not induce a significant autoimmune response and/or induce immune tolerance. Tolerance mechanisms involve clonal deletion, neglect, anergy or suppression in the host, in which the number of high-affinity self-reactive T cells is reduced.

The new antigenic peptide can also be modified by extending or reducing the amino acid sequence of the compound (for example, by adding or deleting amino acids). The peptide can also be modified by changing the order or composition of certain residues. It is easy to understand that certain amino acid residues that are essential for biological activity, such as residues in key contact sites or conserved residues, are usually not changed, so as to have no side effects on biological activity. Non-critical amino acids are not necessarily limited to those naturally present in proteins, such as L-α-amino acids or their D-isomers, but can also include non-natural amino acids, such as β-γ-δ-amino acids, and many derivatives of L-α-amino acids.

Typically, a series of peptides with single amino acid substitutions are used to determine the effects of electrostatic charge, hydrophobicity, etc. on binding. For example, a series of positively charged (e.g., Lys or Arg) or negatively charged (e.g., Glu) amino acid substitutions along the length of the peptide show different sensitivity patterns to various MHC molecules and T cell receptors. In addition, multiple substitutions using smaller, relatively neutral molecules (e.g., Ala, Gly, Pro or similar residues) can be used. The substitutions can be homo-oligomers or hetero-oligomers. The number and type of residues substituted or added depends on the necessary spacing between the necessary contact points and certain functional attributes sought (e.g., hydrophobicity vs. hydrophilicity). The binding affinity to MHC molecules or T cell receptors can also be improved by such substitutions compared to the affinity of the parent peptide. In any case, such substitutions should use selected amino acid residues or other molecular fragments to avoid, for example, steric and charge interference that may destroy binding.

Amino acid substitutions are usually single residues. Substitution, deletion, insertion or any combination thereof can be combined to obtain the final peptide. Substitution variants refer to variants in which at least one residue of the peptide is removed and another residue is inserted in its place. When it is desired to fine-tune the properties of the peptide, such substitutions are usually performed according to Table 1 below.

Initial residue Exemplary Substitutions Ala Ser Arg Lys, His Asn Gln Asp Glu Cys Ser Gln Asn Glu Asp Gly Pro His Lys, Arg Ile Leu, Val Leu Ile, Val Lys Tyr, Trp Met Thr Phe Ser Ser Tyr, Phe Tyr Trp, Phe Val Ile, Leu Pro Gly

Table 1

Substantial changes in function (e.g., affinity for MHC molecules or T cell receptors) are achieved by selecting substitutions that are less conservative than those in the table above, i.e., by selecting residues that have a more significant difference in their role in maintaining (a) the structure of the peptide backbone in the area of the substitution (e.g., sheet or helical conformation) or (b) the charge or hydrophobicity of the molecule at the target site or (c) the bulk of the side chain. Substitutions that are generally expected to produce the greatest changes in peptide properties are (a) substitutions of (or with) hydrophobic residues (e.g., leucine, isoleucine, phenylalanine, valine, or alanyl) for hydrophilic residues (e.g., serine); (b) substitutions of (or with) negatively charged residues (e.g., glutamyl or aspartyl) for residues with positively charged side chains (e.g., lysine, arginine, or histidine); or (c) substitutions of (or with) residues with bulky side chains (e.g., phenylalanine) for residues without side chains (e.g., glycine).

The peptides and polypeptides may also contain isosteres of two or more residues in the neoantigenic peptide or neoantigenic polypeptide. The isosteres defined herein are sequences that can replace two or more residues of the second sequence because the spatial conformation of the first sequence fits the specific binding site of the second sequence. The term specifically includes peptide backbone modifications well known to those skilled in the art. Such modifications include modifications of amide nitrogen, α-carbon, amide carbonyl, complete replacement, extension, deletion or backbone cross-linking of amide bonds. See generally Spatola, Chemistry and Biochemistry of Amino Acids, Peptides and Proteins, Vol. VII (Weinstein ed., 1983).

In addition, the neoantigenic peptides can be conjugated to carrier proteins, ligands or antibodies. The half-life of the peptides can be increased by PEGylation, glycosylation, polysialylation, HESylation, recombinant PEG mimics, Fc fusion, albumin fusion, nanoparticle attachment, nanoparticle encapsulation, cholesterol fusion, iron fusion or acylation.

Modification of peptides and polypeptides with various amino acid mimetics or unnatural amino acids is particularly helpful in improving the stability of peptides and polypeptides in vivo. Stability can be determined in a variety of ways. For example, peptidases and various biological media, such as human plasma and serum, have been used to test stability. See, for example, Verhoef et al., Eur. J. Drug Metab Pharmacokin. 11: 291-302 (1986). The half-life of the peptides disclosed herein is conveniently determined using a 25% human serum (v/v) assay. The protocol is generally as follows. Pooled human serum (AB type, non-heat inactivated) is defatted by centrifugation prior to use. The serum is then diluted to 25% with RPMI tissue culture medium and used to test the stability of the peptides. At predetermined time intervals, a small amount of the reaction solution is removed and added to 6% trichloroacetic acid aqueous solution or ethanol. The turbid reaction sample is cooled (4° C.) for 15 minutes and then spun to precipitate the precipitated serum proteins. The presence of the peptide is then determined by reverse phase HPLC using stability-specific chromatographic conditions.

Peptides and polypeptides can be modified to provide the desired attributes in addition to improved serum half-life. For example, the ability of peptides to induce CTL activity can be enhanced by being connected to a sequence comprising at least one epitope that can induce T helper cell responses. Particularly preferred immunogenic peptide/T helper conjugates are connected by spacer molecules. Spacers are usually composed of relatively small neutral molecules, such as amino acids or amino acid mimetics that are substantially uncharged under physiological conditions. Spacers are usually selected from neutral spacers such as Ala, Gly or other non-polar amino acids or neutral polar amino acids. It should be understood that the optional spacers do not need to be composed of the same residues, and therefore can be hetero-oligomers or homo-oligomers. When present, spacers are usually at least one or two residues, more usually three to six residues. Alternatively, the peptide can be connected to the T helper peptide without spacers.

The neoantigenic peptide can be linked to the T helper peptide directly or via a spacer located at the amino or carboxyl terminus of the peptide. The amino terminus of the neoantigenic peptide or the T helper peptide can be acylated. Exemplary T helper peptides include tetanus toxoid 830-843, influenza virus 307-319, malarial ringworm 382-398 and 378-389

The multiple neoantigenic peptides described herein may also be optionally linked together via spacers.

Transmembrane chimeric polypeptide (or protein) and antigen binding domain binding thereto

The present disclosure provides a set of transmembrane chimeric polypeptides (also referred to herein as pJET or fusion transcript-derived peptides) that provide excellent extracellular neoantigen candidates. The chimeric proteins are derived from fusion transcripts predicted from a bioinformatics pipeline developed to identify genome-wide non-classical splicing regions as defined above. Publicly available RNA-Seq data from TCGA (Cancer Genome Atlas) and CCLE (Broad Institute Cancer Cell Line Encyclopedia) were used (described in the Examples section).

The identification and characterization of fusion transcripts according to the invention (also referred to herein as chimeric transcripts or JETs) has been described in detail in the above section of the present application (see also the Examples).

As previously mentioned, fusion transcripts come from alternative splicing mechanisms known to be essential for producing functional diversity. In fact, the splicing mechanism allows the expression of multiple mRNAs (i.e., transcripts or splice variants) encoding multiple proteins from a single gene through the rearrangement of existing exon and intron sequences. The types of splicing changes observed herein include exon skipping, intron retention, and the use of alternative splicing donor or acceptor sites. In fusion transcripts according to the present invention, TE can be used as a donor (positioned at 5') or an acceptor (positioned at 3'), and accordingly, an exon can be an acceptor or a donor. Therefore, TE-exon splicing causes a part of a "non-coding" genome to be incorporated into a coding genome, thereby exposing a non-coding genome sequence to a translation mechanism. These fusion (or chimeric) transcripts, also referred to as JET (joining exon TE), include ORF (open reading frame), i.e., they are parts that have the ability to be translated into polypeptides or proteins in a reading frame. When TE is the acceptor, the ORF of the fusion transcript is canonical (i.e., the same as the canonical transcript), while when TE is the donor, the ORF can be canonical (usually ORF1), or can be frameshifted by 1 or 2 nucleotides compared to ORF1 (usually ORF2 and 3, respectively). The fusion transcript includes not only the fused TE and exon sequence (corresponding to JET), but may further include exons corresponding to various transcript isoforms, upstream of the fusion breakpoint (between the exon and TE) if the exon is the donor, and downstream of the fusion breakpoint if the TE is the donor.

More specifically, the present disclosure provides transmembrane chimeric polypeptides (or proteins) mentioned in Tables 9-15 and 19-20, SEQ ID NOs: 1-21542. Tables 14 and 19 provide amino acid sequences translated from fusion transcripts (JET), wherein exons are donors. Tables 15 and 20 provide amino acid sequences translated from fusion transcripts (JET), wherein exons are donors.

This set of (transmembrane) chimeric proteins is obtained by further selecting fusion transcripts with exon sequences that are annotated as belonging to transcripts encoding transmembrane proteins in normal proteome databases (e.g., typically UniProt). The sequences of the selected fusion transcripts are then translated (in silico) into fusion (or chimeric) polypeptide sequences (also called translated junctions or pJETs or translated JETs).

Fusion transcripts in which the exon is the donor are translated after the classical ORF of the transcript from the start of the transcript to the first stop codon after the breakpoint between the exon and the TE.

The fusion transcript in which TE is the donor is translated after 3 ORFs (1-3) from the start of TE or from the last stop codon before the breakpoint between TE and exon to the first stop codon after the breakpoint.

Only translated polypeptide sequences containing at least 3 amino acids derived from TE sequences were retained.

In some embodiments, peptide sequences from translated junctions that matched any reference or annotated protein sequence in UniProt were discarded, thus focusing on unannotated chimeric peptides (as exemplified in the Results).

Tables 9-13 provide peptides (usually polypeptides of size) sequences identified from the aforementioned protein genome methods. Typically, the fusion transcripts (i.e., JET) libraries generated by the pipeline of the present application are searched in MS raw data from total proteomics (all peptide groups) or surface proteomics (surface groups) (using various public RNA seq data sets).

Therefore, the present disclosure is particularly directed to chimeric polypeptides (or proteins) expressed on cell membranes. Cell membrane expression of computer-simulated polypeptides partitioned into transmembrane compartments (as described above) can be experimentally verified at different molecular levels by several in vitro methods:

A) The protein or peptide containing the selected translation connection can be ectopically expressed in a host cell, such as a tumor cell line (such as Hela, CHO, etc.) to confirm stability and proper integration within the plasma membrane. An expression vector containing a translation connection and a tag sequence (such as FLAG or HA) can be designed to produce a tagged fusion chimeric protein. The fused protein or peptide also contains an epitope tag, so it can be detected by flow cytometry or microscopy using commercially available anti-tag antibodies. The sequence cloned into the vector can preferably be designed in a form in which the tag sequence is located immediately before or after the TE sequence. This method allows the selection of a translation connection peptide or protein that produces a stable protein, which is expressed in the cell membrane and usually exposes the TE sequence to the extracellular space.

b) Targeted sequencing experiments can also be performed to amplify and detect fusion transcripts in additional tumor specimens or cell lines. This can be performed by conventional PCR, quantitative real-time PCR, or SMRT full-length transcript sequencing (PACBIO Technology).

c) Translated sequences can also be detected in the "translatome" (which represents the entirety of the translating mRNA in the cell) by ribosome profiling or ribo-Seq. Thus, ribosome profiling can monitor the translation process of the junction transcript and predict the abundance of chimeras (translated junction peptides or proteins). By this technique, the regions of the translated JET-derived transcripts (mRNA) can be identified, thereby determining the translation start and stop sites of the fusion polypeptide. Using this method of genomic and proteomic validation of the bridged JET sequence, the translated regions of the fusion transcript can be fully determined.

d) Further experimental validation of protein expression can be performed by targeted mass spectrometry approaches in tumor specimens and cell lines. Targeted proteomics experiments can be performed to quantify small amounts of chimeric proteins (including translated junctions) at a time with extremely high precision, sensitivity, specificity, and throughput.

In a more specific embodiment, the present disclosure refers to a chimeric polypeptide as defined herein, wherein a portion of the sequence derived from the TE nucleotide sequence is exposed on the cell surface. The cell surface exposure of the TE-derived sequence can be predicted in silico based on the predicted topology of the TE-derived sequence.

Predicting that a protein is located on the cell surface generally involves (i) detection of TM domains or lipid anchors; (ii) definition of the orientation of the protein within the membrane, including identification of extracellularly exposed domains; and/or (iii) prediction of subcellular location. There are bioinformatic tools for predicting TM domains, signal peptides and GPI-linked proteins (see Kail L, Krogh A, Sonnhammer ELL (2004) A combined transmembrane topology and signal peptide prediction method. J Mol Biol 338: 1027-1036; Jones DT (2007) Improving the accuracy of transmembrane protein topology prediction using evolutionary information. Bioinformatics 23: 538-544; Reeb J, Kloppmann E, Bemhofer M, Rost B (2015) Evaluation of transmembranehelix predictions in 2014. Proteins 83: 473-484; Krogh A, Larsson B, von Heijne G, Sonnhammer ELL (2001) Predicting transmembrane protein topology with a hidden Markov model: Application to complete genomes. J Mol Biol 305:567-580; Viklund H, Elofsson A (2008) OCTOPUS: Improving topology prediction by two-track ANN-based preference scores and an extended topological grammar. Bioinformatics 24:1662-1668; Fankhauser N, Maser P (2005) Identification of GPI anchor attachment signals by a Kohonen self-organizing map. Bioinformatics 21:1846-1852).

For a given translated linkage, experimental verification of cell surface expression of the TE-derived sequence can also be performed as described above (see point a).

As described above, identification of peptides expressed on the cell surface can be achieved based on a proteomic approach by searching the JET library from the MS raw data of the surface group (all plasma membrane proteins with at least 1, in particular at least 2, at least 3 or at least 4 amino acid residues exposed to the extracellular space). The surface group is therefore a subset of the plasma membrane proteome, which is a subset of the membrane proteome (all membrane proteins as a whole). Unilateral integral membrane proteins attached to the extracellular lipid layer [e.g., via a glycosylphosphatidylinositol (GPI) anchor] are part of the human surface group.

In some embodiments, the present disclosure is more specifically directed to chimeric polypeptides generated from a non-classical ORF downstream of the junction between a TE-derived sequence and an exon-derived sequence. By positioning the tag before or after the non-classical ORF sequence, the cell surface expression of a non-classical ORF translation sequence with a predicted extracellular topology can also be experimentally confirmed as described above (see point a).

In some embodiments, the translational linkage can be selected based on the fusion type (TE donor or acceptor) and the subtype of the membrane protein. This selection can be achieved by:

-Integral membrane proteins

o Type I single-pass proteins (positioned so that their carboxyl termini face the cytosol): selected transcripts are derived from fusions with TE as donor (fusion TE->exon), which in some cases are preceded by a second fusion with TE as acceptor (fusion exon-TE). In the latter case, transcripts are generated by a double fusion "exon-TE-exon" or "metafusion", in which the resulting transcript includes a TE exon exonisated sequence flanking two classical exons.

○ Class II single-pass proteins (with their amino termini facing the cytosol): selected transcripts are derived from fusions with TE as acceptor (fusion exon->TE), which in some cases are followed by a second fusion with TE as donor (fusion TE->exon). In the latter case, transcripts are generated from double fusions "exon-TE-exon" or "meta-fusions", where the resulting transcripts include a TE exon exon sequence flanking two canonical exons.

o Multipass or multispanning proteins (polypeptide chains pass through the membrane multiple times): TE sequences can act as donors or acceptors and/or as part of a metafusion. The breakpoint (or one of the two breakpoints in a metafusion) is located on the extracellular side of the membrane between the two transmembrane helices.

○ Unilateral integrins (integral membrane proteins that are attached to only one side of the membrane and do not span the entire membrane): Transcripts selected are derived from fusions with TE as acceptor (fusion exon->TE), as donor (fusion TE->exon), or as both (meta-fusion).

- Peripheral membrane proteins (attached or bound to the cell membrane): Transcripts selected are derived from fusions with TE as acceptor (fusion exon->TE), as donor (fusion TE->exon) or both (meta-fusion).

Typically, the chimeric protein according to the present disclosure is expressed in more than 1%, particularly more than 5%, and typically more than 10% of tumor samples (from one or more subjects and/or from one or more tumor types, where tumor samples can be obtained from TCGA). Alternatively or additionally, the chimeric protein (or pJET) may be expressed in one or more cell lines (typically from CCLE), particularly in at least 1; 2; 3; 4; 5; 6; 7; 8; 9; 10; 20; 50; 100 cell lines.

Typically, the chimeric protein of the present disclosure is expressed at a higher level in tumor samples than in normal samples (including adjacent tumor samples) and/or cell lines. More specifically, the chimeric protein is preferably expressed in less than 20%, particularly less than 10%, less than 5% or less than 1% of normal samples or cell lines. In some embodiments, the chimeric protein is not detectably expressed in normal samples (including adjacent tumor samples).

The present disclosure also encompasses chimeric polypeptide variants having at least 50%, in particular at least 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% identity to any chimeric polypeptide (or protein) of SEQ ID NO: 1-8202. The variant chimeric polypeptide is expressed on the cell surface membrane. Typically, in the variant, the sequence derived from the TE nucleotide sequence maintains at least 90%, 95%, 96%, 97%, 98%, 99% identity to the TE-derived sequence of the variant-derived peptide. More preferably, the variant chimeric protein does not match any annotated polypeptide or protein in a normal proteome database (e.g., UniProt).

The (transmembrane) chimeric proteins disclosed herein can also be modified by extending or reducing the amino acid sequence of the compound, for example by adding or deleting amino acids. The chimeric protein can also be modified by changing the order or composition of certain residues, and it is easy to understand that certain amino acid residues that are essential for biological activity, such as key contact sites or conserved residues, are usually not changed without adversely affecting the biological activity. Non-critical amino acids are not necessarily limited to those naturally present in proteins, such as L-α-amino acids or their D-isomers, and can also include non-natural amino acids, such as β-γ-δ-amino acids, and many L-α-amino acid derivatives.

Typically, a series of single amino acid substitutions are made to the peptide to determine the effects of net charge, hydrophobicity, etc. on binding. Substitutions may be homo-oligomeric or hetero-oligomeric. The number and type of residues substituted or added will depend on the necessary spacing between necessary contact points and certain functional attributes sought (e.g., hydrophobicity vs. hydrophilicity). In any case, such substitutions should be made with amino acid residues or other molecular fragments selected to avoid, for example, steric and charge interferences that could disrupt binding.

Amino acid substitutions are usually single residues. Substitution, deletion, insertion or any combination thereof can be combined to obtain the final peptide. Substitution variants refer to variants in which at least one residue of the peptide is removed and another residue is inserted in its place. When it is desired to fine-tune the properties of the peptide, such substitutions are usually performed according to Table 1 below.

It should be mentioned that the chimeric polypeptides or proteins disclosed herein can also be processed by the proteasome mechanism and produce a neoantigenic peptide of generally at least 8 amino acids (particularly 8-25 amino acids), which binds to at least one major histocompatibility complex (MHC) molecule of the subject as described in detail above.

The present disclosure also encompasses an antigen binding domain as described herein, which binds to a chimeric protein or fragment thereof as defined above, in particular a neoantigenic tumor sequence (or epitope) of at least 4, 5, 6, 7, or 8 amino acids in length, wherein the dissociation constant (Kd) is about 2× ^10-7 M or less. In certain embodiments, the _Kd is about 2× ^10-7 M or less, about 1× ^10-7 M or less, about 9× ^10-8 M or less, about 9× ^10-9 M or less, about 5× ^10-9 M or less, about 4× ^10-9 M or less, about 3× ^10-9 M or less, about 2× ^10-9 M or less, or about 1× ^10-9 M or less, or about 1× ^10-10 M or less, or about 1× ^10-12 M or less. In certain non-limiting embodiments, the _Kd is about 3× ^10-9 M or less. In certain non-limiting embodiments, K _d is from about 1×10 ^-9 M to about 3×10 ^-7 M. In certain non-limiting embodiments, K _d is from about 1.5×10 ^-9 M to about 3×10 ^-7 M. In certain non-limiting embodiments, K _d is from about 1.5×10 ^-9 M to about 2.7×10 ^-7 M. In certain non-limiting embodiments, K _d is from about 1.5×10 ^-10 M to about 2.7×10 ^-7 M. In certain non-limiting embodiments, K _d is from about 1.5×10 ^-12 M to about 2.7×10 ^-7 M.

The binding of the antigen binding domain (e.g., Fv or its analog) can be confirmed by, for example, enzyme-linked immunosorbent assay (ELISA), radioimmunoassay (RIA), FACS analysis, bioassay (e.g., growth inhibition), Western Blot assay, or fluorescence microscopy. Each of these assays is usually performed by using a labeling agent (e.g., antibody or Fv) specific for the target complex to detect the presence of a specific target protein-antibody complex. For example, Fv can be radiolabeled and used in radioimmunoassay (RIA) (see, e.g., Weintraub, B., Principles of Radioimmunoassays, Seventh Training Course on Radioligand Assay Techniques, The Endocrine Society, March, 1986, which is incorporated herein by reference). Radioisotopes can be detected by using methods such as a g counter or a scintillation counter or autoradiography.

For microscopy, the labeling agent (e.g., antibody or Fv) can be directly conjugated to a fluorophore, or recognized by a secondary antibody conjugated to the fluorophore of the labeling agent. Non-limiting examples of fluorophores (also called fluorescent dyes) include derivatives of cyanines (e.g., Cy3) or rhodamines (e.g., TRITC) or fluoresceins (e.g., FITC).

In certain embodiments, the extracellular antigen-binding domain is labeled with a fluorescent marker. Non-limiting examples of fluorescent markers include green fluorescent protein (GFP), blue fluorescent protein (e.g., EBFP, EBFP2, Azurite and mKalamal), cyan fluorescent protein (e.g., ECFP, Cerulean and CyPet) and yellow fluorescent protein (e.g., YFP, Citrine, Venus and YPet).

In some embodiments, the antigen binding domain disclosed herein binds to a fragment of the amino acid sequence (or epitope) of the chimeric protein described herein (or a tumor neoantigen peptide sequence), the chimeric protein comprising at least one TE-derived amino acid sequence or from any one of SEQ ID NOs: 1424-8202; 8203-10163 and 12831-21542 (generally encoded by a fusion transcript, wherein TE is the donor). In some embodiments, the peptide sequence from the chimeric protein described herein overlaps the breakpoint between the TE-derived amino acid sequence and the exon-derived amino acid sequence. In other embodiments, the peptide sequence is derived from a pure TE sequence. In other embodiments, the peptide sequence is encoded by a non-classical ORF downstream of the junction between the TE-derived amino acid sequence and the exon-derived amino acid sequence.

