Novel connection of cancer antigen epitopes derived, specific t cell receptors (TCRs) for immunotherapy and their discovery platform

Abstract

This article provides peptides derived from neoantigens generated by aberrant splicing. In addition to their use as cancer vaccines, these peptides can also be used to sensitize immune cells. T-cell receptors, cells containing them, methods for inducing immune responses, and therapeutic approaches are also provided. A discovery platform for identifying other cancer neoantigens is also described.

Description

[0001] Cross-references to related applications

[0002] This application claims the benefit of U.S. Provisional Application No. 63 / 541,617, filed September 29, 2023, which is incorporated herein by reference in its entirety.

[0003] Statement on Federally Funded Research

[0004] This invention was made with government support under grant numbers R01 CA222965, R35 NS105068, and R01CA244838 granted by the National Institutes of Health. The government owns certain rights in this invention.

[0005] By referencing and incorporating the sequence list provided in the sequence list XML file

[0006] The sequence list, “UCSF-743WO_SEQLIST”, is provided in XML format. It was created on September 26, 2024, and has a size of 21,701 bytes. The contents of the sequence list XML are incorporated herein by reference in their entirety. Background Technology

[0007] Immunotherapy has emerged as a promising cancer treatment strategy, demonstrating significant clinical efficacy across various disease types. However, numerous cases demonstrate that tumors manage to evade eradication due to intratumoral heterogeneity (ITH). ITH refers to the spatial and temporal diversity of the tumor's cellular and genetic landscape, which allows tumor subgroups to acquire immunotherapy resistance and immune evasion. While immunotherapy has been successful in tumors with high immune infiltration and mutational burden, cancer types with extensive ITH or lower mutational burden are more prone to resistance.

[0008] Current immunotherapies targeting tumor-specific antigens (TSAs) have focused on peptides derived from somatic mutations in typical coding regions of the genome. This approach is challenging in tumors with low mutational burdens due to its reliance on TSAs. To mitigate this, recent studies have explored aberrant alternative splicing events across multiple cancers as additional sources of TSAs. These cancer-specific splicing events (neojunctions) have been shown to be ubiquitous in tumors and capable of generating novel TSAs that promote CD8+ T cell-mediated expansion and responses in selected cancer types. However, no studies have examined the persistence of neojunctions across the entire tumor landscape; therefore, the identification of neojunction-derived targets faces similar challenges due to ITH (intracytoplasmic Tissue Hyperplasia). Summary of the Invention

[0009] This article provides neoantigens across the entire tumor spectrum generated by aberrant splicing events. Peptides derived from these antigens (e.g., any one of SEQ ID NO: 1-3) can be used for various immunotherapeutic applications, including stimulating the immune system (e.g., as cancer vaccines). In some embodiments, the peptide can be presented on antigen-presenting cells by class I MHC. These cells can be used in vitro or in vivo to activate immune cells (e.g., T cells). TCRs that bind MHC-peptide complexes are also provided. CD8+ T cells expressing these TCRs can recognize and kill cells expressing the neoantigens.

[0010] Because this neoantigen is shared among patients and is tumor-wide (i.e., expressed intratumorally in heterogeneous tumors such as glioblastoma (GBM) and low-grade glioma (LGG), the peptides of the present invention are considered to represent a new class of “off-the-shelf” cancer immunotherapies, providing a promising avenue for improving cancer treatment strategies, particularly for heterogeneous cancers that may have low somatic mutational burden, such as gliomas, for example glioblastoma (GBM) and low-grade glioma (LGG).

[0011] It also provides a discovery platform for identifying such antigens. Brief description of the attached diagram

[0013] Those skilled in the art will understand that the accompanying drawings described below are for illustrative purposes only. The drawings are not intended to limit the scope of the invention in any way.