In some embodiments, an antigen binding domain according to the present disclosure binds to a neoantigenic peptide sequence from any one of the chimeric polypeptide neoantigenic peptides or fragments thereof of the present disclosure, wherein the neoantigenic peptide sequence:

a) from any one of SEQ ID NOs: 1-21542 or a fragment thereof, and comprising at least one sequence derived from an amino acid or nucleotide sequence derived from TE,

Optionally (i) a fragment overlapping the breakpoint between the TE-derived amino acid sequence and the exon-derived amino acid sequence, or,

Optionally (ii) pure TE sequence; or

b) from any one of SEQ ID NOs: 1-1423; 8203-10163 and 10164-12830 (typically encoded by a fusion transcript, in which the exon is the donor) or a fragment thereof, and encoded by a non-canonical ORF downstream of the junction between the TE-derived amino acid sequence and the exon-derived amino acid sequence.

Typically, the peptide sequence is derived from the extracellular portion of the chimeric protein.

In certain embodiments, the antigen binding domain comprises an antigen binding portion of a TCR.

In certain embodiments, the antigen binding domain comprises an antigen binding portion of an antibody or fragment thereof. In certain embodiments, the antigen binding domain comprises a heavy chain variable region (VH) and/or a light chain variable region (VL) of an antibody. In certain embodiments, the antigen binding domain comprises a single chain variable fragment (scFv).

In certain embodiments, the antigen binding domain comprises a heavy chain only antibody (VHH) or a variant thereof and/or a VL domain or a variant thereof.

In certain embodiments, the antigen binding domain comprises a Fab, which is optionally cross-linked. In certain embodiments, the antigen binding domain comprises a F(ab)2. In certain embodiments, any of the foregoing molecules may be included in a fusion protein with a heterologous sequence to form an antigen binding domain.

In certain embodiments, the extracellular antigen binding domain is derived from a scFv, Fab or antibody derived from mouse, human or camelid (eg, llama).

Peptide products, polynucleotides and vectors

Protein or peptide can be prepared by any technology known to those skilled in the art, including expression of protein, polypeptide or peptide by standard molecular biology techniques, isolation of protein or peptide from natural sources, or chemical synthesis of protein or peptide. Nucleotide and protein, polypeptide and peptide sequences corresponding to various genes have been previously disclosed and can be found in computerized databases known to those of ordinary skill in the art. One such database is the Genbank and GenPept databases of the National Center for Biotechnology Information located at the National Institutes of Health website. The coding region of a known gene can be amplified and/or expressed using technology disclosed herein or known to those of ordinary skill in the art. Alternatively, various commercial preparations of protein, polypeptide and peptide are known to those skilled in the art.

On the other hand, the present disclosure provides nucleic acids (e.g., polynucleotides) encoding new antigenic peptides as disclosed herein. The polynucleotides may be selected from DNA, cDNA, PNA, CNA, RNA, single-stranded and/or double-stranded, or natural or stable forms of polynucleotides, such as polynucleotides with a thiophosphate backbone or a combination thereof, and as long as it encodes a peptide, it may or may not contain introns. Only peptides containing naturally occurring amino acid residues connected by naturally occurring peptide bonds can be encoded by polynucleotides.

Another aspect of the present disclosure provides an expression vector capable of expressing a new antigenic peptide as disclosed herein. Expression vectors for different cell types are well known in the art and can be selected without undue experimentation. Typically, DNA is inserted into an expression vector (such as a plasmid) in the proper direction and correct reading frame for expression. The expression vector will contain a suitable heterologous transcription and/or translation regulatory nucleotide sequence recognized by the desired host. The polynucleotide encoding the tumor new antigen peptide may be linked to the heterologous regulatory nucleotide sequence, or may be non-adjacent to the heterologous regulatory nucleotide sequence but still operably linked. The vector is then introduced into the host by standard techniques. For example, guidance can be found in Sambrook et al. (1989) Molecular Cloning, A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.

Antigen presenting cells

The present disclosure also includes a population of antigen presenting cells that have been pulsed with one or more of the aforementioned peptides and/or obtained by the aforementioned methods. Preferably, the antigen presenting cells are dendritic cells or artificial antigen presenting cells (see Neal, Lillian R et al. "The Basics of Artificial Antigen Presenting Cells in T Cell-Based Cancer Immunotherapies." Journal of immunology research and therapy vol. 2, 1 (2017): 68-79). Dendritic cells (DC) are specialized antigen presenting cells (APC) with the extraordinary ability to stimulate initial T cells and initiate primary immune responses to pathogens. In fact, the main function of mature DC is to sense antigens and produce mediators that activate other immune cells (especially T cells). Since DC expresses MHC molecules that trigger TCR (signal 1) and co-stimulatory molecules (signal 2) on T cells, it is an effective stimulatory factor for lymphocyte activation. In addition, DC also secretes cytokines that support T cell expansion. T cells need to present antigens in the form of processed peptides to recognize foreign pathogens or tumors. Pathogen/tumor protein-derived peptide epitopes are presented by MHC molecules. MHC class I (MHC-I) and MHC class II (MHC-II) molecules present processed peptides to CD8+T cells and CD4+T cells, respectively. Importantly, DCs home to inflammatory sites containing a large number of T cell populations to promote immune responses. Therefore, DCs may be an important component of any immunotherapy approach because they are closely related to the activation of adaptive immune responses. With regard to vaccines, DC therapy can enhance the immune response of T cells to the desired target in healthy volunteers or patients with infectious diseases or cancer. In one embodiment, APCs are artificial APCs that are genetically modified to express the desired T cell co-stimulatory molecules, human HLA alleles and/or cytokines. This artificial antigen presenting cell (aAPC) can provide sufficient T cell participation, co-stimulation, and the requirement for sustained release of cytokines, thereby allowing controlled expansion of T cells. These cells are not limited by time and limited availability and can be stored in small portions for subsequent generation of T cell lines from different donors, thus representing ready-made reagents for immunotherapy applications. The expression of potent co-stimulatory signals on these aAPCs confers greater efficiency to this system, thereby improving the efficacy of adoptive immunotherapy. In addition, aAPCs can be engineered to express genes that direct the release of specific cytokines to promote the preferential expansion of desired T cell subsets for adoptive transfer; for example, long-lived memory T cells (see review Hasan AH et al., . Artificial Antigen Presenting Cells: An Off the Shelf Approach for Generation of Desirable T-Cell Populations for Broad Application of Adoptive Immunotherapy; Adv Genet Eng. 2015; 4(3): 130, Kim JV, Latouche JB, Rivière I, Sadelain M. The ABCs of artificial antigen presentation. Nat Biotechnol. 2004; 22: 403-410 or Wang C, Sun W, Ye Y, Bomba HN, Gu Z. Bioengineering of Artificial Antigen Presenting Cells and Lymphoid Organs. Theranostics 2017; 7(14): 3504-3516.).

Typically, dendritic cells are autologous dendritic cells pulsed with the new antigenic peptides disclosed herein. The peptide can be any suitable peptide that causes an appropriate T cell response. Antigen presenting cells (or stimulator cells) typically have MHC class I or class II molecules on their surface, and in one embodiment, their own body is substantially unable to load the selected antigen to MHC class I or class II molecules. MHC class I or class II molecules can be easily loaded with selected antigens in vitro.

As an alternative, the antigen presenting cell may comprise an expression construct encoding the tumor neoantigen peptide disclosed herein. The polynucleotide may be any suitable polynucleotide as defined above, and preferably it is capable of transducing dendritic cells, resulting in presentation of the peptide and induction of immunity.

Thus, the present invention includes such APC populations: which can be pulsed or loaded with the neoantigenic peptides described herein, genetically modified (by DNA or RNA transfer) to express at least one neoantigenic peptide described herein, or contain expression constructs encoding tumor neoantigenic peptides described herein. Typically, the APC population is pulsed or loaded, modified to express or contain at least one, at least 5, at least 10, at least 15 or at least 20 different neoantigenic peptides or expression constructs encoding them.

The present disclosure also includes compositions comprising the APCs disclosed herein. The APCs can be suspended in any known physiologically compatible pharmaceutical carrier, such as cell culture medium, physiological saline, phosphate buffered saline, cell culture medium, etc., to form a physiologically acceptable aqueous pharmaceutical composition. Parenteral media include sodium chloride solution, Ringer's dextrose, dextrose and sodium chloride, lactated Ringer's. Other substances, such as antimicrobial agents, may be added as needed. As used herein, "carrier" refers to any substance suitable as a medium for delivering APCs to a suitable in vitro or in vivo site of action. Therefore, the carrier can serve as an excipient for the preparation of therapeutic agents or experimental agents containing APCs. Preferred carriers are capable of maintaining APCs in a form capable of interacting with T cells. Examples of such carriers include, but are not limited to, water, phosphate buffered saline, saline, Ringer's solution, glucose solution, serum-containing solution, Hank's solution, and other physiologically balanced aqueous solutions or cell culture media. The aqueous carrier may also contain suitable auxiliary substances required to approach the physiological conditions of the recipient, such as enhancing chemical stability and isotonicity. Suitable auxiliary substances include, for example, sodium acetate, sodium chloride, sodium lactate, potassium chloride, calcium chloride, sorbitan monolaurate, triethanolamine oleate, and other substances for the production of phosphate buffers, Tris buffers, and bicarbonate buffers.

Vaccine composition

The present disclosure further includes vaccines or immunogenic compositions capable of eliciting a specific T cell response, comprising:

- one or more neoantigenic peptides as defined herein,

- one or more polynucleotides encoding a neoantigenic peptide as defined herein; and/or

- A population of antigen presenting cells as described above (eg autologous dendritic cells or artificial APCs).

Preferably, according to the present disclosure, the neoantigenic peptides encoded by the previously defined tumor-specific fusions are used in the vaccine composition. The neoantigenic peptides may also be referred to as tumor-specific peptides. Preferably, according to the present disclosure, polynucleotides encoding tumor-specific peptides are also used.

A suitable vaccine or immunogenic composition will preferably comprise 1 to 20 neoantigenic peptides, more preferably 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25 different neoantigenic peptides, further preferably 6, 7, 8, 9, 10, 11, 12, 13 or 14 different neoantigenic peptides, and most preferably 12, 13 or 14 neoantigenic peptides.

One or more neoantigenic peptides may be linked to a carrier protein. When the composition comprises two or more neoantigenic peptides, the two or more (e.g., 2-25) peptides may be linearly linked by a spacer molecule as described above (e.g., a spacer comprising 2-6 non-polar or neutral amino acids).

In one embodiment of the present disclosure, different neoantigenic peptides, encoding polynucleotides, vectors or APCs are selected so that a vaccine or immunogenic composition comprises neoantigenic peptides that can bind to different MHC molecules (e.g., different MHC class I molecules). Preferably, the neoantigenic peptide is capable of binding to the MHC class I molecules that appear with the highest frequency, for example, different fragments that can bind to at least 2 preferred, more preferably at least 3 preferred, and even more preferably at least 4 preferred MHC class I molecules. In some embodiments, the composition comprises peptides, encoding polynucleotides, vectors or APCs that can bind to one or more class II molecules. MHC is optionally HLA-A, -B, -C, -DP, -DQ or -DR.

The vaccine or immunogenic composition is capable of eliciting a specific cytotoxic T cell response and/or a specific helper T cell response.

Therefore, in a specific embodiment, the present disclosure also relates to a new antigenic peptide as described above, wherein the new antigenic peptide has a tumor-specific new epitope and is contained in a vaccine or immunogenic composition. Vaccine compositions should be understood to refer to compositions used to produce immunity for preventing and/or treating diseases. Therefore, vaccines are drugs that contain or produce antigens, and are intended for humans or animals to produce specific resistance and protective substances through vaccines. "Immunogenic composition" should be understood to refer to a composition that contains or produces an antigen and is capable of eliciting an antigen-specific humoral or cellular immune response (e.g., a T cell response).

In a preferred embodiment, the length of the new antigenic peptide according to the present disclosure is 8 or 9 residues, or 13 to 25 residues. When the peptide is less than 20 residues, in order to make the peptide more suitable for in vivo immunization, the new antigenic peptide is optionally flanked by additional amino acids to obtain an immune peptide with more amino acids (usually more than 20).

Pharmaceutical compositions (i.e., vaccines or immunogenic compositions) comprising the peptides described herein can be administered to individuals who already have cancer. In therapeutic applications, the composition is administered to the patient in an amount sufficient to induce an effective CTL response to the tumor antigen and to cure or at least partially inhibit symptoms and/or complications. An amount sufficient to achieve this purpose is defined as a "therapeutically effective dose". The effective amount for this purpose depends on, for example, the peptide composition, the mode of administration, the stage and severity of the disease being treated, the patient's weight and general health, and the judgment of the prescribing physician, but the initial immunization (i.e., therapeutic or prophylactic administration) is generally in the range of about 1.0 μg to about 50,000 μg of peptide for a 70 kg patient, followed by increasing the dose of the peptide or increasing the dose of the peptide from about 1.0 μg to about 10,000 μg according to a booster regimen of several weeks to several months, depending on the patient's response and condition, by measuring the specific CTL activity in the patient's blood. It must be remembered that the peptides and compositions of the present invention are generally used in serious disease states, i.e., life-threatening or potentially life-threatening situations, especially when the cancer has metastasized. In such cases, it is likely and likely that the treating physician will desirably administer a substantial excess of these peptide compositions in view of the minimization of foreign matter and the relatively nontoxic nature of the peptides.

For therapeutic use, administration should be initiated upon detection or surgical removal of a tumor, and the dosage should be increased thereafter until at least substantial amelioration of symptoms is achieved and sustained for a period of time.

Vaccines or immunogenic compositions for therapeutic treatment are intended for parenteral, topical, nasal, oral or topical administration. Preferably, the pharmaceutical composition is administered parenterally, such as intravenously, subcutaneously, intradermally or intramuscularly. The composition can be administered at the site of surgical resection to induce a local immune response to the tumor.

The vaccine or immunogenic composition can be a pharmaceutical composition, which additionally comprises a pharmaceutically acceptable adjuvant, an immunostimulant, a stabilizer, a carrier, a diluent, an excipient and/or any other material well known to those skilled in the art. Such materials should be non-toxic and should not interfere with the efficacy of the active ingredient. The carrier is preferably an aqueous carrier, but the exact nature of its carrier or other materials will depend on the route of administration. A variety of aqueous carriers can be used, such as water, buffered water, 0.9% saline, 0.3% glycine, hyaluronic acid, etc. These compositions can be sterilized by conventional, well-known sterilization techniques, or can be aseptically filtered. The resulting aqueous solution can be packaged and used as is, or lyophilized, and the lyophilized preparation is mixed with a sterile solution before administration. The composition may further comprise pharmaceutically acceptable auxiliary substances required to approach physiological conditions, such as pH regulators and buffers, tension regulators, wetting agents, etc., such as sodium acetate, sodium lactate, sodium chloride, potassium chloride, calcium chloride, sorbitan monolaurate, triethanolamine oleate, etc. See, e.g., Butterfield, BMJ. 2015 22; 350 for a discussion of cancer vaccines.

Examples of adjuvants that increase or expand the host's immune response to antigenic compounds include emulsifiers, muramyl dipeptides, avridine, aqueous adjuvants such as aluminum hydroxide, chitosan-based adjuvants, saponins, oils, amphoteric proteins, LPS, bacterial cell wall extracts, bacterial DNA, CpG sequences, synthetic oligonucleotides, cytokines and combinations thereof. Emulsifiers include, for example, potassium, sodium and ammonium salts of lauric acid and oleic acid, calcium, magnesium and aluminum salts of fatty acids, organic sulfonates, such as sodium lauryl sulfate, cetyltriethylammonium bromide, glycerides, polyoxyethylene glycol esters and ethers, and sorbitan fatty acid esters and polyoxyethylene thereof, gum arabic, gelatin, lecithin and/or cholesterol. The auxiliary material comprising an oil component includes mineral oil, vegetable oil or animal oil. Other adjuvants include Freund's complete adjuvant (FCA) or Freund's incomplete adjuvant (FIA). Cytokines useful as additional immunostimulants include interferon alpha, interleukin-2 (IL-2), and granulocyte macrophage colony stimulating factor (GM-CSF), or a combination thereof.

The concentration of the peptides described herein in the vaccine or immunogenic formulation can vary widely, i.e., from less than about 0.1%, typically at or at least about 2% to as high as 20% to 50% or more by weight, and is selected primarily based on liquid volume, viscosity, etc., depending on the particular mode of administration chosen.

The peptides described herein can also be administered via liposomes, which target the peptides to specific cell tissues, such as lymphoid tissue. Liposomes can also be used to increase the half-life of the peptides. Liposomes include emulsions, foams, micelles, insoluble monolayers, liquid crystals, phospholipid dispersions, lamellar layers, etc. In these preparations, the peptide to be delivered is combined as part of the liposome alone or with molecules that bind to receptors that are ubiquitous in lymphocytes (e.g., monoclonal antibodies that bind to CD45 antigens), or with other therapeutic or immunogenic compositions. Therefore, liposomes filled with the desired peptides of the present invention can be directed to the lymphocyte site, where the liposomes subsequently deliver the selected therapeutic/immunogenic peptide composition. The liposomes used in the present invention are formed by standard vesicle-forming lipids, which typically include neutral and negatively charged phospholipids and sterols, such as cholesterol. The choice of lipids is typically based on considerations such as liposome size, acid instability, and stability of the liposomes in the blood. Various methods are available for preparing liposomes, such as Szoka et al., Ann. Rev. Biophys. Bioeng. 9; 467 (1980), U.S. Pat. Nos. 4,235,871, 4501728, 4,501,728, 4,837,028 and 5,019,369.

To target immune cells, the ligand to be incorporated into the liposomes may include, for example, antibodies or fragments thereof that are specific for cell surface determinants of the desired immune system cells. The liposomal suspension containing the peptide may be administered intravenously, topically, topically, etc., at doses that vary depending on the mode of administration, the peptide being delivered, and the stage of the disease to be treated.

For solid compositions, conventional or nanoparticle non-toxic solid carriers can be used, including, for example, pharmaceutical grade mannitol, lactose, starch, magnesium stearate, sodium saccharin, talc, cellulose, glucose, sucrose, magnesium carbonate, etc. For oral administration, a pharmaceutically acceptable non-toxic composition is formed by incorporating any commonly used excipients (e.g., the aforementioned carriers) and generally 10-95% of the active ingredient (i.e., one or more peptides of the present invention), more preferably 25%-75%.

For aerosol administration, the immunogenic peptide is preferably provided in a subdivided form together with a surfactant and a propellant. The usual percentage of the peptide is 0.01%-20% (by weight), preferably 1%-10%. The surfactant must of course be non-toxic and preferably soluble in the propellant. Representatives of this type of agent are esters or partial esters of fatty acids containing 6-22 carbon atoms, such as esters or partial esters of caproic acid, caprylic acid, dodecanoic acid, palmitic acid, stearic acid, linoleic acid, linolenic acid, cholesterol esters and oleic acid with aliphatic polyols or their cyclic anhydrides. Mixed esters, such as mixed or natural glycerides, can be used. The surfactant may account for 0.1%-20% of the weight of the composition, preferably 0.25%-5%. The balance of the composition is usually a propellant. A carrier may also be included as needed, such as lecithin for intranasal delivery.

Cytotoxic T cells (CTLs) recognize antigens in the form of peptides bound to MHC molecules, rather than the complete foreign antigens themselves. The MHC molecules themselves are located on the cell surface of antigen presenting cells. Therefore, CTLs can only be activated when there is a trimeric complex of peptide antigens, MHC molecules and antigen presenting cells (APCs). Accordingly, if not only the peptide is used to activate CTLs, but also APCs with respective MHC molecules are additionally added, the immune response may be enhanced. Therefore, in some embodiments, the vaccine or immunogenic composition according to the present disclosure alternatively or additionally comprises at least one antigen presenting cell, preferably an APC population.

Thus, the vaccine or immunogenic composition can be delivered in the form of cells, such as antigen presenting cells, such as dendritic cell vaccines. Antigen presenting cells (e.g., dendritic cells) can be pulsed or loaded with the neoantigenic peptides disclosed herein, can contain expression constructs encoding the neoantigenic peptides disclosed herein, or can be genetically modified (by DNA or RNA transfer) to express one, two or more neoantigenic peptides disclosed herein, such as at least 2, 3, 4, 5, 6, 7, 8, 9 or 10 neoantigenic peptides.

Suitable vaccines or immunogenic compositions can also be in the form of DNA or RNA associated with the neoantigenic peptides described herein. For example, DNA or RNA encoding one or more neoantigenic peptides or proteins derived therefrom can be used as a vaccine, for example, by direct injection into a subject. For example, DNA or RNA encoding at least 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24 or 25 neoantigenic peptides or proteins derived therefrom.

Many methods are conveniently used to deliver nucleic acids to patients. For example, nucleic acids can be delivered directly as "naked DNA". For example, this method is described in Wolff et al., Science 247: 1465-1468 (1990) and U.S. Patent Nos. 5,580,859 and 5,589,466. Nucleic acids can also be administered using ballistic delivery, such as that described in U.S. Patent No. 5,204,253. Particles consisting only of DNA can be administered. Alternatively, the DNA can be attached to particles, such as gold particles.

Nucleic acid can also be delivered in combination with a cationic compound (e.g., a cationic lipid). For example, lipid-mediated gene delivery methods are described in WO9618372, WO9324640, Mannino & Gould-Fogerite, BioTechniques 6 (7): 682-691 (1988), U.S. Patent No. 5279833, WO9106309, and Felgner et al., Proc. Natl. Acad. Sci. USA 84: 7413-7414 (1987).

Delivery system optionally includes cell permeation peptide, nanoparticle encapsulation, virus-like particle, liposome or any combination thereof. Cell permeation peptide includes TAT peptide, herpes simplex virus VP22, transmembrane peptide (transportan), Antp. Liposome can be used as delivery system. Listeria vaccine or electroporation can also be used.

One or more new antigenic peptides can also be delivered by a bacterial or viral vector comprising a DNA or RNA sequence encoding one or more new antigenic peptides. DNA or RNA can be delivered as a vector itself or in an attenuated bacterial virus or attenuated live virus (such as cowpox or fowlpox). This method involves using cowpox virus as a vector to express a nucleotide sequence encoding the peptide of the present invention. After being introduced into an acutely or chronically infected host or a non-infected host, the recombinant cowpox virus expresses an immunogenic peptide, thereby eliciting a host CTL response. Cowpox vectors and methods that can be used for immunization protocols are described, for example, in U.S. Pat. No. 4,722,848. Another vector is BCG (Bacille Calmette Guerin, BCG). Stover et al. (Nature 351: 456-460 (1991) describes a BCG vector. Based on the description herein, a person skilled in the art will be aware of a variety of other vectors that can be used for therapeutic administration or immunization of the peptides of the present invention, such as Salmonella typhimurium, etc.

Suitable methods for administering nucleic acids encoding peptides described herein include the use of minigene constructs encoding multiple epitopes. In order to generate a DNA sequence encoding a selected CTL epitope (minigene) for expression in human cells, the amino acid sequence of the epitope is reverse translated. The human codon usage table is used to guide the codon selection for each amino acid. These DNA sequences encoding epitopes are directly adjacent to form a continuous polypeptide sequence. In order to optimize expression and/or immunogenicity, additional elements can be incorporated into the minigene design. Examples of amino acid sequences that can be reverse translated and included in the minigene sequence include: helper T lymphocytes, epitopes, leader (signal) sequences, and endoplasmic reticulum retention signals. In addition, by including synthetic (e.g., polyalanine) or naturally occurring flanking sequences near the CTL epitope, the presentation of the CTL epitope on the MHC can be improved.

The minigene sequence is converted to DNA by assembling oligonucleotides encoding the positive and negative strands of the minigene. Overlapping oligonucleotides (30-100 bases long) are synthesized, phosphorylated, purified, and annealed under appropriate conditions using well-known techniques. The ends of the oligonucleotides are ligated using T4 DNA ligase. This synthetic minigene encoding the CTL epitope polypeptide can then be cloned into the desired expression vector.

The vector contains standard regulatory sequences well known to those skilled in the art to ensure expression in target cells. Therefore, the DNA or RNA encoding one or more new antigenic peptides is usually operably linked to one or more of the following:

-Promoters that can be used to drive expression of nucleic acid molecules. AAV ITR can be used as a promoter and is conducive to eliminating the need for additional promoter elements. For ubiquitous expression, the following promoters can be used: CMV (particularly human cytomegalovirus immediate early promoter (hCMV-IE)), CAG, CBh, PGK, SV40, RSV, ferritin heavy chain or light chain, etc. For brain expression, the following promoters can be used: Synapsinl is used for all neurons, CaMKIIalpha is used for excitatory neurons, GAD67 or GAD65 or VGAT is used for γ-aminobutyric acid neurons, etc. Promoters for driving RNA synthesis may include: PolIII promoters, such as U6 or HI. The use of Pol II promoters and intron boxes can be used to express guide RNA (gRNA). Typically, the promoter includes a downstream cloning site for minigene insertion. For examples of suitable promoter sequences, see in particular U.S. Patent Nos. 5,580,859 and 5,589,466.

- Transcriptional transactivators or other enhancer elements may also increase transcriptional activity, such as the regulatory R region of the 5′ long terminal repeat (LTR) of human T-cell leukemia virus type 1 (HTLV-1), which has been shown to induce higher cellular immune responses when bound to the CMV promoter.

- Translation optimization sequences, such as the Kozak sequence flanking the AUG start codon (ACCAUGG) within the mRNA, and codon optimization.

Additional vector modifications may be required to optimize minigene expression and immunogenicity. In some cases, introns are required for efficient gene expression, and one or more synthetic or naturally occurring introns can be integrated into the transcribed region of the minigene. The addition of mRNA stabilizing sequences to increase minigene expression can also be considered. Immunostimulatory sequences (ISS or CpG) have recently been proposed to play a role in the immunogenicity of DNA vaccines. If these sequences are found to enhance immunogenicity, they can be included in the vector, outside the minigene coding sequence.

In some embodiments, a bicistronic expression vector can be used to allow production of both the minigene-encoded epitope and a second protein that enhances or reduces immunogenicity.

DNA vaccines or immunogenic compositions described herein can be enhanced by co-delivery of cytokines (e.g., IL-2, IL-12, IL-18, GM-CSF, and IFNγ) that promote cell-mediated immune responses. CXC chemokines (e.g., IL-8) and CC chemokines (e.g., macrophage inflammatory protein (MIP) -1α, MIP-3α, MIP-3β, and RANTES) can increase the effectiveness of immune responses. The immunogenicity of DNA vaccines can also be enhanced by co-delivery of plasmid-encoded cytokine induction molecules (e.g., LeIF), costimulatory molecules, and adhesion molecules (e.g., B7-1 (CD80) and/or B7-2 (CD86)). Auxiliary (HTL) epitopes can be connected to intracellular targeting signals and expressed separately from CTL epitopes. This will allow HTL epitopes to be directed to cell compartments different from CTL epitopes. If desired, this can promote HTL epitopes to enter MHC class II pathways more effectively, thereby improving CTL induction. In contrast to CTL induction, reduction of immune responses by co-expression of immunosuppressive molecules (eg, TGF-β) may be beneficial in certain diseases.

Once the expression vector is selected, the minigene is cloned into the polylinker region downstream of the promoter. The plasmid is transformed into an appropriate strain of E. coli and DNA is prepared using standard techniques. The orientation and DNA sequence of the minigene and all other elements contained in the vector are confirmed by restriction mapping and DNA sequence analysis. Bacterial cells containing the correct plasmid can be stored as master cell banks and working cell banks.

Purified plasmid DNA for injection can be prepared using a variety of formulations. The simplest of these is to reconstitute the lyophilized DNA in sterile phosphate buffered saline (PBS). A variety of methods have been described, and new technologies may emerge. As described above, nucleic acids can be conveniently formulated with cationic lipids. In addition, glycolipids, fusogenic liposomes, peptides and compounds, collectively referred to as protective, interactive, non-condensing (PINC), can also be complexed with purified plasmid DNA to affect variables such as stability, intramuscular dispersion or transport to specific organs or cell types.

Vaccines or immunogenic compositions comprising peptides can be administered in combination with vaccines or immunogenic compositions comprising polynucleotides encoding the peptides. For example, administration of peptide vaccines and DNA vaccines can be alternating in a prime-boost regimen. For example, it is contemplated that priming with a peptide immunogenic composition and boosting with a DNA immunogenic composition is as likely as priming with a DNA immunogenic composition and boosting with a peptide immunogenic composition.

The present disclosure also includes a method of producing a vaccine composition comprising the steps of:

a) optionally, identifying at least one neoantigenic peptide according to the aforementioned method;

b) producing the at least one neo-antigenic peptide, at least one polypeptide encoding the neo-antigenic peptide, or at least one vector comprising a polypeptide as described herein; and

c) optionally adding a physiologically acceptable buffer, excipient and/or adjuvant and producing a vaccine using the at least one neoantigenic peptide, polypeptide or vector.

Another aspect of the present disclosure is a method of producing a DC vaccine, wherein the DCs are presented with at least one neoantigenic peptide of the present disclosure.

Antibody TCR, CAR and their derivatives

The present disclosure also relates to an antibody or antigen-binding fragment thereof that specifically binds to a chimeric polypeptide (or protein) as disclosed herein (most preferably a chimeric protein of SEQ ID NO: 1-8202) or a neoantigenic peptide that is normally bound to an MHC or HLA molecule.

In some embodiments, the antibody or antigen-binding fragment thereof comprises or consists of an antigen-binding domain (which binds to a chimeric protein) as defined above.

Typically, the antibody or antigen-binding fragment thereof binds to a neoantigenic peptide, which typically binds to an MHC or HLA molecule or (transmembrane) chimeric protein as defined above with a dissociation constant ( _Kd ) of about 2× ^10-7 M or less. In certain embodiments, the _Kd is about 2× ^10-7 M or less, about 1× ^10-7 M or less, about 9× ^10-8 M or less, about 9× ^10-9 M or less, about 5× ^10-9 M or less, about 4× ^10-9 M or less, about 3× ^10-9 M or less, about 2× ^10-9 M or less, or about 1× ^10-9 M or less. In certain non-limiting embodiments, the _Kd is about 1× ^10-9 M to about 3× ^10-7 M. In certain non-limiting embodiments, the _Kd is about 1.5× ^10-9 M to about 3× ^10-7 M. In certain non-limiting embodiments, _Kd is from about 1.5 x 10 ^"9 M to about 2.7 x 10 ^"7 M.

In order to promote the penetration and recognition of tumor cells by lymphocytes T (LT), another strategy is to use antibodies that can recognize more than one antigen target at the same time, especially two antigen targets at the same time. Bispecific antibodies come in many forms. BiTE (bispecific T cell engager) was the first to be developed. These fusion proteins consist of two scFvs (heavy chain variable domain VH and light chain variable domain VL) of two antibodies, which are connected by a binding peptide: one recognizes the LT marker (CD3+) and the other recognizes the tumor antigen. The purpose is to promote the recruitment and activation of LTs in contact with the tumor, thereby leading to tumor cell lysis (see review Patrick A. Baeuerle and Carsten Reinhardt; Bispecific T-Cell Engaging Antibodies for Cancer Therapy, Cancer Res 2009; 69: (12). June 15, 2009 and Galaine et al., Innovations & Therapeutiques en Oncologie, vol. 3-n°3-7, 2017).

Thus, in a particular embodiment, the antibody is a bispecific T cell engager targeting a chimeric protein as defined herein, and in particular comprises an antigen binding domain as defined previously.

The term "antibody" is used in the broadest sense herein, including polyclonal and monoclonal antibodies, including complete antibodies and functional (antigen-binding) antibody fragments, including fragment antigen-binding (Fab) fragments, F(ab')2 fragments, Fab' fragments, Fv fragments, recombinant IgG fragments, variable heavy chain (VH) regions capable of specific binding to antigens, single-chain antibody fragments, including single-chain variable fragments (scFv) and single-domain antibodies (e.g., VHH antibodies, sdAbs, sdFvs, nanobodies) fragments. The term includes immunoglobulins in genetically engineered and/or other variant-modified forms, such as intrabodies, peptide antibodies, chimeric antibodies, fully human antibodies, humanized antibodies and heteroconjugate antibodies, multispecific (e.g., bispecific) antibodies, antibodies, double antibodies, three antibodies and four antibodies, tandem double scFvs, tandem three scFvs. Unless otherwise indicated, the term "antibody" should be understood to include functional antibodies and fragments thereof. The term also includes intact or full-length antibodies, including antibodies of any class or subclass, including IgG and its subclasses, IgG1, IgG2, IgG3, IgG4, IgM, IgE, IgA and IgD. In some embodiments, the antibody includes a light chain variable region and a heavy chain variable region, for example in the form of an scFv.

Antibodies include variant polypeptide forms having one or more amino acid substitutions, insertions or deletions in the native amino acid sequence, provided that the antibody retains or substantially retains its specific binding function. Conservative substitutions of amino acids are well known and described above.

The present disclosure further includes a method of producing an antibody or antigen-binding fragment thereof, comprising the step of selecting an antibody that binds to a tumor neoantigenic peptide as defined herein, said antibody typically binding to an MHC or HLA molecule or chimeric protein as defined herein with a dissociation constant ( _Kd ) of about 2× ^10-7 M or less. In certain embodiments, the Kd is about 2× ^10-7 M or less, about 1× ^10-7 M or less, about 9× ^10-8 M or less, about 1× ^10-8 M or less, about 9× ^10-9 M or less, about 5× ^10-9 M or less, about 4× ^10-9 M or less, about 3× ^10-9 M or less, about 2× ^10-9 M or less, or about 1× ^10-9 M or less, or about 1× ^10-10 M or less, or about 1× ^10-12 M or less.

In certain embodiments, the antibodies are of murine, human, or camelid (eg, llama) origin.

In some embodiments, the antibody is selected from a library of human antibody sequences. In some embodiments, the antibody is produced by immunizing an animal with a chimeric protein of any one of SEQ ID NOs: 1-8202 as defined above, or a portion thereof (particularly an extracellular portion), followed by a selection step.

Antibodies including chimeric antibodies, humanized antibodies or human antibodies may be further affinity matured and selected as described above. Humanized antibodies contain CDR regions derived from rodent sequences; generally, rodent CDRs are implanted into human frameworks, and some human framework residues may be reverse mutated to original rodent framework residues to maintain affinity, and/or one or more CDR residues may be mutated to increase affinity. Fully human antibodies do not contain mouse sequences and are usually produced by phage display technology of human antibody libraries or immunization of transgenic mice whose natural immunoglobulin loci have been replaced by fragments of human immunoglobulin loci.

Antibodies produced by the method and immune cells expressing the antibodies or fragments thereof are also included in the present disclosure.

The present disclosure also includes pharmaceutical compositions comprising one or more antibodies disclosed herein alone or in combination with at least one other agent (e.g., a stabilizing compound), which can be administered in any sterile, biocompatible pharmaceutical carrier, and optionally formulated with sterile pharmaceutically acceptable buffers, diluents, and/or excipients. Pharmaceutically acceptable carriers are typically enhanced or stabilized compositions, and/or can be used to promote the preparation of compositions. Pharmaceutically acceptable carriers include solvents, dispersion media, coatings, antibacterial and antifungal agents, isotonic agents, and absorption delay agents that are physiologically compatible and, in some embodiments, pharmaceutically inert.

Administration of the pharmaceutical compositions comprising the antibodies disclosed herein can be accomplished orally or parenterally. Parenteral methods include topical, intra-arterial (directly to the tumor), intramuscular, spinal, subcutaneous, intramedullary, intrathecal, intraventricular, intravenous, intraperitoneal, or intranasal administration.

Thus, in addition to the active ingredient, these pharmaceutical compositions may contain a suitable pharmaceutically acceptable carrier comprising excipients and adjuvants which facilitate processing of the active compound into pharmaceutically acceptable preparations. Further details of techniques for formulation and administration may be found in the latest edition of Remington's Pharmaceutical Sciences (Ed. Maack Publishing Co, Easton, Pa.).

Depending on the route of administration, the active compounds (ie, antibodies, bispecific and multispecific molecules) can be coated in a material to protect the compound from acids and other natural conditions that might render the compound inactive.

Said composition is normally sterile, preferably fluid.Suitable fluidity can be maintained by, for example, using a coating (e.g., lecithin), maintaining the required particle size in the case of a dispersion, and using a surfactant. In many cases, it is preferred to include an isotonic agent in the composition, such as sugar, a polyol (e.g., mannitol or sorbitol) and sodium chloride. The long-term absorption of the injectable composition can be achieved by including an agent that delays absorption in the composition, such as aluminum monostearate or gelatin.

Pharmaceutical compositions for oral administration can be formulated with pharmaceutically acceptable carriers well known in the art in dosages suitable for oral administration. Such carriers enable the pharmaceutical compositions to be formulated into tablets, pills, dragees, capsules, liquids, gels, syrups, slurries, suspensions, etc. for consumption by patients.

Pharmaceutical preparations for oral administration can be obtained by combining the active compound with a solid excipient, optionally grinding the resulting mixture, and processing the granular mixture after adding suitable excipients (if necessary) to obtain tablets or dragee cores. Suitable excipients are carbohydrate or protein fillers, such as sugars, including lactose, sucrose, mannitol or sorbitol; starch from corn, wheat, rice, potato or other plants; cellulose, such as methyl, cellulose, hydroxypropylmethylcellulose or sodium carboxymethylcellulose; and gums, including gum arabic and tragacanth; and proteins, such as gelatin and collagen. If necessary, disintegrants or solubilizers, such as cross-linked polyvinyl pyrrolidone, agar, alginic acid or its salts, such as sodium alginate, can be added.

Dragee cores are provided with suitable coatings, for example concentrated sugar solutions, which may also contain gum arabic, talc, polyvinyl pyrrolidone, carbopol gel, polyethylene glycol and/or titanium dioxide, lacquer solutions and suitable organic solvents or solvent mixtures. Dyes or pigments may be added to the tablets or dragee coatings for product identification or to characterize the amount of active compound, i.e., the dosage.

Pharmaceutical preparations that can be used orally include push-fit capsules made of gelatin, as well as soft, sealed capsules made of gelatin and coatings, such as glycerol or sorbitol. The push-fit capsules may contain the active ingredients mixed with fillers or binders, such as lactose or starches, lubricants, such as talc or magnesium stearate, and optionally stabilizers. In soft capsules, the active compounds may be dissolved or suspended in suitable liquids, such as fatty oils, liquid paraffin, or liquid polyethylene glycols, with or without stabilizers.

Pharmaceutical preparations for parenteral administration include aqueous solutions of the active compound. For injection, the pharmaceutical composition of the present invention can be prepared in an aqueous solution, preferably in a physiologically compatible buffer, such as Hank's solution, Ringer's solution or physiologically buffered saline. Aqueous injection suspensions may contain substances that increase the viscosity of the suspension, such as sodium carboxymethylcellulose, sorbitol or dextran. In addition, the suspension of the active compound can be prepared into a suitable oily injection suspension. Suitable lipophilic solvents or carriers include fatty oils (such as sesame oil) or synthetic fatty acid esters (such as ethyl oleate or triglycerides) or liposomes. Optionally, the suspension may also include a suitable stabilizer or an agent that can increase the solubility of the compound to prepare a highly concentrated solution.

For topical or nasal administration, penetrants appropriate to the particular barrier to be permeated are used in the formulation. Such penetrants are well known in the art.

The pharmaceutical compositions of the present disclosure can be prepared according to methods well known and routinely practiced in the art. For example, Remington: The Science and Practice of Pharmacy, Mack Publishing Co., 20th ed., 2000; and Sustained and Controlled Release Drug Delivery Systems, J R. Robinson, ed., Marcel Dekker, Inc., New York, 1978. The pharmaceutical compositions are preferably prepared under GMP conditions.

The present disclosure also encompasses recombinant T cell receptors (TCRs) targeting neoantigenic peptides as defined herein bound to MHC or HLA molecules.

The present disclosure further includes a method of producing a TCR or an antigen-binding fragment thereof, the method comprising the step of selecting a TCR that binds to a tumor neoantigenic peptide as defined herein, optionally bound to an MHC or HLA molecule, with a dissociation constant ( _Kd ) of about 2× ^10-7 M or less. In certain embodiments, the _Kd is about 2× ^10-7 M or less, about 1× ^10-7 M or less, about 9× ^10-8 M or less ^, about 1× ^10-8 M or less, about 9× ^10-9 M or less, about 5× ^10-9 M or less, about 4× ^10-9 M or less, about 3× ^10-9 M or less, about 2× ^10-9 M or less, or about 1× ^10-9 M or less, or about 1× ^10-10 M or less, or about 1× ^10-12 M or less.

Nucleic acids encoding TCRs can be obtained from a variety of sources, such as by amplifying naturally occurring TCR DNA sequences by polymerase chain reaction (PCR), followed by expression of antibody variable regions, followed by the above-described selection steps. In some embodiments, TCRs are obtained from T cells isolated from patients or cultured T cell hybridomas. In some embodiments, TCR clones for target antigens have been generated in transgenic mice engineered with human immune system genes (e.g., HLA system or HLA). See, for example, tumor antigens (see, for example, Parkhurst et al. (2009) Clin Cancer Res. 15: 169-180 and Cohen et al. (2005) J Immunol. 175: 5799-5808. In some embodiments, phage display is used to isolate TCRs for target antigens (see, for example, Varela-Rohena et al. (2008) Nat Med. 14: 1390-1395 and Li (2005) Nat Biotechnol. 23: 349-354).

"T cell receptor" or "TCR" refers to a molecule that comprises a variable α chain and a β chain (referred to as TCRRa and TCRb, respectively) or a variable γ chain and a δ chain (referred to as TCRg and TCRd, respectively) and is capable of specifically binding to an antigenic peptide bound to an MHC receptor. In some embodiments, the TCR is in the αβ form. Typically, TCRs in the αβ and γδ forms are roughly similar in structure, but the T cells expressing them may have different anatomical locations or functions. TCRs can be found on the cell surface or in a soluble form. Typically, TCRs are located on the surface of T cells (or T lymphocytes), which are generally responsible for recognizing antigens that are bound to major histocompatibility complex (MHC) molecules through their extracellular binding domains. In some embodiments, the TCR may also include a constant domain, a transmembrane domain, and/or a short cytoplasmic tail (see, e.g., Janeway et ah, Immunobiology: The Immune System in Health and Disease, 3rd Ed., Current Biology Publications, p. 4: 33, 1997). For example, in some aspects, each chain of a TCR may have an N-terminal immunoglobulin variable domain, an immunoglobulin constant domain, a transmembrane region, and a short cytoplasmic tail at the C-terminus. In some embodiments, the TCR binds to a constant protein of a CD3 complex involved in mediating signal transduction. Unless otherwise indicated, the term "TCR" is understood to include its functional TCR fragments. The term also includes complete or full-length TCRs, including TCRs in αβ or γδ forms.

Therefore, for the purposes of this article, reference to a TCR includes any TCR or functional fragment, such as an antigen binding portion of a TCR, which binds to a specific antigen peptide bound to an MHC molecule, i.e., an MHC peptide complex. "Antigen binding portion" or "antigen binding fragment" of a TCR is used interchangeably and refers to a molecule that contains a portion of a domain of a TCR but binds to an antigen (e.g., an MHC peptide complex) bound to the full TCR. In some cases, the antigen binding portion comprises a variable domain of a TCR, such as a variable α chain and a variable β chain of a TCR, which is sufficient to form a binding site that binds to a specific MHC peptide complex, such as typically each chain comprising three complementary determining regions.

In some embodiments, the variable domains of the TCR chains combine to form a loop or complementary determining region (CDR) similar to that of immunoglobulins, which confers antigen recognition and determines the specificity of the peptide by forming the binding site of the TCR molecule. As with immunoglobulins, CDRs are usually separated by framework regions (FRs) (see Jores et al., Pwc. Nat'l Acad. Sci. U.S.A. 87:9138, 1990; Chothia et al., EMBO J. 7:3745, 1988, see also Lefranc et al., Dev. Comp. Immunol. 27:55, 2003). In some embodiments, CDR3 is the primary CDR responsible for recognizing the processed antigen, although it has also been shown that CDR1 of the α chain interacts with the N-terminal portion of the antigenic peptide, while CDR1 of the β chain interacts with the C-terminal portion of the peptide. CDR2 is believed to be able to recognize MHC molecules. In some embodiments, the variable region of the β chain may contain a further high variability (HV4) region.

In some embodiments, the TCR chain comprises a constant domain. For example, as with immunoglobulins, the extracellular portion of the TCR chain (e.g., α chain, β chain) can comprise two immunoglobulin domains, a variable domain (e.g., Va or Vp; typically 1-116 amino acids according to Kabat numbering, Kabat et al., "Sequences of Proteins of Immunological Interest, US Dept. Health and Human Services, Public Health Service National Institutes of Health, 1991, 5th ed.) and a constant domain adjacent to the cell membrane (e.g., an alpha chain constant domain or Ca, typically 117-259 amino acids based on Kabat, a beta chain constant domain or Cp, typically 117-295 amino acids based on Kabat). For example, in some cases, the extracellular portion of the TCR formed by the two chains comprises two constant domains proximal to the membrane and two variable domains comprising CDRs distal to the membrane. The constant domain of the TCR domain comprises a short linker sequence in which cysteine residues form a disulfide bond to form a connection between the two chains. In some embodiments, the TCR may have additional cysteine residues on each of the alpha and beta chains such that the TCR comprises two disulfide bonds in the constant domain.

In some embodiments, the TCR chain may include a transmembrane domain. In some embodiments, the transmembrane domain is positively charged. In some cases, the TCR chain includes a cytoplasmic tail. In some cases, the structure allows TCR to be combined with other molecules (such as CD3). For example, a TCR comprising a constant domain with a transmembrane region can anchor the protein to the cell membrane and bind to the constant subunit of a CD3 signaling device or a complex.

Typically, CD3 is a multiprotein complex that can have three different chains (γ, δ, and ε) in mammals and a ζ chain. For example, in mammals, the complex can contain a homodimer of a CD3y chain, a CD35 chain, two CD3 chains, and a CD3ζ chain. The CD3y, CD35, and CD3 chains are highly related cell surface proteins of the immunoglobulin superfamily, containing a single immunoglobulin domain. The transmembrane regions of the CD3y, CD35, and CD3 chains are negatively charged, a feature that allows these chains to bind to positively charged T cell receptor chains. The intracellular tails of the CD3y, CD35, and CD3 chains each contain a conserved motif, called an immunoreceptor tyrosine-based activation motif or ITAM, and each CD3ζ chain has three conserved motifs. In general, ITAMs are involved in the signaling ability of the TCR complex. These accessory molecules have negatively charged transmembrane regions and play a role in propagating signals from the TCR to the cell. The CD3- and ζ- chains form the T cell receptor complex together with the TCR.

In some embodiments, TCR can be a heterodimer of two chains α and β (or optional γ and δ), or can be a single-chain TCR construct. In some embodiments, TCR is a heterodimer comprising two separate chains (α and β chains or γ and δ chains) connected by, for example, a disulfide bond or a disulfide bond.

Although T cell receptors (TCRs) are transmembrane proteins and do not exist naturally in soluble form, antibodies can be secreted and membrane-bound. Importantly, the advantage of TCRs over antibodies is that, when presented in the context of MHC molecules, they can in principle recognize peptides generated by all degraded intracellular and extracellular proteins. Therefore, TCRs have important therapeutic potential.

The present disclosure also relates to soluble T cell receptors (sTCRs) comprising an antigen recognition portion directed against a tumor neoantigenic peptide as disclosed herein (see in particular Walseng E, S, Fallang LE, Yang W, Vefferstad A, Areffard A, et al. (2015) Soluble T-Cell Receptors Produced in Human Cells for Targeted Delivery. PLoS ONE 10 (4): e0119559). In a specific embodiment, the soluble TCR can be fused with an antibody fragment against a T cell antigen, optionally wherein the targeted antigen is CD3 or CD16 (see, e.g., Boudousquie, Caroline et al. "Polyfunctional response by ImmTAC (IMCgp100) redirected CD8+ and CD4+ T cells." Immunology vol. 152, 3 (2017): 425-438. doi: 10.1111 / imm.12779).

In certain embodiments, the present disclosure includes a recombinant HLA-independent (or non-HLA-restricted) T cell receptor (referred to as "HI-TCR") that binds to a (transmembrane) chimeric protein as defined herein (particularly a new antigenic peptide as defined above in any one of SEQ ID NOs: 1-8202). The intended "HI-TCR" that is very suitable for the present invention is described in International Application No. WO2019/157454. Therefore, the HI-TCR according to the present disclosure generally comprises an antigen binding chain comprising: (a) an antigen binding domain (as defined above) that binds to an antigen in an HLA-independent manner, such as an antigen binding fragment of an immunoglobulin variable region; and (b) a constant domain that is capable of binding to (and thus activating) a CD3ζ polypeptide. Because TCRs generally bind to antigens in an HLA-dependent manner, the antigen binding domain that binds in an HLA-independent manner is heterologous. Preferably, the antigen binding domain or fragment thereof comprises: (i) an antigen binding domain comprising or consisting of a heavy chain variable region (VH) of an antibody and/or (ii) a light chain variable region (VL) of an antibody. The constant domain of the TCR is, for example, a natural or modified TRAC polypeptide, or a natural or modified TRBC polypeptide. The constant domain of the TCR is, for example, a natural TCR constant domain (α or β) or a fragment thereof. Unlike chimeric antigen receptors that typically contain intracellular signaling domains, HI-TCRs do not directly generate activation signals; instead, the antigen binding chain binds to the CD3ζ polypeptide and thereby activates the polypeptide. Immune cells comprising recombinant TCRs provide excellent activity when the density of antigen on the cell surface is less than about 10,000 molecules/cell.

The CD3 ζ polypeptide is, for example, a natural CD3 ζ polypeptide or a modified CD3 ζ polypeptide. The CD3 ζ polypeptide is optionally fused to the intracellular domain of a costimulatory molecule or a fragment thereof. Alternatively, the antigen binding domain optionally comprises a costimulatory region, such as an intracellular domain, which can stimulate immune responsive cells when the antigen binding chain binds to the antigen. Examples of costimulatory molecules include CD28, 4-1BB, OX40, ICOS, DAP-10, fragments thereof, or combinations thereof.

In some embodiments, the recombinant HI-TCR is expressed by a transgene integrated into an endogenous locus of an immune responsive cell, for example, a CD3 δ genome, a CD3 ε locus, a CD247 locus, a B2M locus, a TRAC locus, a TRBC locus, a TRDC locus, and/or a TRGC locus. In most embodiments, the expression of the recombinant HI-TCR is driven by an endogenous TRAC or TRBC locus. In some embodiments, a transgene encoding a portion of the recombinant HI-TCR is integrated into an endogenous TRAC and/or TRBC locus in a manner that destroys or eliminates the endogenous expression of a TCR comprising a natural TCR α chain and/or a natural TCR β chain. The destruction prevents or eliminates mismatches between the recombinant TCR and the natural TCR α chain and/or the natural TCR β chain in the immune responsive cell. The endogenous locus may also include a modified transcription terminator region, for example, a TK transcription terminator, a GCSF transcription terminator, a TCRA transcription terminator, an HBB transcription terminator, a bovine growth hormone transcription terminator, an SV40 transcription terminator, and a P2A element.

In some embodiments of the present disclosure, the recombinant TCR and HI-TCR generally comprise an extracellular antigen binding domain that can dimerize with a second extracellular antigen binding domain. Generally, the second extracellular antigen binding domain binds to a tumor antigen, preferably wherein the tumor antigen is selected from pHER95, CD19, MUC16, MUC1, CAIX, CEA, CD8, CD7, CD10, CD20, CD22, CD30, CD70, CLL1, CD33, CD34, CD38, CD41, CD44, CD49f, CD56, CD74, CD133, CD138, EGP-2, EGP-40, EpCAM, Erb-B2, Erb-B3, Erb-B4, FBP, fetal acetylcholine receptor, folate receptor-a, GD2, GD3, HER-2, hTERT, IL-13R -a2, kappa-light chain, KDR, LeY, L1 cell adhesion molecule, MAGE-A1, mesothelin, MAGEA3, p53, MART1, GP100, proteinase 3 (PR1), tyrosinase, survivin, hTERT, EphA2, NKG2D ligand, NY-ESO-1, carcinoembryonic antigen (h5T4), PSCA, PSMA, ROR1, TAG-72, TAG-R2, WT-1, BCMA, CD123, CD44V6, NKCS1, EGF1R, EGFR-VIII, CD99, CD70, ADGRE2, CCR1, LILRB2, LILRB4, PRAME and ERBB.

The present disclosure also encompasses chimeric antigen receptors (CARs) for chimeric polypeptides (proteins) disclosed herein and, in particular, transmembrane chimeric proteins of any one of SEQ ID NOs: 1-8202 defined herein. In a preferred embodiment, CAR comprises an antigen binding domain as defined above. CAR is a fusion protein comprising an antigen binding domain, typically derived from an antibody linked to a TCR complex signaling domain. By selecting a suitable antigen binding domain, CAR can be used to direct immune cells (such as T cells or NK T cells) to tumor neoantigen peptides as previously defined.

The antigen binding domain of CAR is usually based on scFv (single-chain variable fragment) from antibodies. In addition to the extracellular antibody binding domain at the N-terminus, CAR may generally include a hinge domain as a spacer to extend the antigen binding domain from the plasma membrane of the immune effector cell it expresses; transmembrane (TM) domain; intracellular signaling domain (e.g., signaling domain or equivalent from the CD3 molecule ζ chain of the TCR complex); and optional one or more costimulatory domains, which can assist in the signaling or function of cells expressing CAR. Signaling domains from costimulatory molecules (including CD28, OX-40 (CD134) and 4-1BB (CD137)) are added alone (second generation) or in combination (third generation) to enhance the survival of CAR-modified T cells and increase their proliferation. Potential costimulatory domains also include ICOS-1, CD27, GITR and DAP10.

Therefore, CAR may include

(1) in its extracellular portion, one or more antigen binding molecules, such as one or more antigen binding fragments, domains or parts of antibodies, or one or more antibody variable domains, and/or antibody molecules, and generally one or more antigen binding domains as defined above.

(2) In its transmembrane portion, the transmembrane domain is derived from human T cell receptor-α or -β chain, CD3ζ chain, CD28, CD3-ε, CD45, CD4, CD5, CD8, CD9, CD16, CD22, CD33, CD37, CD64, CD80, CD86, CD134, CD137, ICOS, CD154 or GITR. In some embodiments, the transmembrane domain is derived from CD28, CD8 or CD3-ζ.

(3) one or more costimulatory domains, for example, costimulatory domains derived from human CD28, 4-1BB (CD137), ICOS-1, CD27, OX40 (CD137), DAP10 and GITR (AITR). In some embodiments, CAR comprises a costimulatory domain of CD28 and 4-1BB.

(4) An intracellular signaling domain comprising one or more ITAMs, for example, the intracellular signaling domain is CD3-ζ, or a variant thereof lacking one or two ITAMs (e.g., ITAM3 and ITAM2), or the intracellular signaling domain is derived from FcεRIγ.

CAR can be designed to recognize tumor neoantigen peptides alone or in combination with HLA or MHC molecules.

The parts used to bind antigens include three major categories, single-chain antibody fragments (scFvs) from antibodies, Fab's from libraries, or natural ligands that bind to their cognate receptors (for first-generation CARs). Successful examples of each of the above categories are specifically reported in Sadelain M, Brentjens R, Riviere I. The basic principles of chimeric antigen receptor (CAR) design. Cancer discovery. 2013; 3 (4): 388-398 (see Table 1 in particular) and are included in this application.

Antibodies include chimeric, humanized or human antibodies, and can be further affinity matured and selected as described above. Chimeric or humanized scFvs derived from rodent immunoglobulins (e.g., mice, rats) are generally used because they are easily derived from fully characterized monoclonal antibodies. Humanized antibodies include rodent sequence-derived CDR regions; generally, rodent CDRs are implanted into human frameworks, and some human framework residues may be reversely mutated to original rodent framework residues to maintain affinity, and/or one or more CDR residues may be mutated to improve affinity. Fully human antibodies do not contain mouse sequences and are generally produced by immunization of transgenic mice whose phage display technology of human antibody libraries or their natural immunoglobulin loci have been replaced by human immunoglobulin locus fragments. Antibody variants having one or more amino acid substitutions, insertions or deletions in the native amino acid sequence can be produced, wherein the antibody retains or substantially retains its specific binding function. Conservative substitutions of amino acids are well known and as described above. Other variants with improved affinity for antigens can also be produced.

Typically, CARs include a previously defined antigen-binding domain from an antibody molecule, such as a single-chain antibody fragment (scFv) derived from the variable heavy (VH) and variable light (VL) chains of a monoclonal antibody (mAb).

In some aspects, the antigen binding domain of CAR is connected to one or more transmembrane and intracellular signaling domains. In some embodiments, CAR includes a transmembrane domain fused to the extracellular region of CAR. In one embodiment, a transmembrane domain naturally bound to a domain in CAR is used. In some cases, the transmembrane domain is selected or modified by amino acid substitution to avoid binding of the domain to the transmembrane domain of the same or different surface membrane proteins, thereby minimizing the interaction with other members of the receptor complex.

In some embodiments, the transmembrane domain is derived from a natural or synthetic source. If the source is natural, the domain can be derived from any membrane-bound or transmembrane protein. The transmembrane region includes a transmembrane region derived from (i.e., at least comprising) α, β or ζ chains of a T cell receptor, CD28, CD3ε, CD45, CD4, CD5, CD8, CD9, CD16, CD22, CD33, CD37, CD64, CD80, CD86, CD 134, CD137, CD154, ICOS or GITR. The transmembrane domain can also be synthetic. In some embodiments, the transmembrane domain is derived from CD28, CD8 or CD3-ζ.

In some embodiments, a short oligo- or polypeptide linker is present, e.g., a linker of 2 to 10 amino acids in length, and forms a connection between the transmembrane domain and the cytoplasmic signaling domain of the CAR.

CAR generally includes at least one intracellular signaling component.The first generation CAR generally has an intracellular region from CD3ζ chains, which is the main transmitter of the signal from endogenous TCR.The second generation CAR generally further includes an intracellular signaling domain from various costimulatory protein receptors (e.g., CD28, 41BB (CD28), ICOS) to the cytoplasmic tail of CAR, to provide additional signals to T cells.Co-stimulatory domains include domains derived from human CD28, 4-1BB (CD137), ICOS-1, CD27, OX 40 (CD137), DAP10 and GITR (AITR).Considering the combination of two costimulatory domains, such as CD28 and 4-1BB, or CD28 and OX40.The third generation CAR combines a variety of signaling domains, such as CD3z-CD28-4-1BB or CD3z-CD28-OX40, to enhance effectiveness.

The intracellular signaling domain can be derived from the intracellular component of the TCR complex, such as the TCR CD3+ chain that mediates T cell activation and cytotoxicity, such as the CD3ζ chain. Other intracellular signaling domains include FcεRIγ. The intracellular signaling domain may comprise a modified CD3ζ polypeptide that lacks one or two of its three immunoreceptor tyrosine-based activation motifs (ITAMs), wherein ITAMs are ITAM1, ITAM2, and ITAM3 (numbered from the N-terminus to the C-terminus). The intracellular signaling region of CD3-ζ is residues 22-164 of SEQ ID NO: 4. ITAM1 is located around amino acid residues 61-89, ITAM2 is located around amino acid residues 100-128, and ITAM3 is located around residues 131-159. Thus, the modified CD3ζ polypeptide may inactivate any of ITAM1, ITAM2, or ITAM3. Alternatively, the modified CD3 zeta polypeptide may inactivate any two ITAMs, such as ITAM2 and ITAM3, or ITAM1 and ITAM2. Preferably, ITAM3 is inactivated, such as deleted. More preferably, ITAM2 and ITAM3 are inactivated, such as deleted, and ITAM1 is retained. For example, a modified CD3 zeta polypeptide retains only ITAM1, and the remaining CD3 zeta domain is deleted (residues 90-164). As another example, ITAM1 is replaced by the amino acid sequence of ITAM3, and the remaining CD3 zeta domain is deleted (residues 90-164). For example, see Bridgeman et al., Clin. Exp. Immunol. 175(2): 258-67 (2014), Zhao et al., J. Immunol. 183(9): 5563-74 (2009), Maus et al., WO 2018/132506; Sadelain et al., WO/2019/133969, Feucht et al., Nat Med. 25(1): 82-88 (2019).

Therefore, in some aspects, the antigen binding domain is connected to one or more cell signaling modules. In some embodiments, the cell signaling module includes a CD3 transmembrane domain, a CD3 intracellular signaling domain, and/or other CD transmembrane domains. CAR may also further include a portion of one or more additional molecules, such as Fc receptor gamma, CD8, CD4, CD25, or CD16.

In some embodiments, after CAR connection, the cytoplasmic domain or intracellular signaling domain of CAR activates at least one normal effector function or response of the corresponding non-engineered immune cell (usually T cell).For example, CAR can induce the function of T cells, such as cytolytic activity or T helper activity, secretion of cytokines or other factors.

In some embodiments, the intracellular signaling domain comprises a cytoplasmic sequence of a T cell receptor (TCR), and in some aspects also comprises a cytoplasmic sequence of a co-receptor that acts synergistically with the receptor in its natural environment to initiate signal transduction following antigen-specific receptor binding, and/or a variant of that molecule, and/or any synthetic sequence having the same functional capability.

T cell activation is described in some aspects as being mediated by two types of cytoplasmic signaling sequences: antigen-dependent primary activation (primary cytoplasmic signaling sequence) is initiated by TCR, and secondary or costimulatory signals (secondary cytoplasmic signaling sequence) are provided in an antigen-independent manner. In some aspects, CAR includes one or two such signaling components.

In some aspects, CAR includes a primary cytoplasmic signaling sequence that regulates the primary activation of the TCR complex in a stimulatory or inhibitory manner. The primary cytoplasmic signaling sequence that acts in a stimulatory manner may include a signaling motif, which is referred to as an immunoreceptor tyrosine-based activation motif or ITAM. Examples of ITAMs comprising primary cytoplasmic signaling sequences include sequences derived from TCR ζ, FcR γ, FcR β, CD3 γ, CD3 δ, CD3 ε, CDS, CD22, CD79a, CD79b, and CD66d. In some embodiments, the cytoplasmic signaling molecule in CAR includes a cytoplasmic signaling domain, a portion thereof, or a sequence derived from CD3 ζ.

CAR can also include the signaling domain and/or transmembrane portion of a co-stimulatory receptor, such as CD28, 4-1BB, OX40, DAP10, and ICOS. In some aspects, the same CAR includes activation and co-stimulatory components; alternatively, the activation domain is provided by one CAR, and the co-stimulatory component is provided by another CAR that recognizes another antigen.

Thus, in some embodiments, a CAR may include:

(1) In its extracellular portion, one or more antigen-binding molecules, such as one or more antigen-binding fragments, domains or portions of antibodies, or one or more antibody variable domains (heavy and/or light chains), and/or antibody molecules.

(2) In its transmembrane portion, the transmembrane domain is derived from human T cell receptor-α or -β chain, CD3ζ chain, CD28, CD3-ε, CD45, CD4, CD5, CD8, CD9, CD16, CD22, CD33, CD37, CD64, CD80, CD86, CD134, CD137, ICOS, CD154 or GITR. In some embodiments, the transmembrane domain is derived from CD28, CD8 or CD3-ζ.

(3) one or more costimulatory domains, for example, costimulatory domains derived from human CD28, 4-1BB (CD137), ICOS-1, CD27, OX40 (CD137), DAP10 and GITR (AITR). In some embodiments, CAR comprises a costimulatory domain of CD28 and 4-1BB.

(4) one or more intracellular signaling domains comprising one or more ITAMs in its intracellular signaling domain, for example: an intracellular signaling domain or a portion thereof from CD3-ζ, or a variant thereof lacking one or two ITAMs (for example: ITAM3 and/or ITAM2, see also the above detailed description and references), FcRγ, FcRβ, CD3γ, CD3δ, CD3ε, CDS, CD22, CD79a, CD79b and/or CD66d, in particular, an intracellular domain selected from CD3-ζ, or a variant thereof lacking one or two ITAMs (for example: ITAM3 and ITAM2), or an intracellular signaling domain of FcεRIγ and variants thereof.

CAR or other antigen-specific receptors may also be inhibitory CAR (e.g., iCAR), and include intracellular components that hinder or inhibit reactions (e.g., immune responses). Examples of such intracellular signaling components are found on immune checkpoint molecules, including PD-1, CTLA4, LAG3, BTLA, OX2R, TIM-3, TIGIT, LAIR-1, PGE2 receptors, EP2/4 adenosine receptors (including A2AR). In some aspects, engineered cells include inhibitory CAR, which includes the signaling domain of such inhibitory molecules or the signaling domain derived from such inhibitory molecules, so that it is used to hinder the reaction of cells. For example, when the antigen recognized by the activation receptor (e.g., CAR) is also expressed on the surface of normal cells or may also be expressed on the surface of normal cells, this CAR is used to reduce the possibility of off-target effects.

Exemplary antigen receptors (including CARs and recombinant TCRs) and methods for engineering and introducing the receptors into cells, including, for example, those disclosed in International Patent Application Publication Nos. WO200014257, WO2013126726, WO2012/129514, WO2014031687, WO2013/166321, WO2013/071154, WO2013/123061, WO2019157454, U.S. Patent Application Publication Nos. US2002131960, US201328 7748, US20130149337, U.S. Patent Nos. 6,451,995, 7,446,190, 8,252,592, 8,339,645, 8,398,282, 7,446,179, 6,410,319, 7,070,995, 7,265,209, 7,354,762, 7,446,191, 8,324,353, and 8,479,118, and those described in European Patent Application No. EP2537416, and/or Sadelain et al., Cancer Discov. 2013 April; 3(4): 388-398; Davila et al. (2013) PLoS ONE 8(4): e61338, Turtle et al., Curr. Opin. Immunol., 2012 October; 24(5): 633-39, Wu et al., Cancer, 2012 March 18(2): 160-75. In some aspects, the genetically engineered antigen receptors comprising CAR are those described in U.S. Pat. No. 7,446,190 and International Patent Application Publication No. WO/2014055668A1.

The present disclosure also includes polynucleotides encoding the aforementioned antibodies, antigen-binding fragments or derivatives thereof, TCR and CAR, and vectors comprising the polynucleotides.

Immune Cells

The present disclosure further includes immune cells, in particular isolated immune cells targeting one or more tumor neoantigen peptides as described above. In a more specific embodiment, the present disclosure includes immune cells, in particular isolated immune cells expressing recombinant CAR or TCR as defined above.

As used herein, the term "immune cell" includes cells derived from the hematopoietic system and that play a role in immune response. Immune cells include lymphocytes (such as B cells and T cells), natural killer cells, myeloid cells (such as monocytes, macrophages, eosinophils, mast cells, basophils and granulocytes).

As used herein, the term "T cell" includes cells carrying a T cell receptor (TCR), in particular a TCR for a tumor neoantigen peptide disclosed herein. T cells according to the present disclosure may be selected from the following groups: inflammatory T lymphocytes, cytotoxic T lymphocytes, regulatory T lymphocytes, mucosal-associated constant T cells (MAIT), YδT cells, tumor-infiltrating lymphocytes (TIL) or helper T lymphocytes (including type 1 and type 2 helper T cells and Th17 helper cells). In another embodiment, the cell may be derived from the following group: CD4+T-lymphocytes and CD8+T-lymphocytes. The immune cells may be from a healthy donor or a cancer patient. In some embodiments, the immune cells are allogeneic or autologous cells. In some embodiments, the immune cells are selected from T cells, natural killer T cells, CD4+/CD8+ T cells, TIL/tumor-derived CD8 T cells, central memory CD8+ T cells, Treg, MAIT, Yδ T cells, human embryonic stem cells, and pluripotent stem cells that can differentiate into lymphocytes.

Immune cells can be extracted from blood or derived from stem cells. Stem cells can be adult stem cells, embryonic stem cells, particularly non-human stem cells, umbilical cord blood stem cells, progenitor cells, bone marrow stem cells, induced pluripotent stem cells, totipotent stem cells or hematopoietic stem cells. Representative human cells are CD34+ cells.

T cells can be obtained from a variety of non-limiting sources, including peripheral blood mononuclear cells, bone marrow, lymph node tissue, umbilical cord blood, thymus tissue, infection site tissue, ascites, pleural effusion, spleen tissue and tumor. In certain embodiments, T cells can be obtained from a unit of blood collected from a subject using any number of techniques known to those skilled in the art (e.g., FICOLL ^TM separation). In one embodiment, cells in the subject's circulating blood are obtained by single sampling. In certain embodiments, T cells are separated from PBMC. PBMC can be separated from buffy coat cells obtained by whole blood density gradient centrifugation, for example, by LYMPHOPREP ^TM gradient, PERCOLL ^TM gradient or FICOLL ^TM gradient centrifugation. T cells can be obtained by consuming mononuclear cells (e.g., by using CD14 ) are isolated from PBMCs. In some embodiments, red blood cells can be lysed prior to density gradient centrifugation.

In another embodiment, the cells may be from a healthy donor, a subject diagnosed with cancer. The cells may be autologous or allogeneic.

In allogeneic immune cell therapy, immune cells are collected from healthy donors rather than patients. Typically, these are HLA matched to reduce the likelihood of graft-versus-host disease. Alternatively, generic "off-the-shelf" products that do not require HLA matching include changes designed to reduce graft-versus-host disease, such as disruption or removal of TCR α β receptors. See review Graham et al., Cells. 2018 Oct; 7(10): 155. Since a single gene encodes the α chain (TRAC), rather than two genes encoding the β chain, the TRAC locus is typically a typical target for removal or disruption of TCR α β receptor expression. Alternatively, inhibitors of TCR α β signaling can also be expressed, such as a truncated form of CD3ζ that can be used as a TCR inhibitory molecule. Destruction or removal of HLA class I molecules is also employed. For example, Torikai et al., Blood. 2013; 122: 1341-1349 used ZFN to knock out the HLA-A locus, while Ren et al., Clin. Cancer Res. 2017; 23: 2255-2266 knocked out beta-2 microglobulin (B2M) required for HLA class I expression. Ren et al. simultaneously knocked out TCRαβ, B2M, and immune checkpoint PD1. In general, immune cells are activated and expanded for adoptive cell therapy. The immune cells disclosed herein can be expanded in vivo or in vitro. Immune cells, particularly T cells, can generally be activated and expanded using methods known in the art. Typically, T cells are expanded by contacting an antigen attached to a surface having a reagent that stimulates CD3/TCR complex-related signals and a ligand that stimulates a co-stimulatory molecule on the surface of the T cell.

In one embodiment of the present disclosure, immune cells may be modified to target previously defined tumor neoantigen peptides. In specific embodiments, the immune cells may express recombinant antigen receptors on their cell surfaces that target the neoantigen peptides. "Recombinant" refers to an antigen receptor that is not encoded by the cell in its native state, i.e., it is heterologous and non-endogenous. Therefore, the expression of recombinant antigen receptors can be regarded as introducing new antigen specificity into immune cells, causing the cells to recognize and bind to the aforementioned peptides. Antigen receptors can be isolated from any useful source. In some embodiments, the cells contain one or more nucleic acids encoding one or more antigen receptors introduced by genetic engineering, wherein the antigen contains at least one tumor neoantigen peptide according to the present disclosure.

Antigen receptors according to the present disclosure are genetically engineered T cell receptors (TCR) and components thereof, as well as functional non-TCR antigen receptors, such as the aforementioned chimeric antigen receptors (CAR).

Methods by which immune cells can be genetically modified to express recombinant antigen receptors are well known in the art. Nucleic acid molecules encoding antigen receptors can be introduced into cells in the form of, for example, vectors or any other suitable nucleic acid constructs. Vectors and components required therefor are well known in the art. Nucleic acid molecules encoding antigen receptors can be produced using any method known in the art, such as molecular cloning using PCR. Antigen receptor sequences can be modified using common methods (e.g., site-directed mutagenesis).

In some embodiments of the present disclosure, the immune cell is a cell in which (a) the SUV39h1 gene is inactivated, (b) the antigen-specific receptor is a modified TCR comprising a heterologous (or recombinant) antigen-binding domain and a natural TCR constant domain or a fragment thereof as defined above, and the antigen-specific receptor is capable of activating a CD3ζ polypeptide. For example, the immune cell may further comprise at least one chimeric costimulatory receptor (CCR) and/or at least one chimeric antigen receptor, for example, as defined above.

In a related aspect, immune cells, particularly allogeneic immune cells, can be designed to reduce graft-versus-host disease so that the cells contain inactivated (e.g., destroyed or deleted) TCR α β receptors. In this case, nucleic acids encoding the antigen binding domains of HI-TCR (generally as defined above) are conveniently inserted into the endogenous TRAC loci and/or TRBC loci of immune cells. Insertion or other minor mutations of HI-TCR nucleic acid sequences can destroy or eliminate endogenous expression of TCRs containing natural TCR α chains and/or natural TCR β chains. Insertion or mutation can reduce endogenous TCR expression by at least about 50%, 60%, 70%, 75%, 80%, 85%, 90%, 95%, 98% or 99%. Since a single gene encodes the α chain (TRAC), rather than two genes encoding the β chain, the TRAC locus is usually a target for reducing TCR α β receptor expression. Therefore, nucleic acid encoding an antigen-specific receptor (e.g., CAR or TCR) can be integrated into a position of the TRAC locus, preferably the 5' region of the first exon (SEQ ID NO: 3), which significantly reduces the expression of functional TCR alpha chain. See, for example, Jantz et al., WO 2017/062451, Sadelain et al., WO 2017/180989, Torikai et al., Blood, 119(2): 5697-705 (2012); Eyquem et al., Nature. 2017 Mar 2; 543(7643): 113-117. The expression of endogenous TCR can be reduced by at least about 50%, 60%, 70%, 75%, 80%, 85%, 90%, 95%, 98% or 99%. In these embodiments, expression of the nucleic acid encoding the antigen-specific receptor is optionally under the control of the endogenous TCR-α or endogenous TCR-β promoter.

Optionally, the immune cell also includes a modified CD3 with a single active ITAM domain, and optionally, CD3 may further include one or more or two or more costimulatory domains. In some embodiments, CD3 includes two costimulatory domains, optionally CD28 and 4-1BB. The modified CD3 with a single active ITAM domain may include, for example, a modified CD3 ζ intracellular signaling domain in which ITAM2 and ITAM3 are inactivated or ITAM1 and ITAM2 are inactivated. In some embodiments, the modified CD3 ζ polypeptide retains only ITAM1, and the remaining CD3 ζ domains are missing (90-164 residues). As another example, ITAM1 is replaced by the amino acid sequence of ITAM3, and the remaining CD3 ζ domains are missing (90-164 residues).

The modified immune cells disclosed herein may comprise a combination of two or more, or three or more, or four or more of the aforementioned aspects.

For example, the modified immune cell is an immune cell in which (a) the antigen-specific receptor is a modified TCR comprising a heterologous (or recombinant) antigen binding domain (generally as defined above) and a natural TCR constant domain or a fragment thereof, and the antigen-specific receptor is capable of activating a CD3 ζ polypeptide, and/or the antigen-specific receptor is a CAR, and optionally (b) the SUV39H1 gene is inactivated, and optionally (c) the immune cell comprises a modified CD3 with a separate active ITAM domain, for example, wherein ITAM2 and ITAM3 are inactivated, and optionally (d) the TCR is controlled by an endogenous TRAC and/or TRBC promoter, and optionally (e) the expression of a natural TCR-α chain and/or a natural TCR-β chain is destroyed or eliminated. In further embodiments, the cell may comprise at least one chimeric costimulatory receptor (CCR).

The present disclosure also relates to methods for providing immune cells, in particular T cell populations targeting the tumor neoantigen peptides disclosed herein, in particular immune cells and in particular T cell populations expressing TCRs, in particular HLA-independent TCRs (HITCRs) or CARs as defined above.

The T cell population may include CD8+ T cells, CD4+ T cells, or CD8+ and CD4+ T cells.

The immune cell population generated according to the present disclosure can be enriched in immune cells that are specific (i.e., targeted) to the tumor neoantigen peptides or chimeric proteins of the present disclosure, in particular any one of the transmembrane chimeric proteins of SEQ ID NO: 1-8202. That is, the immune cell population generated according to the present disclosure will have an increased number of immune cells targeting one or more tumor neoantigen peptides (i.e., enriched in clonal types targeting neoantigen peptides) or one or more chimeric proteins. For example, compared to the immune cells in a sample isolated from a subject, the immune cell population of the present disclosure will have an increased number of immune cells targeting tumor neoantigen peptides or chimeric proteins. That is, the composition of the immune cell population will be different from that of a "natural" immune cell population (i.e., a population that has not undergone the identification and amplification steps discussed herein) because the percentage or proportion of immune cells targeting tumor neoantigen peptides or chimeric proteins will increase.

According to the present disclosure, the immune cell population may have at least about 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95 or 100% of T cells targeting the tumor neoantigen peptides or chimeric proteins disclosed herein. For example, the immune cell population may have about 0.2%-5%, 5%-10%, 10-20%, 20-30%, 30-40%, 40-50%, 50-70% or 70-100% of immune cells targeting the tumor neoantigen peptides or chimeric proteins disclosed herein.

For example, when those cells are exposed to tumor neoantigen peptides or chimeric proteins, the amplified tumor neoantigen peptide/or chimeric protein reactive immune cell population may have higher activity than the unamplified immune cell population. Reference to "activity" may represent the response of an immune cell population to restimulation of a tumor neoantigen peptide (e.g., a peptide corresponding to a peptide used for amplification) or a mixture of tumor neoantigen peptides or a chimeric protein as defined herein (or a fragment thereof, typically an extracellular fragment thereof) or a mixture of chimeric proteins (or a fragment thereof, typically an extracellular fragment thereof). Suitable methods for determining the response are known in the art. For example, the production of cytokines can be measured (e.g., the production of IL2 or IFNy can be measured). Reference to "higher activity" includes, for example, 1-5, 5-10, 10-20, 20-50, 50-100, 100-500, 500-1000 times the activity increase. In one aspect, the activity can be more than 1000 times higher.

In a preferred embodiment, the present disclosure provides a plurality of immune cells or immune cell populations (i.e., more than one), wherein the plurality of immune cells comprises immune cells (particularly T cells) that recognize clonal tumor neoantigen peptides and T cells that recognize different clonal tumor neoantigen peptides. Therefore, the present disclosure provides a plurality of immune cells, particularly T cells, that recognize different clonal tumor neoantigen peptides. Different immune cells, particularly T cells, in a plurality or population may optionally have different TCRs that recognize different epitopes of the same tumor neoantigen peptide or chimeric protein.

In a preferred embodiment, the number of clonal tumor neoantigen peptides or chimeric proteins or epitopes of one or more chimeric proteins recognized by multiple T cells is 2 to 1000. For example, the number of recognized clonal neoantigens can be 2, 3, 4, 5, 6, 7, 8, 9, 10, 20, 50, 100, 150, 200, 250, 300, 350, 400, 450, 500, 550, 600, 650, 700, 750, 800, 850, 900, 950 or 1000, preferably 2 to 100. There may be multiple immune cells, particularly T cells, that have different TCRs but recognize the same clonal neoantigens.

The immune cell (particularly T cell) population may consist entirely or mainly of CD8+ T cells, or entirely or mainly of a mixture of CD8+ T cells and CD4+ T cells, or entirely or mainly of CD4+ T cells.

In certain embodiments, the T cell colony is generated by T cells isolated from a cancer patient or a healthy donor. For example, the T cell colony can be generated by T cells in a sample isolated from a patient carrying a tumor. The sample can be a tumor sample, a peripheral blood sample, or a sample from other tissues of the subject.

In a specific embodiment, the immune cell population is generated from a tumor sample in which a tumor neoantigen peptide is identified. In other words, the immune cell (particularly T cell) population is isolated from a biological specimen derived from a tumor of a cancer patient. Such T cells are referred to herein as "tumor infiltrating lymphocytes (TIL)".

T cells can be isolated using methods well known in the art. For example, T cells can be purified from a single cell suspension generated from a sample based on CD3, CD4 or CD8 expression. T cells can be enriched from a sample using a Ficoll-paque gradient.

Cancer treatment methods

In any embodiment, the cancer therapeutic products described herein can be used in methods of inhibiting cancer cell proliferation. The cancer therapeutic products described herein can also be used for the preventive treatment of cancer in patients with cancer or at risk of cancer.

Cancers that may be treated using the methods of treatment described herein include any solid or non-solid tumor as defined above. Of particular interest in light of the present disclosure are breast cancer, melanoma, and lung cancer.

Cancer also includes cancers that are refractory to treatment with other chemotherapeutic drugs. The term "refractory" as used herein refers to cancers (and/or their metastases) that do not show or only show a weak antiproliferative response (e.g., do not show or only show a weak inhibition of tumor growth) after treatment with another chemotherapeutic agent. These cancers cannot be satisfactorily treated with other chemotherapeutic drugs. Refractory cancers include not only (i) cancers in which one or more chemotherapeutic drugs have failed during patient treatment, but also (ii) cancers that can be shown to be refractory by other means (e.g., biopsy and culture in the presence of chemotherapeutic drugs).

The treatments described herein are also applicable to the treatment of previously untreated patients in need thereof.

The subject according to the present disclosure is typically a patient in need thereof who has been diagnosed with cancer or is at risk of developing cancer. The subject is typically a human, dog, cat, horse, or any animal in need of a tumor-specific immune response.

The present disclosure also relates to the previously defined neoantigenic peptides, APC populations, vaccines or immunogenic compositions, polynucleotides encoding the neoantigenic peptides or vectors for use in cancer vaccine therapy or for treating cancer in a subject, wherein the peptide binds to at least one MHC molecule of the subject.

The present disclosure also provides a method of treating cancer in a subject, comprising administering to the subject a therapeutically effective amount of the vaccine or immunogenic composition described herein to treat the subject. The method may additionally include the step of identifying a subject with cancer.

The present disclosure also relates to a method for treating cancer, comprising producing an antibody or an antigen-binding fragment thereof by the method described herein, and administering the antibody or the antigen-binding fragment thereof or an immune cell expressing the antibody or the antigen-binding fragment thereof to a subject suffering from cancer in a therapeutically effective amount to treat the subject.

The present disclosure also relates to antibodies (including variants and derivatives thereof), T cell receptors (TCRs) (including variants and derivatives thereof), non-HLA restricted TCRs (HI TCRs) or CARs (including variants and derivatives thereof) directed against transmembrane chimeric proteins as described herein, or tumor neoantigen peptides that are normally bound to MHC or HLA molecules, for use in treating cancer in a subject.

In some embodiments, the antibody, TCR (particularly non-HLA restricted TCR) or CAR binds to a transmembrane chimeric protein as defined herein and in particular any one of SEQ ID NOs: 1-8202. Typically, the antibody TCR (particularly non-HLA restricted TCR) or CAR comprises an antigen binding domain as defined above (which binds to any one of SEQ ID NOs: 1-8202).

The present disclosure also relates to antibodies (including variants and derivatives thereof), T cell receptors (TCRs) (including variants and derivatives thereof) or CARs (including variants and derivatives thereof) directed against the tumor neoantigen peptides described herein (usually bound to MHC or HLA molecules) or against the chimeric proteins described herein, or immune cells targeting the neoantigen peptides or chimeric proteins as defined above, for adoptive cell or CAR-T cell therapy of a subject, wherein the tumor neoantigen peptides bind to at least one MHC molecule of the subject. In some embodiments, the antibody, TCR (particularly non-HLA restricted TCR) or CAR binds to a transmembrane chimeric protein as defined herein and in particular any one of SEQ ID NOs: 1-8202. Typically, the antibody TCR (particularly non-HLA restricted TCR) or CAR comprises an antigen binding domain as defined above (which binds to any one of SEQ ID NOs: 1-8202 transmembrane chimeric proteins). Therefore, typically in some embodiments, immune cells target transmembrane chimeric proteins as defined herein. Generally, one skilled in the art will be able to select suitable antigen receptors that bind and recognize previously defined tumor neoantigenic peptides to redirect immune cells for cancer cell therapy. In certain embodiments, the immune cells used in the disclosed methods are redirected T cells, such as redirected CD8+ and/or CD4+ T cells.

In some embodiments, the cancer treatment, vaccination therapy and/or adoptive cell cancer therapy described above is administered in combination with an additional cancer therapy. Specifically, the T cell composition according to the present disclosure can be administered in combination with checkpoint blockade therapy, co-stimulatory antibodies, chemotherapy and/or radiotherapy, targeted therapy or monoclonal antibody therapy.

Checkpoint inhibitors include, but are not limited to, PD-1 inhibitors, PD-L1 inhibitors, Lag-3 inhibitors, Tim-3 inhibitors, TIGIT inhibitors, BTLA inhibitors, V domain Ig inhibitors of T cell activation (VISTA) inhibitors and CTLA-4 inhibitors, such as IDO inhibitors. Co-stimulatory antibodies deliver positive signals through immunomodulatory receptors (including but not limited to ICOS, CD137, CD27OX-40 and GITR). In a preferred embodiment, the checkpoint inhibitor is a CTLA-4 inhibitor.

The chemotherapeutic entity used herein refers to an entity that is destructive to cells, i.e., an entity that reduces cell viability. The chemotherapeutic entity can be a cytotoxic drug. The chemotherapeutic agents considered include, but are not limited to, alkylating agents, anthracyclines, epothilones, nitrosoureas, ethyleneimine/methylmelamine, alkyl sulfonates, alkylating agents, antimetabolites, pyrimidine analogs, epipodophyllotoxins, enzymes such as L-asparaginase; biological response modifiers such as IFNa, IL-2, G-CSF and GM-CSF; platinum coordination complexes such as cisplatin, oxaliplatin and carboplatin; anthracenediones; substituted ureas such as hydroxyurea; methylhydrazine derivatives, including N-methylhydrazine (MIH) and procarbazine; adrenocortical inhibitors , such as mitotane (o,ρ'-DDD) and aminoglutamine; hormones and antagonists, including adrenocortical steroid antagonists, such as prednisone and its equivalents, dexamethasone and aminoglutamine; progestins, such as hydroxyprogesterone, medroxyprogesterone acetate and megestrol acetate; estrogens, such as diethylstilbestrol and ethinyl estradiol equivalents; antiestrogens, such as tamoxifen; androgens, including testosterone propionate and flumethoxetine/equivalents; antiandrogens, such as flutamide, gonadotropin-releasing hormone analogs and leuprolide; and non-steroidal antiandrogens such as flutamide.

"In combination" may mean that the additional treatment is administered before, simultaneously with, or after the administration of the T cell composition according to the invention.

In addition or as an alternative to checkpoint blockade, the T cell compositions disclosed herein may also be genetically modified using gene editing techniques (including but not limited to TALEN and Crispr/ca) to make them resistant to immune checkpoints. This method is known in the art, for example, see US20140120622. Gene editing techniques can be used to prevent the expression of immune checkpoints expressed by T cells, including but not limited to PD-1, Lag-3, Tim-3, TIGIT, BTLA CTLA-4 and combinations thereof. The T cells discussed herein can be modified by any of these methods.

T cells according to the present disclosure may also be genetically modified to express molecules that increase homing to tumors and/or delivery of inflammatory mediators to the tumor microenvironment, including but not limited to cytokines, soluble immunomodulatory receptors and/or ligands.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1: Tumor neoantigen peptides (or TE-derived epitopes) with predicted affinity to MHC alleles less than 500 nM were identified in the mouse tumor B16F10-OVA cell line (A) and the MCA101-OVA cell line (B) by the computer simulation method according to the present invention, and in both cell lines (C).

Figure 2: (A) RT-PCR gel of the fusion transcript sequence encoding the new antigen peptide N25 amplified from the cDNA of the mouse tumor B16F10-OVA and MCA101-OVA cell lines. (B) RT-PCR gel of the fusion transcript sequence encoding the new antigen peptide N26 amplified from the cDNA of the mouse tumor B16F10, B16F10-OVA and MCA101-OVA cell lines.

Figure 3: (A) ELISPOT detection of peptide-reactive IFNg secreting cells in inguinal lymph nodes of animals immunized with DMSO (negative control), OVA (ovalbumin) (positive control), N25 peptide or N26 peptide. (B) IFNg spots of 10^5 cells of animals immunized with DMSO (negative control), SIINFEKL (positive control), N25 or N26 peptide.

Figure 4: (A) Development of tumor volume (mm ³ ) of mice previously immunized with DMSO, OVA or N25L peptide a few days after injection of tumor cells B16F10-OVA into said immunized mice. (B) Development of tumor volume (mm ³ ) of mice previously immunized with DMSO, OVA or N26L peptide a few days after injection of tumor cells B16F10-OVA into said immunized mice.

Figure 5: TCGA datasets of 784 luminal, 100 HER2+, 197 TNBC, 112 normal breast tissues, 516 primary lung adenocarcinomas (primary tumors), and 59 normal lung tissues (solid tissue normal) were analyzed by the method for identifying tumor neoantigenic peptides encoded by fusion transcript sequences. (A) The number of fusion transcript sequences (TE-exon fusions) in different subtypes of breast cancer (HER2+, TNBC, normal breast tissue, and luminal). (B) The number of fusion transcript sequences (TE-exon fusions) in different subtypes of lung cancer (primary lung adenocarcinoma, normal lung tissue).

Figure 6: 8-9 amino acid long peptides predicted from TE gene fusion products of each sample were tested by computer simulation for binding to predicted HLA alleles expressed in the same sample. Peptides with predicted affinity less than 500 nM for at least one HLA-A, -B or -C allele in each sample are shown. (A) Breast cancer samples of different subtypes (HER2+, TNBC, normal breast tissue and luminal). (B) Lung cancer samples of different subtypes (non-small cell lung cancer, normal lung tissue).

Figure 7: Distribution of tumor-specific peptides per patient in breast tumor subtypes. (A) The number of tumor-specific HLA binding peptides for breast cancer patients of each subtype is shown. (B) The number of shared predicted tumor neoantigen peptides in luminal subtype samples (n=784) (abscissa). (C) The number of shared predicted tumor neoantigen peptides in HER2+ subtype samples (n=100) (abscissa). (D) The number of shared predicted tumor neoantigen peptides in TNBC subtype samples (n=197) (abscissa).

Figure 8: (A) Number of tumor-specific HLA binding peptides in each primary lung adenocarcinoma (LUAD) sample (lung cancer). (B) Distribution of tumor-specific peptides in lung adenocarcinoma per patient. Number of predicted tumor neoantigen peptides shared among primary tumor subtype samples (n=516) (abscissa).

Figure 9: Reconstruction of fusion nucleotide sequences when the donor is an exon (A) and the donor is a TE (B).

Figure 10: Binding of chimeric transcript-derived peptides to HLA-A2. Binding of the predicted peptides in the highest frequency chimeric fusions to the HLA-A2 allele was verified by flow cytometry using a tetramer formation assay. Results are shown as percent binding relative to the positive control. The dashed line indicates the threshold that is considered confirmed binding to this allele.

Figure 11: Binding of ER-derived peptides to HLA-A2 molecules. Peptide-HLA-A*02:01 complex formation for synthetic chimeric transcript-derived peptides. The percentage of complex formation relative to the positive control (CMV pp65495-503) is shown. Mutated (MelA Mut) and non-mutated (MelA) sequences of Melan-A were used as strong binding peptide and weak binding peptide controls, respectively. "Negative" indicates staining background. The dotted line indicates the minimum complex formation value (50% of the positive control) required for the peptide to be considered as binding well to HLA-A*0201.

Figure 12: Immunogenicity of fusion transcript-derived peptides and reactive CD8+T cell production. (A) Frequency of pJET (fusion transcript-derived peptide)-specific tetramer-positive CD8+T cells expanded from 6 different healthy donors in an in vitro immunogenicity assay using 6 different healthy donors. (B) Cytokine secretion of CTL clones after stimulation with different concentrations of specific peptides. The generated CTL clones and their peptide specificity are listed on the right. (C) Killing assay for CTL-clone 9 co-cultured with target cells loaded with 2 different peptide concentrations and anti-MHC-1 antibodies or isotype controls (left figure) or with different ratios of unloaded target cells (right figure). (D) Killing assays were performed when CTL clones 9, 80 and 64 were co-cultured with target cells unloaded with peptides and anti-MHCI-I antibodies or isotype controls. Effector: target ratios are shown in each individual panel. H1650 was used as the target cell for each panel of this figure.

Figure 13: Expression of TCRs that recognize fusion-derived peptides. Jurkat reporter cells transduced with TCR sequences derived from CTL clone 9 were co-cultured with target cells alone or loaded with 2 different concentrations of peptides. The inset shows the percentage of Jurkat-positive cells for the 3 reporter genes assessed by flow cytometry, using the H1650 cell line as target cells (upper panel) or the H1395 cell line as target cells (lower panel). Negative control: Jurkat cells were not transduced. No peptide: Transduced Jurkat cells were co-cultured with target cells that were not loaded with peptides. Positive control: Transduced Jurkat cells stimulated with PMA/ionomycin.

Figure 14: A. After co-culturing with target cells loaded with relevant/specific or irrelevant/unrelated peptides (Melan-A), Jurkat cells transduced with TCRs derived from CTL clones were activated, which recognized chimeric transcript-derived peptides. B. In the presence or absence of anti-MHC-I blocking antibodies (W6/32) or isotype controls, after co-culturing with target cells loaded with relevant or unrelated peptides (Melan-A), Jurkat cells transduced with TCRs derived from CTL clones were activated, which recognized chimeric transcript-derived peptides. PMA/viomycin was used as a positive control for activation, and target cells not loaded with peptides were used as negative controls for activation. H1395LUAD cell lines were used as target cells. Each TCR source from CTL-clones is shown at the top, with peptide specificity in brackets, showing the amino acid sequence of the chimeric transcript-derived peptides recognized by each of these TCRs. The peptide sequence is the specific/relevant peptide used to load the target cells in each case. Melan-A and MelA Mut both refer to unrelated peptides (ELAGIGILTV).

Figure 15: Tumor infiltrating lymphocytes recognizing fusion transcript-derived peptides. Percentage of tetramer-positive CD8 T cells for the indicated fusion transcript-derived peptides found in tumor infiltrating lymphocytes (TILs) expanded in the presence of a mixture of fusion transcript-derived peptides + IL2 (A) or with IL-2 alone (B).

Figure 16: Recognize the CD8+T cell phenotype of fusion transcript-derived peptides in LUAD patient-derived samples. Recognize the percentage of tetramer-positive CD8T cells of fusion transcript-derived peptides present in tumors, near tumors, lymph nodes and blood samples of LUAD patients 2 (A, upper figure) and 3 (B, upper figure). In the figure below (A) and (B), the percentage of initial (CCR 7+CD45+), central memory (CM, CCR7+CD45-), effector memory (EM, CCR7-CD45-) and terminal effector (TE, CCR7-CD45+) cells in tetramer-positive parental cell populations is shown.

Figure 17: A. Heat map summarizing the frequency of CD8+ T cells recognizing chimeric transcript-derived peptides found ex vivo without T cell expansion. Only peptide specificities found in at least one tissue are shown (total patients evaluated = 4). B. Percentages of CCR7 and CD45RA in tetramer-positive cells summarized in A after ex vivo staining for patients 2 and 5. (No data available for patient 1). C. Heat map summarizing specific tetramer-positive cells recognizing chimeric transcript-derived peptides after in vitro expansion on CD8+ T cells in tumor, near-tumor, or tumor-draining LN samples from the 5 analyzed patients. Only peptide specificities found in at least one tissue are shown. Black boxes highlight peptide specificities found ex vivo in the same tissue and patient.

Figure 18: Immunopeptidomic analysis of lung tumor samples. Fusion transcript-derived peptide sequences were searched in the public MHC-I immunopeptidome dataset. Each column represents a different sample. Each row represents a different peptide sequence (explicitly stated on the right). The filled boxes indicate in which sample each fusion transcript-derived peptide was found. The publication describing each sample dataset is annotated at the top.

Figure 19: FACS histograms showing untransfected negative control (left) and ABHD1-JET transfection conditions (right). The red histogram corresponds to anti-Myc staining and the grey line shows non-antibody (buffer) conditions.

Detailed ways

1. Example 1: Identification of tumor neoantigen peptides encoded by fusion transcript sequences

1.1 Proof of concept in mice

In order to detect single and shared tumor neoantigenic peptides generated by fusion transcript sequences, a bioinformatics pipeline was developed. The pipeline is designed to identify tumor-specific mRNA sequences consisting of a portion of the TE sequence and a portion of the exon sequence. The pipeline is meant to determine MHC alleles. For each human sample, the seq2hla (v2.2) tool (bitbucket.org/sebastian_BOEgel/seq2hla) can be used to determine class I and class II MHC alleles. For mouse models, the mouse H-2 allele is generally well known. Bioinformatics methods include aligning transcripts from RNA sequencing with a reference genome. In the proof-of-concept analysis described in this article, mm10 was used for mice and hg19 was used for humans. Different versions of assembled genomes can be used, such as hg19, hg38, mm9, or mm10. The alignment was performed using STAR (v2.5.3a) (github.com/alexdobin/STAR) and the settings are as follows:

- To allow for multiple hit alignments, the parameter outFilterMultimapNmax, which sets the maximum number of loci and allowed alignment read length, was set to 1000, and

- To detect abnormal connections (fusions), the parameter chimSegmentMin, which sets the minimum length of a fused segment, was set to 10, and the parameter chimJunctionoverhangMin, which sets the minimum overhang of a fused junction, was set to 10.

Normal (from SJ.out.tab output file) and abnormal (from Chimeric.out.joint output file) connections were annotated using Ensembl and repeatmasker databases. Normal connections define all connections that match the parameters used for the alignment (maximum intron length ≤ 1000000 bp (set by -alignItronMax), same chromosome and good orientation), and abnormal connections are connections that do not match at least one of the aforementioned criteria. This means that TE/exon connections can be both connection types, but exon/exon connections must be in the normal file (SJ.out.tab). Transcript sequences containing connections between TE sequences and exon sequences were extracted in silico. From regions in the transcript sequence that overlap with the connection, or downstream of the connection when out-of-frame (reading frame non-canonical), the software predicts peptides of all possible 8 or 9-mers in all reading frames. Then, the binding affinity of all these possible peptides to the MHC alleles previously defined for the matched samples was determined by netMHCpan (v3.4) (cbs.dtu.dk/services/NetMHCpan/). There are currently more than a dozen different prediction algorithms used to predict peptide binding affinity, among which NetMHC is the most widely used and effective algorithm in neoantigen prediction pipelines.

Peptides with a p-value below 500 nM or a percentile below 2% were considered potential neoantigens. Each splice site (donor or acceptor) was uniquely annotated as TE or exon. The 5' end was classified as a qualified "donor" and the 3' end as a qualified "acceptor".

Predicted HLA-binding peptides shared between cancer and normal tissues were excluded from further analysis.

The method has been applied to RNAseq data obtained from seven well-characterized murine tumor cell lines (B16F10, B16F10-OVA, MCA101, MCA101-OVA, MC38, MC38-GFP, MC38-GFP-OVA). The cell line with extended OVA corresponds to the same model, but further expresses ovalbumin. In this study, the cell lines were considered as similar models, that is, for example, an assay performed on a cell line from B16F10-OVA was considered a duplicate of an assay performed on a cell line from B16F10.

Using these parameters, a list of candidate peptides was obtained ( FIGS. 1A , 1B, and 1C ), some of which were specific to a particular cell line ( FIGS. 1A and 1B ) and others that were shared between the two tumor cell lines ( FIG. 1C ).

For validation, we selected a series of peptides expressed in B16F10-OVA or MCA101-OVA for which the predicted affinity was below 500 nM. Peptides were selected and an attempt was made to optimize the ratio between the number of reads and the predicted affinity against MHC-1.

Four predicted tumor neoantigenic peptides were selected and characterized by identifying TE and exon sequences (Table 2).

Table 2: Characterization of 4 predicted tumor neoantigenic peptides selected by this method

1.2 Verification of fusion transcript sequence by RT-PCR

First, verification was performed by conventional RT-PCR using a pair of primers, one of which was in the TE sequence and the other in the exon sequence.

To extract and reverse transcribe RNA, 3-5.10 ⁶ cells were lysed in 500 μL Trizol and 100 μL phenol-chloroform was added to the lysate before centrifugation. The aqueous phase was collected, mixed with 100% EtOH in a 1:1 ratio, and transferred to an RNAeasy minikit column. RNA was then collected according to the manufacturer's instructions (including DNAse column treatment). After RNA elution, DNA contaminants were further removed by using Turbo DNAse (Fisher scientific) according to the manufacturer's instructions. RNA concentration was measured using nanodrop, and 1 μg RNA was used for reverse transcription. First-strand synthesis was performed using Superscript III (Life technologies) using oligodT (15) as a primer according to the manufacturer's instructions. Primers were ordered from Eurogentec. PCR reactions were performed using Taq polymerase. After identifying the optimal conditions for each reaction, PCR products were extracted from agarose gels and sequenced using GATC lightrun. Sequence alignment was checked using APE software.

Using this method, bands matching the predicted sizes of N25, N26, N90, and N94 were detected in the cell lines identified in Table 2, respectively (see N25 in Figure 2A). Interestingly, although N26 was only detected in MCA and MC38 cells by RNAseq in silico as described in the pipeline, using RT-PCR, we detected a band corresponding to N26 in B16F10-OVA cells (Figure 2B), indicating that this sequence is shared between three independent tumor cell lines (MCA, MC38, and B16F10). By reanalyzing the RNAseq data, we found that the N26 junction was present in B16F10-OVA cells, but below the detection threshold of the algorithm. In addition, sequencing of the RT-PCR product showed an exact match to the sequence predicted by the algorithm.

1.3 Immunization of mice

To validate these candidates in vivo, short (9-mers) peptides corresponding to neoantigenic peptides that bind to MHC class I sequences were synthesized. For in vivo experiments, long (27-mers) peptides were synthesized that included flanking regions of the predicted 9-mer MHC-binding short peptides, as this length is more suitable for in vivo immunization. B16F10OVA and MCA101-OVA were maintained in RPMI, glutamine, 10% FCS, 1% penicillin-streptomycin and passaged using TrypLE. Cells were maintained in culture for up to one month, and new vials were thawed for each in vivo experiment. C57BL6J recipient mice were immunized with 100 μg of long peptide (N25L or N26L), SIINFEKL peptide (short OVA peptide), OVA (Sigma), or DMSO (each containing 50 μg polyI:C) by subcutaneous injection into the flank. Immunizations were repeated 7 days after the initial immunization. After 3 days (10 days after the initial immunization), the animals were sacrificed and the number of peptide-specific IFNg-secreting CD8 T cells in the inguinal lymph nodes was detected by ELISPOT (Figure 3A). 10μg.mL ^-1 of short peptides (N25, N26 or SIINFEKL) or DMSO was used to re-stimulate T cells. Alternatively, 7 days after the secondary immunization, the animals were subcutaneously injected with 2.5.10 ⁵ B16F10-OVA or 5.10 ⁵ MCA-OVA cells in PBS. We found that N25 and, to a lesser extent, N26 were able to induce an immune response (Figure 3B).

1.4 In vivo treatment of mouse tumors

To test whether these peptides have a protective effect on tumor cells, we immunized C57BL6 mice with 100 mg of peptide N25L or N26L, or OVA (control peptide) and 50 μg polyI:C in PBS on days 0 and 7, and injected 2.5.10 ⁵ B16F10-OVA cells into mice immunized with OVA, N25L and N26L on day 14. B16F10 OVA and MCA101-OVA were maintained in RPMI, glutamine, 10% FCS, 1% penicillin-streptomycin and passaged using TrypLE. Cells were maintained in culture for up to one month and a new vial was thawed for each in vivo experiment. Immunizations were performed by subcutaneous injection into the flank with 100 μg of long peptide (N25L or N26L), OVA (Sigma) or DMSO (each containing 50 μg polyI:C). Immunizations were repeated 7 days after the initial immunization.

10 μg.mL ^-1 of short peptides (N25, N26 or SIINFEKL) or DMSO was used to re-stimulate T cells. Alternatively, 7 days after the secondary immunization, the animals were subcutaneously injected with 2.5.10 ⁵ B16F10-OVA or 5.10 ⁵ MCA-OVA cells. Tumor size was measured twice a week using a manual caliper, and animal health was monitored throughout the experimental time frame (Figures 4A and 4B). Animals were sacrificed when the tumor volume reached 1 mm ^3. Strikingly, we observed that N25L significantly delayed the formation of B16OVA tumors in a more effective manner than OVA. In addition, we obtained similar results after N26L immunization.

2. Example 2: Identification of human lung adenocarcinoma neoantigenic peptides derived from fusion transcripts (consisting of TE elements and exon sequences)

2.1 Materials and methods

RNA extraction. Samples of tumors and adjacent tissues were cut into 1 mm ³ fragments and resuspended in 700 μl of RTL lysis buffer (Quiagen) supplemented with 1% β-mercaptoethanol and homogenized using a Perecellys 24 tissue homogenizer (Bertin Technoogies). Total RNA was isolated using the RNeasy Micro Kit (Qiagen) according to the manufacturer's instructions. Total RNA of tumor cell lines was extracted from 5.10 ⁶ tumor cell lines using the same process.

PCR and sequencing. Primers were designed using APE software. For each sample, 1 μg of RNA was reverse transcribed into cDNA using SuperScript III reverse transcriptase (ThermoFisher) according to the supplier's instructions. PCR reactions were performed using GoTaq G2 hot start polymerase (Promega). All primers were used at a concentration of 0.5 μM. Reactions were performed in a VeritiTM 96-well thermal cycler (ThermoFisher). PCR products were loaded into LabChip GX (Caliper LifeSciences) and analyzed by LabChip GX software (v4.2).

For samples with amplified products of expected size, PCR reactions were repeated. Then, PCR products were run in 2% agarose gel without SYBR dye (1/10000) (Invitrogen). According to the manufacturer's instructions, QIAquick gel extraction kit (Qiagen) was used to cut specific bands and purify DNA products. Finally, these products were sequenced by EuroFinsScientific. Serial Cloner software was used to compare the resulting sequence with the expected sequence.

Tetramer formation. HLA-A2 monomers were purchased from The formation of tetramers was assessed with synthetic ER-derived peptides following the manufacturer’s instructions. Briefly, synthetic HLA-A2 monomers were incubated with synthetic peptides at 18 °C for 48 h. Tetramerization was completed by further incubation of the monomers with biotinylated agarose. Finally, tetramer formation was measured by flow cytometry using PE-conjugated anti-β2-microglobulin antibodies. As a positive control, we used a peptide derived from CMV provided by the manufacturer.

In experiments evaluating the presence of specific CD8+ T cells, the tetramerization step was performed by incubating the monomers with different combinations of fluorescent streptavidin (PE, APC, PE-Cy5, PE-CF594, BV421, BV711, and FITC).

Initiate natural CTL. PBMCs were obtained from HLA-A2+ healthy blood donors by Ficoll gradient separation. CD14+, CD4+ and CD8+ cells were purified by positive selection using magnetic beads (Miltenyi Biotec). When CD4+ and CD8+ T cells were cryopreserved until the day of the experiment, 10 ⁶ cells/mL of CD14+ fractions were cultured for 5 days in the presence of IL-4 (50ng/mL) and GM-CSF (10ng/mL) to obtain moDC. After this period of time, moDC were matured with LPS and incubated with synthetic ER-derived peptides at a final concentration of 1μg/mL for 2 hours. Finally, peptide-loaded moDC were co-cultured with autologous CD4+ and CD8+ T cells in culture medium supplemented with IL-2 (10U/ml) and IL-7 (100ng/ml). Stimulation of specific CD8 + CTL populations by ER-derived peptides was assessed by MHC-I tetramer staining by flow cytometry using a two-color tetramer combination for each peptide.

Tetramer staining. The cells were resuspended in PBS, stained with live/dead Aqua-405nm (ThermoFisher) for 20 minutes at 4°C, and washed once. Thereafter, the cells were resuspended in PBS-1% BSA containing a mixture of SA-coupled tetramers and incubated in the dark for 20 minutes at room temperature. Without further washing, the surface antibodies were added to PBS-1% BSA and the cells were incubated in the dark at 4°C for 20 minutes. The surface antibodies were a combination of anti-CD3-BV650 + anti-CD8-PECy7, combined with anti-CCR7-AF700 + anti-CD45RA-BUV395 if necessary. Finally, the cells were washed twice and resuspended in FACS buffer for flow cytometry analysis.

CTL-clone generation. Tetramer-positive cells are sorted by single-cell FACS (ARIA-sorter, BD) in a U-shaped bottom 96-well plate. The sorted cells are collected in 100 μl of RPMI 10% human serum AB (Sigma-Aldrich) containing 150,000 feeder cells. Finally, 100 μl of AIM-medium containing IL-2 (3000 IU/ml) and anti-CD3 (100 μg/ml, OKT3 clone from Miltenyi) is added, and the cells are cultured for up to 15-20 days. When obvious cell growth is observed in the wells, we use new feeder cells for a second round of expansion, which lasts up to 15 days. During this period, cells are fed and lysed with the same medium (AIM-RPMI 50/50+5% human serum) containing only 500 IU/ml of IL-2 when necessary. Finally, the specificity of the amplified clones was checked by FACs-tetramer staining, and only clones with more than 85% tetramer-positive clones were used for further analysis.

Killing assay. For the killing assay, the xCELLigence RTCA S16 real-time cell analyzer was used. The H1650 cell line was plated at 0,5x10 ⁶ cells/ml in pre-coated 16-well plates. One day later, the cells were incubated for 1 hour with different concentrations of the corresponding synthetic peptides or without incubation. Thereafter, the cells were washed twice with medium and incubated for another 30 minutes with the same concentration of anti-MHC-I antibodies (W6/32 clone, 50 μg/well) or isotype controls or without incubation. Without additional washing, CTL-clones were added in the corresponding ratios. The complete assay was done in a final volume of 200 μl of serum-free medium at 37°C and connected to the xCELLigence system. The impedance changes (cell index) were measured in real time for 40 hours. Each condition was repeated.

Cytokine secretion and Jurkat cell activation. 50.000 H1650 cells were plated in 96-well plates in a medium supplemented with 5% fetal bovine serum. The next day, cells were incubated with synthetic peptides at different final concentrations for 1-2 hours. Afterwards, cells were washed twice, CTL-clones were added at a ratio of 1:1, and co-cultured with target cells loaded with peptides for 18 hours. Culture supernatants were collected and cytokine concentrations were analyzed by cytokine bead array (CBA, BD Biosciences) according to the manufacturer's instructions.

The same experiment was performed using transduced Jurkat cells instead of CTL clones and two different types of target cells (H1650 and H1395 cell lines). In this assay, Jurkat cells were evaluated by flow cytometry analysis of the expression of reporter markers after co-culture with peptide-loaded target cells. PMA/ionomycin was used as a positive control to activate Jurkat cells.

Tissue and blood samples. Lung tumors, near tumors and lymph node samples were cut into small pieces and digested for 40 minutes at 37°C using a mixture of type I collagenase (2 mg/ml), hyaluronidase (2 mg/ml) and DNasa (25 μg/ml) in a final volume of 2 ml of culture medium ( _CO2 independent culture medium +5). After digestion, single cell suspensions were collected and washed by cell strainers. Tumor and near tumor suspensions were enriched in lymphocyte fractions by ficoll gradients. Afterwards, cells were stained for tetramer analysis using FACs as described above.

Blood samples were plated on a ficoll gradient and PBMCs were isolated. PBMCs were then enriched for CD8+ T cells using the EasyStep Human CD8+ T Cell Enrichment Kit (STEMCELL Technologies). Finally, the enriched cells were stained for tetramer analysis as described above.

Tumor infiltrating lymphocyte (TIL) culture. Tumor tissue was cut into small pieces (1-3 mm ³ size, 6-12 pieces at most). Each tumor fragment was transferred from a 24-well plate to a single well and cultured in RPMI 10% human serum + IL-26000 IU/ml in a final volume of 2 ml. Cells were fed/lysed during 15-20 days as needed and frozen for storage or analyzed for tetramer staining.

TCR cloning. Total RNA was extracted from CTL-clones and reverse transcribed into cDNA using SuperScript III (ThermoFisher). TCRα and β were amplified by PCR as described by Li et al. 2019. DNA products were run in a 2% agarose gel and sequenced after extraction of gel bands (Qiagen). TCR V regions (α and β) were ligated to the mouse TCR constant chain, cloned into the PEW-PEF1A-inatEGFP vector, and amplified in transformed bacteria.

Jurkat transduction. Lentiviral particles were produced by a HEK-293FT cell line transfected with a TCR-expressing plasmid together with an envelope (pvSVG) and packaging (psPAX2) plasmid. After 64 hours, the supernatant was collected and the lentiviral particles were concentrated using a 100 kDa centrifugal filter (Sigma-Aldrich). The lentiviral suspension was transferred to TCR-negative Jurkat cells expressing reporter genes (NFAT-GPF, NF-KB-CFP, and AP-1-mCherry) by spin inoculation. After 5 days, the transduction efficiency was assessed by FACS using an anti-mouse TCR-β antibody (clone H57-597). The Jurkat cells are described in Rosskopf S.et al.2018.

Mass spectrometry data analysis. Public immunopeptidomics raw data derived from MHC-eluted peptides were analyzed using ProteomeDiscoverer 1.4 (ThermoFisher) with the following parameters: no enzyme, peptide length 8-15aa, precursor mass tolerance 20ppm and fragment mass tolerance 0.02Da. Methionine can be used as a variable modification, and a 1% false discovery rate (FDR) was applied. MS/MS spectra were retrieved for the human proteome from Uniprot/SwissProt (updated on March 6, 2020), combined with a list of proteins derived from all fusion transcripts from the lung TCGA project. Finally, peptides matching translated fusion transcripts present in the Uniprot database or normal lung samples were discarded.

2.2 Results: Identification of fusion transcript sequences encoding tumor neoantigenic peptides in human subjects

2.2.1 Characterization of neoantigens

First, TE-exon fusion transcripts were characterized in normal samples from the TCGA public database. In 679 normal samples from 19 different tissues (bile duct, bladder, brain, breast, cervix, colon, head and neck, kidney, liver, pancreas, PCPG, prostate, rectum, sarcoma, skin, thymus, thyroid, uterus), a total of 8876 unique fusions were identified. Specific fusions of each tissue type were found, with only a very small amount of pan-tissue fusion transcripts. These results show that a special tissue-specific regulatory mechanism is associated with these fusion transcripts.

The number of fusions identified in 514 LUAD samples from TCGA was then compared to 59 normal-related lung samples from TCGA. On average, 235 fusions were identified in NSCLC samples, while 200 fusions were identified in healthy lung samples (Wilcoxon p-value = ^9x10-10 ). A total of 8269 unique fusions were identified in NSCLC tumors.

The first type of fusion obtained was called TSF (tumor-specific fusion), which was found in at least 1% of tumor samples and not in any normal samples. 210 fusions were therefore defined as TSF.

Some high-frequency fusion transcripts in tumors and low-frequency fusion transcripts in normal cells may also be good candidates for neoantigens. Therefore, the second category of fusions called TAFs (tumor-associated fusions) are specifically defined as fusions present in less than 4% of normal tissues, especially less than 2% of normal tissues and more than 10% of tumors, and are overexpressed in tumors compared with normal tissue samples.

Fusion sequence:

- To reconstruct the fusion nucleotide sequence, the sequence of the donor located on chromosome "donor_chromosome_X" from "donor_start_X" to "donor_breakpoint_X" on strand "donor_strand_X" and the sequence of the acceptor located on chromosome "acceptor_chromosome_X" from "acceptor_breakpoint_X" to "acceptor_end_X" on strand "acceptor_strand_X" were extracted from the Ensembl HG19 human assembly database. It should be noted that the use of the Ensembl HG19 human database is not restrictive, and other suitable databases may also be used, such as the NCBI Reference Sequence Database (RefSeq).

- If the fusion occurs on the minus strand, care should be taken to take the reverse complement of the sequence.

- A "fusion sequence" consists of a donor sequence followed by an acceptor sequence.

Nucleotide sequence of the fusion transcript:

All “fusion transcripts” were reconstructed based on the known classical transcripts involving exons.

When the donor is an exon (see Table 9A)

It starts with a canonical transcript from the donor exon and replaces the complete canonical exon sequence with the fusion sequence. In this case, the fusion transcript stops after the recipient's TE sequence.

When the donor is TE (Table 9B)

This sequence starts at the canonical position of the receptor exon in the transcript and all exons upstream are ignored. The canonical sequence of the receptor exon is replaced with the fusion sequence and the transcript is reconstructed until the end.

Each k-sized nucleotide sequence (i.e., 24-75 nucleotides) of the fusion transcript (translation of the first k-mer starts from the first nucleotide of the fusion transcript, translation of the second k-mer starts from the second nucleotide of the fusion transcript, etc.) is then translated into a peptide sequence.

The peptides obtained were then further analyzed using NetMHCpan for MHC binding prediction. Thus, the affinity of each k-mer present in the sequence for binding to at least one known human allele was predicted (see Example 1 for further explanation).

The peptides are then further screened against a reference proteome, typically for human subjects, against all sequences present in Uniprot (representing all sequences encoded in the human exome). If a peptide has the same amino acid sequence or differs only in the first or last amino acid, it is considered equivalent to a peptide in Uniprot. All these equivalent sequences are then discarded from the candidate list. Therefore, 117 peptide sequences derived from these 230 fusion transcripts are predicted to bind to HLA-A2:01 (see Table 3 below).

Table 3: Peptide LUAD

2.2.2 Validation of HLA-A2-associated peptides

Given that the HLA-A2 allele is expressed in nearly 50% of the Caucasian population and that different technical tools exist, validation has focused primarily on peptides associated with HLA-A2.

In the following paragraphs, TE-exon-derived transcripts may be used interchangeably with “fusion transcripts” and the term “TE-derived peptides” may be used interchangeably with “fusion transcript-derived peptides”.

Expression of TE-exon-derived transcripts in lung adenocarcinoma samples

To experimentally validate the predicted TE-exon transcripts, expression in LUAD tumor samples and tumor cell lines was first verified by PCR. Therefore, specific primers for each chimeric fusion were designed so that one primer binds to the TE portion of the fusion and the other primer binds to the exon portion of the fusion. The results were further confirmed by sequencing the PCR products.

Specifically, specific primers are designed so that the forward primer binds in the "donor" sequence of the reconstructed fusion sequence and the reverse primer binds in the "acceptor" sequence of the reconstructed fusion sequence. PCR is run on RNA derived from lung tumor samples and human tumor cell lines. The amplified product is inoculated on an agarose gel, and the bands found in the expected size are cut and sequenced. Finally, the PCR product of sequencing is compared with the reconstructed fusion sequence.

Using this approach, it was possible to confirm the presence of the predicted fusion transcripts in both LUAD tumor samples and tumor cell lines. Table 4 below summarizes the results for the 8 most frequent chimeric fusions with predicted peptides that are associated with high affinity binding to the HLA-A2 allele.

Table 4: Validation of the most frequent fusion transcripts. The most frequent fusion peptides in 15 LUAD tumor samples and 6 LUAD tumor cell lines were validated by PCR. The "yes" or "no" status in the table below indicates the presence or absence of a PCR product of the expected size. When the PCR product was further validated by sequencing, it was indicated as "yes".

Binding of ER-derived peptides to HLA-A2 molecules

Once the expression of the chimeric transcripts was confirmed, derivative peptides were synthesized and their binding to HLA-A2 was confirmed. Since monomer stabilization and tetramer formation are only possible in the presence of high affinity binding peptides, the formation of HLA-A2 tetramers in the presence of synthetic peptides was estimated by flow cytometry. All predicted peptides were able to stabilize the formation of tetramers, showing a fluorescence percentage higher than 50% relative to the positive control. As a positive control, a known HLA-A2 high affinity binding peptide derived from cytomegalovirus (CMV) was used. The results confirmed the predicted high affinity binding to the HLA-A2 allele. Figure 10 shows the results of the 10 peptides obtained from the highest frequency fusion peptides.

Figure 11 shows a set of new peptides, also derived from high frequency chimeric transcripts, whose binding to HLA-A2 was confirmed using the same peptide-MHC-I complex formation assay. As positive controls for complex formation, we used CMV pp65495-503 (NLVPMVATV) and a mutant sequence of Melan-A (MelA mut, ELAGIGILTV), both of which are known good binders of HLA-A2. The non-mutated sequence of Melan-A (MelA) was used as a control for low binding peptides. Negatives are recombinant HLA-A2 molecules without any peptide.

Immunogenicity of ER-derived peptides

The next step after validation in conjunction with the HLA-A2 allele is to test the immunogenicity of the predicted peptide. Therefore, a priming assay was performed to test the ability of the identified peptide to amplify specific cytotoxic T cells. PBMCs from HLA-A2+ healthy donors were used to produce monocyte-derived dendritic cells (moDC). After moDC was loaded into a synthetic peptide mixture, autologous co-culture was performed with CD4+ and CD8+ T cells. Finally, the amplification of specific CD8+ T cells was analyzed by flow cytometry using two-color tetramer staining. As a control for specific amplification, co-culture was performed in the absence of peptide. By using this method in one donor, specific CD8+ T cells that recognize the 6 highest frequencies of chimeric fusion-derived peptides (RLLHLESFL, LLGETKVYV, AILPKANTV, RLADHLSFC, FLIVAIEILI, YLWTTFFPL) can be identified and amplified. The result was demonstrated by increasing the percentage of tetramer-positive cells by at least one order of magnitude compared to the control test in total CD8+ T cells.

The same experiment was performed to evaluate the responses of an additional 5 donors. Figure 12A summarizes the results obtained after analyzing a total of 6 donors, where we found specific CD8+ T expansion for 23 of the 29 most frequent fusion transcript-derived peptides (YLWTTFFPL, FLGTRVTRV, RLADHLSFC, LLGETKVYV, MLVTWELAL, MLMKTVWQA, SLMQSGSPV, AILPKANTV, AMDGKELSL, LLDRFGYHV, GLLNISHTA, ILTASITSI, ILSGYGPCV, RQAPGFHHA, GLPSHVELA, ILHSLVTGV, LLHLESFLV, VLLTNTIWL, LLTSWHLYL, RLLHLESFL, YLPYFLKSL, VLMWTMAHL, YLQGLPLPL). As a positive result for amplification, a mutated Melan-A peptide (ELAGIGILTV) was used. These experiments demonstrated that the peptides were able to induce an immune response and confirmed the immunogenicity of the ER-derived peptides.

Production of cytotoxic T lymphocyte clones recognizing ER-derived peptides

To generate cytotoxic T lymphocyte (CTL) clones, CD8+ tetramer-positive T cells expanded in the immunogenicity assay (FIG. 12A) were FACS-sorted. Ten clones were generated that recognized five different ER-derived peptides: YLWTTFFPL, LLGETKVYV, MLVTWELAL, MLMKTVWQA, RLADHLSF. These peptides are listed as peptides 9, 86, 53, 80, and 64, respectively, in Table 3. Reference will be made to these numbers to indicate the specificity of each generated CTL clone. For example, CTL-clone 9 recognized ER-derived peptide 9. In a second set of experiments, a new CTL-clone 17 was generated that recognized peptide 17 (LLDRFGYHV).

To evaluate the cytotoxic capacity of the generated CTL-clones, two different functional assays were performed using the H1650 cell line as target cells. This is a LUAD-derived tumor cell line expressing the HLA-A2 allele.

First, the ability of CTL clones to secrete cytokines after exposure to ER-derived peptides was measured. After co-culture of CTL clones with target cells loaded with specific ER-derived peptides for 18 hours, the secretion of INF-γ, TNF and granzyme B (Gr-B) was measured in the culture supernatant. All CTL clones were activated after exposure to specific ER-derived peptides and secreted cytokines in a dose-dependent manner (Figure 12B).

In a second set of experiments, the killing capacity of CTL-clones was evaluated. CTL-clones were co-cultured with target cells loaded or not with ER-derived peptides under different conditions. Real-time impedance changes in the target cell monolayer were measured using the xCELLigence system. In these assays, a decrease in the cell index correlated with a decrease in the number of cells in the monolayer reflecting cell viability.

When CTL-clone 9 was co-cultured with target cells loaded with ER-derived peptide 9 at a ratio of 1:1, a decrease in the cell index was observed over time compared to control cells (target cells alone). This decrease in the cell index was inhibited when co-culture was performed in the presence of a blocking anti-MHC-I antibody (+ anti-MHC-I). The decrease in the cell index was not inhibited when co-culture was performed using an isotype control (+ isotype) at the same concentration. In addition, the decrease increased when the target cells were loaded with a higher concentration of peptide (1 pM compared to 1 uM) (Figure 12C, left). These results indicate that cytotoxic T cells are able to recognize the peptide encoded by the fusion transcript described herein and kill target tumor cells expressing such peptides.

It was subsequently confirmed that the ER-derived peptides are naturally expressed and presented by target cells, and therefore the target cells can be killed by co-culturing them with CTL-clones without the addition of external peptides. For this purpose, co-culture of CTL-clone 9 with different ratios of H1650 target cells was performed. The right figure of Figure 12C shows that CTL-9 is able to kill target cells at an effector-target ratio of 4:1 compared to control cells (target cells only). In addition, at a higher ratio (8:1), the killing effect increases. No target cells were found to be killed at a lower ratio (2:1).

Finally, similar experiments were performed on CTL-clone 9, CTL-clone 64, and CTL-clone 80, showing that specific killing of target cells could also be inhibited when co-cultured in the presence of anti-MCH-I antibodies ( FIG. 12D ).

Taken together, these results confirm that cytotoxic T cells that recognize a variety of different peptides encoded by the fusion transcripts described herein are able to recognize and kill tumor cells expressing the specific fusion transcript-derived peptides, and that this effect is due to specific recognition of the peptides in the context of MHC-I molecules. In addition, the fact that CTL clones are able to kill target cells without the addition of external peptides provides clear evidence that the fusion transcript-derived peptides are naturally expressed and presented in LUAD tumor cell lines.

Generation of engineered T cells that recognize fusion-derived peptides

Jurkat cells transduced with the lentiviral vector encoding CTL-9TCR sequences were co-cultured with two different target cells H1650 and H1395. Both were cell lines expressing HLA-A2 alleles derived from LUAD. Jurkat cell activation mediated by TCR was assessed by the increase of the fluorescence of reporter genes (NFAT-GPF, NF-KB-CFP and AP-1-mCherry) in flow cytometry. Preliminary results show that Jurkat cells are activated when co-cultured with two target cells compared with negative control (untransduced Jurkat cells). In addition, when co-culturing with target cells loaded with specific peptides, the activation increases in a dose-dependent manner. PMA/ ionomycin is used as a positive control (Figure 13).

These results were repeated in another set of experiments, and similar results were obtained with Jurkat cells transduced with lentiviral vectors encoding TCR sequences from CTL-86, CTL-53, and CTL-17. The transduced Jurkat cells were activated by co-culturing with a target tumor cell line loaded with the corresponding ER-derived peptide (specific/relevant peptide) but not loaded with a control Melan-A peptide (irrelevant/irrelevant peptide, ELAGIGILTV) ( FIG. 14A ). As expected, activation was inhibited by blocking with anti-HLA-I antibodies ( FIG. 14B ). Thus, the TCR expressed by the generated CTL clones is specific for the corresponding HLA-A2 presenting the ER-derived peptide.

These results are consistent with those shown in Figure 12C and D, indicating that LUAD-derived tumor cells express TE-derived peptides. In addition, these results also highlight the technical relevance of CTL-cloned TCR sequences in the development of engineered T cells.

CD8+ cells that recognize fusion-derived peptides exist in LUAD patients

We then aimed to identify the presence of CTL cells recognizing the fusion-derived peptides in LUAD tumor samples.

In the first set of experiments, tumor infiltrating lymphocytes (TILs) amplified with a mixture of TE-derived peptides and Il-2 or with Il-2 alone were analyzed by tetramer staining. As shown in Figures 15A and B, CD8+ T cells that recognized the fusion-derived peptides were found in the TILs of LUAD patients.

It was subsequently shown that tetramer positive cells can be detected, and the phenotype in the unamplified CD8+T cells derived from fresh tumor samples was further evaluated. Using this strategy, CD8+T cells present in the tumor, near tumor, invasive lymph nodes and blood derived from LUAD patient samples were analyzed. Based on the expression of surface markers CCR 7 and CD45RA, the cell phenotypes of initial (CCR7+CD45+), central memory (CM, CCR7+CD45RA-) effector memory (EM, CCR7-CD45-) and terminal effector (TE, CCR7-CD45+) T cells were determined. Interestingly, the tetramer positive cells found in tumor tissue preferentially share memory phenotypes, and initial cells (CCR7+CD45+) are mainly found in cells derived from lymph nodes (Figure 16 A and B). Patients 2 and 3 are identical in Figures 14 and 15.

All tested samples were from HLA-A2+ patients.

The presence of tetramer-positive cells with a memory phenotype in tumor tissue and the presence of tetramer-positive cells in TILs demonstrated that these patients had developed an immune response to TE-derived peptides. In addition, the presence of naive tetramer-positive cells in lymph nodes indicated that an immune response could be generated against these, especially TE-derived peptides.

In the second cohort, five primary, untreated LUAD tumors, adjacent tumors, tumor-draining lymph nodes, and blood samples from HLA-A2+ patients with LUAD cancer were analyzed. Half of each sample was analyzed directly ex vivo by isolating CD8+ T cells without ex vivo expansion, and the other half was cultured in vitro for 20 days with chimeric transcript-derived peptides mixed with IL-2 (patients 1 and 2) or with IL-2 alone (patients 3, 4, 5) in order to expand the specific T cell population that recognizes the chimeric transcript-derived peptides in the sample. T cells were identified by two-color tetramer staining. Antibodies against CCR7 and CD45RA were also added to the non-expanded cells to distinguish between naïve and memory cells. Cells labeled with five or more dual tetramers were considered for expansion.

Figure 17A shows a summary of the 7 tetramer-positive T cell populations found in 4 patients analyzed directly ex vivo (one patient sample could not be analyzed for technical reasons). CCR7/CD45RA labeling showed that all tetramer-positive T cells detected in tumor samples had a clear effector/memory phenotype, while in blood and lymph nodes, tetramer-positive T cells specific for "chimeric transcript-derived peptides" had different proportions of poorly differentiated, CCR7+ naive and/or central memory phenotypes.

Thus, these results indicate that human primary NSCLC tumors contain memory T cells specific for the chimeric transcript-derived peptide ( FIG. 17B ).

A summary of the expanded tetramer+CD8+T cells is shown in Figure 17C. For most peptide specificities, T cells were expanded from tumors and matched invaded lymph nodes (LN) in only 2 patients, and in some cases from matched near-tumor samples (Jt) (Figure 17C). Five of the 7 specific tetramer-positive populations were also found in the same patient at day 20, and no T cells were expanded in tissues found ex vivo (bold boxes in Figures 17A and 17C).

Therefore, these results provide evidence that chimeric transcript-derived peptide-specific T cells exist in tumors, tumor-draining lymph nodes, and sometimes juxtatumor tissues and blood of LUAD patients before and after ex vivo expansion, which is consistent with the presence of chimeric transcript-derived peptide-specific immune responses in LUAD patients.

Identification of peptides in LUAD biopsies by mass spectrometry.

ER-derived peptides require MHC class I molecules to be presented on the surface of tumor cells in order to be recognized by cytotoxic T cells. To confirm that the predicted peptides are expressed on MHC class I molecules, public data from MHC I immunopeptidomes derived from three LUAD biopsies were used (Laumont CM et al., "Noncoding regions are the main source of target tumor-specific antigens" Sci Transl Med. 2018 10 (470)). The raw data uploaded to the PRIDE database from MHC-I immunopurification of three LUAD tumors (PXD009752, PXD009754 and PXD009755) were analyzed using OpenMS software. It should be noted that data-dependent acquisition in proteomics only allows the identification of sequences contained in the target database (usually the entire human proteome); the peptides according to the present application have not been previously identified because they are derived from non-coding sequences. The MS/MS identification of the peptide sequences predicted in this paper that were included in the target database was reanalyzed. Five peptides were found in three sample biopsies (peptide IDs: 3304, 269, 757, 1810, 3953). For this analysis, all predicted peptides derived from chimeric fusions that bind to any MHC I allele present in at least five samples in TCGA were considered. The results confirm the expression of chimeric fusion-derived peptides on MHC class I molecules in LUAD tumors.

We then extended the analysis to new lung immunopeptidomics datasets (Bulik-Sullivan et al. Nat. Biotec 2018, Chong et al. Nat. Comm. 2020 and Javitt et al. Front Immunol 2019). It is worth noting that all datasets were generated using fresh lung tumor samples, except for Javitt et al. Front Immunol 2019, which contained LUAD tumor cell lines. In the second analysis, ProteomeDiscoverer 1.4 software was used to identify ER-derived peptides. Four datasets were considered, and 23 unique ER-derived peptides were present in at least one of the 19 immunopeptidomic samples in total. In Figure 18, the ER-derived peptides (rows) identified in each MHC sample (column) are indicated by gray boxes. The peptide sequences found are shown on the right. Interestingly, some of these peptide sequences were observed in more than one MHC sample, indicating that they are shared between samples. These results confirm that the fusion transcript-derived peptides are processed and presented by HLA-I molecules on the surface of tumor cells.

The peptide RLADHLSFC is derived from a fusion transcript in which the fused gene portion is a tumor suppressor gene (fusion ID: chr22:29117506:->chr 22:29115473:-/related gene: CHEK2), and the peptide GLPSHVELA is derived from a fusion transcript in which the gene portion is an oncogene (fusion ID: chr 6:117763597:->chr 6:117739669:-/related gene: ROS1). Interestingly, both peptides were found to be immunogenic (Figure 12A), especially for the peptide RLADHLSFC, and the results shown in Figure 12D indicate that both peptides can be expressed by the H1650 cell line. In addition, we found TILs that recognized the peptide GLPSHVELA (Figure 13A), indicating that this fusion transcript-derived peptide can be expressed in LUAD tumor samples.

3. Example 3: Identification of new antigenic peptides derived from fusion transcripts composed of TE elements and exon sequences from various cancer samples.

Table 5: 9184 samples from 32 different cancer types (from TCGA)

The RNA datasets from the aforementioned cancer samples were analyzed according to the methods described above.

16,580 fusion transcripts were identified.

4 transmembrane chimeric protein

4.1 Identification of transmembrane fusion proteins

The present disclosure provides a first selection of transmembrane chimeric protein candidates obtained from fusion transcripts predicted by a bioinformatic pipeline developed using publicly available RNA-Seq data from TCGA (The Cancer Genome Atlas) and CCLE (Broad Institute Cancer Cell Line Encyclopedia) to identify genome-wide non-canonical splicing regions (described in the Examples section).

The identification and selection of such transmembrane chimeric protein candidates are described in detail below.

We first identified all transcripts originating from splicing events between TE and exonic sequences in transcriptome mRNA data from TCGA and CCLE databases.

This step has been described in detail in the above sections of this application (Detailed Description and previous Examples).

This step has been described in detail in the above-mentioned part (detailed description and previous embodiments) of the application. As previously mentioned, fusion transcripts come from alternative splicing mechanisms known to be essential for producing functional diversity. Through the rearrangement of existing exon and intron sequences, it allows a single gene to express multiple mRNAs and encode multiple proteins. The types of splicing changes observed include exon skipping, intron retention and the use of alternative splicing donor or acceptor sites. In these fusion transcripts, TE can be used as a donor (positioned at 5') or an acceptor (positioned at 3'), and accordingly, an exon can be an acceptor or a donor. Therefore, TE-exon splicing causes a part of a "non-coding" genome to be incorporated into the coding genome, thereby exposing the non-coding genome sequence to the translation mechanism. These fusion (or chimeric) transcripts, also referred to as JET (joining exon TE), include ORF (open reading frame), i.e., they are parts in the reading frame that have the ability to be translated into polypeptides or proteins. When the TE is the acceptor, the ORF of the fusion transcript is canonical (ie, identical to the canonical transcript), whereas when the TE is the donor, the ORF may be canonical, or may be frameshifted by 1 or 2 nucleotides.

The fusion transcript includes not only the fused TE and exon sequences (corresponding to JET), but may further include exons corresponding to various transcript isoforms, upstream of the fusion breakpoint (between the exon and TE) if the exon is the donor, and downstream of the fusion breakpoint if the TE is the donor.

The identification of transmembrane neoantigenic peptides disclosed herein further comprises the step of selecting fusion transcripts having translated exon sequences annotated in a proteomic database (e.g., UniProt) as belonging to transcripts encoding membrane proteins.

The sequence of the selected fusion transcript is then translated into a fusion peptide (also called a translated junction) sequence.

The complete sequence of each fusion transcript was translated according to the following rules:

- A fusion transcript in which the exon is the donor is translated from the start of the transcript after the classical ORF of the transcript to the first stop codon after the breakpoint between the exon and the TE.

- A fusion transcript in which TE is the donor is translated after 3 ORFs (1-3) from the start of TE or from the last stop codon before the breakpoint between TE and exon to the first stop codon after said breakpoint.

- Only translated peptide sequences containing at least 3 amino acids derived from TE sequences were retained.

In general, peptide sequences from translated junctions that matched any reference or annotated protein sequence in UniProt were discarded, thus focusing on unannotated chimeric peptides (as exemplified in Tables 9 to 12).

Validation of hits by ectopic expression

By applying the above rules, a short list of complete transmembrane chimeric proteins was selected. These chimeric proteins are expected to be produced by TE-receptor fusion or meta-fusion. The corresponding genes and associated chimeric IDs are:

Table 6

Table 7

From Tables 6 and 7 above, 27 TE-receptor fusions and 17 meta-fusion transcripts with added c-Myc sequences were synthesized and cloned into pCDNA 3 plasmids (commercially available). These plasmids were used to ectopically express the predicted chimeric proteins including the c-Myc tag in the HEK293 cell line. After cell transfection and protein expression, c-Myc from the extracellular region was detected and quantified using anti-Myc Alexa Fluor 6472233S clone 9B11 antibody.

The following 19 JET-derived transcripts were validated as positive by this approach, thus demonstrating that the corresponding chimeric proteins were stably translated and inserted into the membrane in the above experimental setup.

Table 8

Figure 19 shows an example of flow cytometry results (transfected with the following constructs: ABHD1ENST00000316470chr 2:27346930:+＞chr 2:27347727:+).

4.2 Total proteomics

Mass spectrometry-based proteomics has emerged as a powerful tool to interrogate the actual protein content of a particular cell preparation. To confirm that JET is indeed translated into protein, mass spectrometry output files (referred to as raw files) generated from cell lines and fresh tumors were analyzed to identify distinct populations of JET-derived peptides. This study was grouped into two different analyses, each providing different and complementary types of information, demonstrating that JET-derived proteins can be reliably detected in tumor samples or tumor cell lines.

First, it was demonstrated that the discovery of JET-derived proteins in the CCLE dataset was highly reproducible. Therefore, the connection of the computer-simulated translation of all those JET mRNA sequences predicted in more than 7 different cell lines in the CCLE cohort was used and queried against the mass spectrometry raw files from Nusinow et al. 2020 (which is a proteomic analysis of 375 cell lines from CCLE). These cell lines were grouped in TMT 6plex, generating a total of 29MS/MS output files. MS-based proteomic analysis identified 57 JET-derived proteins that contained at least 1 peptide overlapping with the splicing connection, and genes involved in splicing events were annotated as being located at the plasma membrane according to Uniprot (Table 9).

Table 9:

The numbers in the columns in the table below refer to the following items.

Despite the inherent limitations of isotopic labeling methods such as TMT, which result in significant label cross-talk between channels, it was confirmed that JET was identified in more than one TMT group and therefore in more than one tumor sample.

Surface protein enrichment:

It has been described that transmembrane proteins are underrepresented in the whole proteomics experiment due to the lysis protocol used. Therefore, a second approach has been used to enrich for extracellularly exposed proteins by biotin labeling in the H1650 lung cell line. Two different analyses were performed using the mass spectrometry raw files obtained from this experiment. First, in order to understand from a general point of view whether fusion-derived proteins (chimeric proteins) are expressed on the plasma membrane, all junctions expressed in more than 7 cell lines (tumor-specific and non-tumor-specific) from the CCLE cohort were analyzed in the mass spectrometry files of this experiment. The identification of 10 chimeric peptides was provided, 6 of which involved the junctions of amino acids where TE and exon were found. From the junction sequence, 4 of them were related to proteins annotated in the membrane compartment. The prediction of transmembrane helices was performed by the TMHMM algorithm (http://www.cbs.dtu.dk/) and the topology of the translated sequences was studied. Based on the predicted topology of the sequences, only candidate sequences were retained for which the TE was predicted to be exposed to the extracellular compartment. A total of 10 fusion-derived peptides were identified, 6 of which overlapped with the splice sites (Table 10).

Table 10:

In addition, peptides involving the classical gene product RHOT2 were found, which carried TEs expected to be exposed to the extracellular compartment. Based on these analyses, it was further suggested that tumor-specific fusion-derived proteins could also be identified using this approach. Therefore, lung tumor-specific JETs from the lung TCGA and CCLE cohorts (as described above) were used to interrogate them to MS files. Of the 16 identified sequences, 8 involved junctions, and the identified amino acids were derived from TEs and classical gene products. Based on the annotated subcellular locations of the classical proteins involved, 5 of the 16 chimeric peptides are likely to be located in the plasma membrane, while the rest (11) may belong to other membrane compartments or contaminant cytosols (Table 11).

Table 11:

Based on the predicted sequences of the translated junctions, the presence of transmembrane helical domains was calculated using the TMHMM algorithm. This yielded two candidate genes of interest, involving the ATP11A and PARM1 gene products.

These initial results reveal the presence of chimeric peptides in the surface group and the power of our methodology to study epigenetic products of translation exposed to the cell surface. Likewise, the demonstrated applicability of our pipeline paves the way for further analysis in different cell models, expanding the scope of our studies and allowing the identification of a larger population of JETs (exon-transposable element junctions).

The original datasets deposited into the ProteomeXchange Consortium through the PRIDE collaborative repository were analyzed with the dataset identifier PXD016582. These analyses correspond to patient-derived organoid clones from single tumor cells that retain heterogeneity and recapitulate the characteristics of colorectal cancer. Four tumor clones were isolated from the same patient and maintained in organoid culture together with a normal colon organoid line generated from a tumor-free colon mucosal tissue biopsy from the same patient. Tumor clones T1, T3, T4, and T5 are morphologically different from normal organoids from the same patient. For more details, see Demmers, L.C., Kretzschmar, K., Van Hoeck, A. et al. Single-cell derived tumor organoids display diversity in HLA class I peptide presentation (see Nat Commun 11, 5338 (2020). https://doi.org/10.1038/s41467-020-19142-9).

Following the same pipeline as described above, the following peptides were identified from the JET transcript:

Table 12:

The second set of results comes from the proteomic profiles of 375 cell lines collected in the Cancer Cell Line Encyclopedia (CCLE) using TMT 10-plex reagents and SPS-MS3 acquisition. For more information on sample sources and characteristics, please see the following publications: Nusinow DP, Szpyt J, Ghandi M, Rose CM, McDonald ER, Kalocsay M, Jané-Valbuena J, Gelfand E, Schweppe DK, Jedrychowski M, Golji J, Porter DA, Rejtar T, Wang YK, Kryukov GV, Stegmeier F, Erickson BK, Garraway LA, Sellers WR, Gygi SP. Quantitative Proteomics of the Cancer Cell Line Encyclopedia. Cell. 2020 Jan 23; 180(2): 387-402.e16. The original files have been uploaded to the MassIVE repository at UCSD.

Table 13:

4.3 Methods

Total Proteomics:

To discover fusions or JETs (Junctioning Exons - TEs) at the proteomic level, publicly available data of lung primary tumors published in Cell by Stewart et al. in 2019 were used (raw files downloaded from the PRIDE database with accession code PXD010357).

In brief, total protein extracts were obtained from cell lines or tumors and subsequently digested with trypsin. The resulting peptides were chemically labeled with isotopic tags using TMT, and different samples were analyzed with internal reference standards in the same experiment. Peptides were fractionated offline by HPLC, and different fractions of each experiment were run separately on a mass spectrometer (Orbitrap Fusion or Orbitrap Fusion Lumos from Thermo-Fisher). For more details on the experimental procedures and analysis, please refer to the corresponding publications.

Proteome Discoverer 2.4 (Thermo-Fisher) and Sequest-HT were used as search engines to query the raw output files from the mass spectrometry run. Custom databases were used to query mass spectrometry peaks, both of which included Swissprot and TrEMBL classical sequences, as well as computer-simulated translations of lung tumor-specific JET sequences predicted from different data sets (lung TCGA and CCLE). Protein cleavage was specified as trypsin with a maximum of 2 missed cleavages allowed. The peptide FDR was set to 1%, while the protein FDR was allowed to be 100% to focus our search on the study of peptides. The mass tolerance of the peptide was 4.5ppm, and the fragment tolerance was 0.02Da. The urethanization of cysteine was set to a fixed modification. For quantification, the signal reported by TMT was obtained by using MS2 or MS3 fragments and paired with MS2 scans for peptide identification. According to the Uniprot annotation, only connections containing identified peptides overlapping with the connection and involving genes located in the plasma membrane were retained.

Surface enrichment proteomics

Lung adenocarcinoma cell line H1650 (ATCC) was grown in RPMI medium supplemented with fetal bovine serum and 1% penicillin/streptomycin at 37°C and 5% CO _2. 10 million cells at approximately 85% confluence were used for enrichment of their surface proteome using the Pierce Cell Surface Protein Biotinylation and Isolation Kit (Thermo Scientific, catalog number A44390).

The culture medium was removed from the adherent cells, washed with PBS, and biotinylated using Sulfo-NHS-SS-Biotin provided in the kit. After incubation at room temperature for 10 minutes, the labeling solution was removed and the cells were washed twice with ice-cold TBS. The cells were then disrupted in ice-cold TBS, pelleted by centrifugation at 500g for 3 minutes at 4°C, and lysed using the lysis buffer supplemented with protease inhibitors provided in the kit. To completely disrupt the cells, they were incubated on ice for 30 minutes in the presence of buffer. The resulting extract was cleared by centrifugation at 15000g for 5 minutes at 4°C.

To capture the labeled protein, the extract was incubated with NeutrAvidin agarose beads on the column provided in the kit on an end-to-end rotator at room temperature for 30 minutes. After incubation, the unbound material was discarded by centrifugation of the column. The beads were washed with the provided wash buffer and immediately centrifuged to discard the flow-through. Washed 4 times with wash buffer, followed by 3 more washes with 20mM Tris-HCl (pH 8). Elution of the biotinylated protein was performed using 100 μl elution buffer (10mM Tris-HCl (pH 8) 10mM DTT), and allowed to incubate with the beads for 45 minutes at room temperature in an end-to-end rotator. The column was centrifuged at 1000 g for 2 minutes to finally recover the concentrated protein.

The eluate was used for subsequent analysis. The protein was alkylated using 5.5 mM CAA for 30 minutes at room temperature in the dark. After verification of the pH value (between 7-9), the protein was then digested in solution using trypsin at a 1:100 ratio (protein: enzyme) at room temperature. Trypsinization was stopped by acidifying the sample with TFA.

The obtained peptides were desalted using an internally packed microcolumn of C18 material. The column was washed with 70% ACN 0.1% TFA and balanced with 0.1% TFA. The sample was loaded and further washed with 0.1% TFA, and then the peptides were eluted with 40% ACN 0.1% TFA. The cleaned peptides were then completely dried and subsequently analyzed by LC-MS/MS in an Orbitrap Fusion (Thermo Scientific).

Tables 14 and 15 below refer to the detailed identification of proteins (or synonyms as peptides used herein) of SEQ ID NOs: 1-1423 translated from fusion transcripts (wherein exons are donors) and chimeric proteins of SEQ ID NOs: 1424-8202 translated from fusion transcripts (wherein TE is donor), respectively. This set of (transmembrane) new antigenic peptides is obtained by selecting fusion transcripts with exon sequences that are annotated as belonging to transcripts encoding transmembrane proteins in normal proteome databases (such as UNIPROT herein). The breakpoint column gives the breakpoint positions between the exon-derived aa sequence and the TE-derived aa sequence. The last column of each table refers to various chimeric proteins (identified by their SEQ ID NOs) derived from splice variants of the same JET (or fusion).

Tables 16-18 refer to meta-fusions. A meta-fusion is a combination of two fusions.

For example, in the following table, metafusion_id:chr1:154709520:->chr1:154705620:-|chr1:15474451:->chr1:154709564:--is constructed from the two chimera ids that are part of the metafusion. The column numbers refer to the following:

Table 18 provides the metafusion_id, transcript, ORF and name of the metafusion peptides of SEQ ID NOs: 9166-10163 (or synonyms as used herein for the peptides).

Tables 19 and 20 relate to detailed identification of the translated fusion peptides of SEQ ID NOs: 10164-12830 translated from the fusion transcript with exon as donor and the translated fusion peptides of SEQ ID NOs: 12331-21452 translated from the fusion transcript with TE as donor, respectively. The column numbers refer to the following:

Column 11 (position) gives the SEQ ID of the peptide translated from the corresponding transcript (referred to as tx in this column) and is associated with each fusion id. The SEQ ID is obtained by adding 10163 to the number in column 11 of Table 19 and 12330 to the number in column 11 of Table 20. Column 12 gives the breakpoint position of each translated peptide in column 11.

Table 19bis and Table 20bis refer to the corresponding peptides obtained from the exon or TE donor fusions of Table 19 and Table 20 and are mentioned in these tables. The tables provide the fusion id, ORF, transcript name and TE name involved in the fusion.

Table 14

Table 14

Table 14

Table 14

Table 14

Table 14

Table 14

Table 14

Table 14

Table 14

Table 14

Table 14

Table 14

Table 14

Table 14

Table 14

Table 14

Table 14

Table 14

Table 14

Table 14

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 15

Table 16

Table 16

Table 16

Table 16

Table 16

Table 17

Table 17

Table 17

Table 17

Table 17

Table 18

Table 18

Table 18

Table 18

Table 18

Table 18

Table 18

Table 18

Table 18

Table 18

Table 18

Table 18

Table 18

Table 18

Table 18

Table 18

Table 18

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 19bis

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Table 20bis

Sequence Description:

Corresponding tables and sequence sources SEQ ID NO Table 14: pJET from Exon Donor Fusion 1-1423 Table 15: pJET from TE donor fusions 1424-8202 Table 9: pJET from proteome 8203-8259 Table 10: pJET from proteome 8260-8269 Table 11: pJET from proteome 8270-8285 Table 12: pJET from proteome 8286-8387 Table 13: pJET from proteome 8388-9165 Table 18: pJET from translation of meta-fusion 9166-10163 Table 19: pJET from translation of exon donor fusions 10164-12830 Table 20: pJET from translation of TE donor fusions 12831-21542 Table LUAD: pJET 21543-21659 pJET from Table 11-18 21660-21698

Claims (30)

1. A chimeric polypeptide comprising or consisting of any one of SEQ ID NOs: 1-21542, or a fragment thereof, optionally having at least 4, 5, 6, 7 or 9 amino acids, wherein the chimeric polypeptide is expressed on a cell membrane. 2. The chimeric polypeptide according to claim 1, which is expressed in more than 1%, in particular more than 5%, and usually more than 10% of tumor samples. 3. The chimeric polypeptide according to any one of claims 1 or 2, which is expressed at a higher level in tumor samples than in normal samples. 4. The chimeric polypeptide according to any one of claims 1 to 3, which is expressed in less than 20%, in particular less than 10%, less than 5% or less than 1% of a normal sample. 5. The chimeric polypeptide according to any one of claims 1 to 4, wherein a portion of the sequence derived from the TE nucleotide sequence is exposed on the cell surface. 6. An antigen binding domain that binds the chimeric polypeptide or fragment thereof according to any one of claims 1 to 5 with a Kd binding affinity of less than about ^10-5 M. 7. The antigen binding domain of claim 6, which binds to a new antigenic peptide sequence from the chimeric polypeptide of any one of claims 1-5, wherein the new antigenic peptide sequence a) is derived from any one of SEQ ID NOs: 1-21542 or a fragment thereof, and comprises at least one sequence derived from a TE-derived amino acid sequence, optionally (i) a fragment overlapping the breakpoint between the TE-derived amino acid sequence and the exon-derived amino acid sequence, or optionally (ii) a pure TE sequence; or b) is derived from any one of SEQ ID NOs: 1-1423, 8203-12830 or a fragment thereof, and is encoded by an atypical ORF downstream of the junction between the TE-derived amino acid sequence and the exon-derived amino acid sequence. 8. An antigen binding domain according to any one of claims 6 or 7, comprising one or more, typically one or two, immunoglobulin regions. 9. The antigen binding domain of any one of claims 6 to 8, comprising the heavy chain variable region (VH) of an antibody, or optionally comprising the three CDRs of VH. 10. The antigen binding domain of any one of claims 6 to 9, comprising the light chain variable region (VL) of an antibody, or optionally comprising the three CDRs of VL. 11. An antibody comprising an antigen binding domain according to any one of claims 6 to 10, optionally wherein the antibody is selected from a whole IgG, a scFv, a BiTE or a multispecific antibody. 12. A chimeric antigen receptor (CAR) or a recombinant non-HLA restricted T cell receptor (TCR) comprising an antigen binding domain as defined in any one of claims 6 to 10. 13. A recombinant non-HLA restricted TCR according to claim 12, wherein the extracellular antigen binding domain is capable of dimerizing with a second extracellular antigen binding domain. 14. The recombinant non-HLA restricted TCR of claim 13, wherein the second extracellular antigen binding domain binds to a tumor antigen, preferably wherein the tumor antigen is selected from pHER95, CD19, MUC16, MUC1, CAIX, CEA, CD8, CD7, CD10, CD20, CD22, CD30, CD70, CLL1, CD33, CD34, CD38, CD41, CD44, CD49f, CD56, CD74, CD133, CD138, EGP-2, EGP-40, EpCAM, Erb-B2, Erb-B3, Erb-B4, FBP, fetal acetylcholine receptor, folate receptor-a, GD2, GD3, HER -2, hTERT, IL-13R-a2, kappa-light chain, KDR, LeY, L1 cell adhesion molecule, MAGE-A1, mesothelin, MAGEA3, p53, MART1, GP100, proteinase 3 (PR1), tyrosinase, survivin, hTERT, EphA2, NKG2D ligand, NY-ESO-1, carcinoembryonic antigen (h5T4), PSCA, PSMA, ROR1, TAG-72, TAG-R2, WT-1, BCMA, CD123, CD44V6, NKCS1, EGF1R, EGFR-VIII, CD99, CD70, ADGRE2, CCR1, LILRB2, LILRB4, PRAME, and ERBB. 15. The CAR according to claim 12, comprising: a) extracellular, comprising the antigen-binding domain of any one of claims 6 to 10, b) transmembrane domain, c) optionally one or more co-stimulatory domains d) an intracellular signaling domain comprising a modified CD3 zeta intracellular signaling domain in which ITAM2 and ITAM3 have been inactivated. 16. The CAR of claim 15, wherein the transmembrane domain is from CD28, CD8 or CD3-ζ. 17. The CAR according to any one of claims 15 or 16, wherein the one or more co-stimulatory domains are selected from: 4-1BB, CD28, ICOS, OX40 and DAP10. 18. The CAR of any one of claims 15 to 17, wherein the intracellular signaling domain comprises the intracellular signaling domain of a CD3-zeta polypeptide, or a fragment thereof, optionally a CD3-zeta polypeptide in which the immunoreceptor tyrosine-based activation motif 2 (ITAM2) and the immunoreceptor tyrosine-based activation motif 3 (ITAM3) are inactivated. 19. A method for producing an antibody, a non-HLA restricted TCR or a CAR as defined in claims 11 to 18, said antibody, a non-HLA restricted TCR or a CAR comprising an antigen binding domain as defined in any one of claims 6 to 10, said method comprising the step of selecting an antibody, a non ^- HLA restricted TCR or a CAR that binds to a neoantigenic peptide or a cell expressing a neoantigenic peptide according to any one of claims 1 to 5 with a Kd binding affinity of about 10-6 M or less. 20. An antibody, TCR or CAR produced by the method of claim 19, optionally wherein the TCR is a non-HLA restricted TCR. 21. A polynucleotide encoding a neoantigenic peptide as defined in claims 1-5 or an antibody, CAR or non-HLA restricted TCR as defined in any one of claims 11-18, optionally linked to a heterologous regulatory sequence. A vector comprising the polynucleotide of claim 21 . 23. An immune cell comprising the CAR or non-HLA restricted TCR defined in any one of claims 12 to 18. 24. The immune cell according to claim 23, which is an allogeneic or autologous cell selected from the following: T cells, natural killer T cells, CD4+/CD8+ T cells, TIL/tumor-derived CD8 T cells, central memory CD8+ T cells, Treg, MAIT, YδT cells, human embryonic stem cells and pluripotent stem cells that can differentiate into lymphocytes. 25. The immune cell according to any one of claims 23-24, which is Suv39h1 deficient. 26. A pharmaceutical composition comprising an effective amount of immune cells as defined in any one of claims 23 to 25 and a pharmaceutically acceptable excipient. 27. The chimeric polypeptide of any one of claims 1-5, the antigen binding domain of any one of claims 6-10, the antibody of claim 11, the non-HLA restricted TCR or CAR of any one of claims 12-18, the polynucleotide of claim 21, the vector of claim 22, the immune cell of any one of claims 23-25, or a composition comprising the same, for inhibiting cancer cell proliferation or for treating cancer in a subject in need thereof, optionally wherein the composition further comprises a pharmaceutical excipient. 28. The chimeric polypeptide of any one of claims 1-5, the antigen binding domain of any one of claims 6-10, the antibody of claim 11, the non-HLA restricted TCR or CAR of any one of claims 12-18, the polynucleotide of claim 21, the vector of claim 22, the immune cell of any one of claims 23-25, or a composition comprising the same, optionally in combination with a pharmaceutical excipient, for use in cancer cell therapy. 29. The chimeric polypeptide of any one of claims 1-5, the antigen binding domain of any one of claims 6-10, the antibody of claim 11, the non-HLA restricted TCR or CAR of any one of claims 12-18, the polynucleotide of claim 21, the vector of claim 22, the immune cell of any one of claims 23-25, or a composition comprising the same for use according to claim 27 or 28, optionally in combination with a pharmaceutical excipient, which is administered in combination with at least one other therapeutic agent. 30. The neoantigenic peptide of any one of claims 1-5, the antigen binding domain of any one of claims 6-10, the antibody of claim 11, the non-HLA restricted TCR or CAR of any one of claims 12-18, the polynucleotide of claim 21, the vector of claim 22, the immune cell of any one of claims 23-25, or a composition comprising the same for use according to claim 29, optionally in combination with a pharmaceutical excipient, wherein the at least one other therapeutic agent is a chemotherapeutic agent or an immunotherapeutic agent, optionally a checkpoint inhibitor.