[0014] Figure 1 This is a novel neoantigen discovery pipeline for identifying common and tumor-wide targets for cellular immunotherapy across various cancers. TCGA RNA sequencing data from multiple cancers are analyzed to obtain unannotated, protein-coding, and cancer-specific splice linkages (GTEx positive sample rate <1%; neolinks). Patient-conserved linkages (TCGA positive sample rate ≥10%; common neolinks) are retained for downstream intratumoral heterogeneity (ITH) analysis. Tumor sequencing data extracted from multiple intratumoral regions are used to evaluate the ITH of each common neolink. Independent prediction algorithms are used to assess proteasome processing and MHC-I binding of peptide sequences translated from common intratumorally conserved neolinks. The expression of these neolinks and their peptide derivatives is validated by RNA sequencing and mass spectrometry analysis of patient-derived tumor samples and cell lines. T cell receptors (TCRs) are cloned and characterized for high-scoring predicted candidate molecules by in vitro sensitization of PBMC-derived CD8+ T cells using the corresponding neoantigens, followed by 10x V(D)J single-cell sequencing. These neoantigen-reactive TCR sequences were transduced into TCR-deficient Jurkat76 cells and PBMC-derived CD8+ T cells to demonstrate their neoantigen-specific immunogenicity and tumor-specific killing power.

[0015] Figure 2A-2I Characterizing presumed new connections across multiple cancer types. A. Flowchart for identifying presumed, spatially conserved, cancer-specific alternative splicing events (new connections) from RNA sequencing data collected from The Cancer Genome Atlas (TCGA). B. TCGA tumor set with corresponding multi-biopsy RNA sequencing data available for analysis. C. Tumor purity of TCGA tumor samples. Samples with confirmed tumor purity of 60% or higher (solid portion) are retained for downstream analysis. D. Cross-patient frequency (positive sample rate; PSR) of presumed new connections identified in each tumor type. Public new connections are defined as those with a PSR ≥ 10% (red line). E. Total number of public new connections detected per sample across tumor types. F. Log2 (reading frequency) of public new connections across tumor types. GH. Distribution of public new connections based on splice type (G) and frameshift state (H). I. Expression of all public new connections (log2 (CPM)) across all TCGA tumor types in all studies.

[0016] Figure 3A-3G A subset of neolinks are expressed across the entire tumor. A. Overview of the tumor-wide characterization of neolinks by studying RNA sequencing of multiple intratumoral regions in various cancers. B. A heatmap showing the log2 (CPM) of neolinks (rows) in five samples (columns) within the same tumor in colon, kidney, liver, and stomach cancer. Neolinks found across all five intratumoral samples are highlighted in yellow. C. A heatmap showing intratumoral heterogeneity of neolinks (rows) in multiple samples (columns) for each tumor in liver and prostate cancer. The intensity of each cell indicates the percentage of regions within the same tumor with putative expression of each neolink. D. 3D models of the human brain and glioma (green). Approximately 10 spatially mapped and maximum distance biopsies (red) were taken in each tumor. Whole exome sequencing, RNA sequencing, and further analysis were performed on each of these regions. E. Heatmap showing intratumoral heterogeneity of neoconnections (rows) across various glioma subtypes (columns), including glioblastoma (blue), astrocytoma (yellow), and oligodendroglioma (red). The intensity of each cell indicates the percentage of regions within the same tumor with putative expression of each neoconnection. F. Distribution of glioma-specific neoconnections (n=789) based on the number of intratumoral samples detected in all spatially mapped tumors. G. Distribution of glioma-specific neoconnections (blue) based on the number of tumors detected expressing them across their entire tumor range.

[0017] Figure 4A-4KNovel epitopes derived from neolinks were predicted to be processed and presented by MHC-I molecules. AB. RNA sequencing validation of detectable levels of neolink expression (colored) compared to classical splicing (gray) expression from patient-derived LGG(A) and GBM(B) cell lines. C. Mass spectrometry analysis of CPTAC-derived GBM samples (n=99) indicating detectable levels of neolink-derived peptides (purple) expressed at intensities comparable to other peptides (gray). D. Schematic diagram (D) used to identify high-confidence neolinks for downstream analysis. Forty-four neolinks were selected because they were found to be expressed on all three sequencing platforms. E. Schematic diagram illustrating the mechanism of neoantigen generation through the introduction of neolinks. Multimer partitioning was then used to generate a peptide library for predictive analysis. (WT sequence is SEQ ID NO: 20; NJ-derived sequences are SEQ ID NO: 21-23, respectively). FG. Histogram of presentation scores of the top 10 percentile n-mers classified by HLA allele (F) or n-mer length (G). HI. A bar graph shows the overall score of the highest-scoring candidates for cross-HLA allele binding based on frameshift state (H) or alternative splicing class (I). J. A schematic diagram (J) depicting the immunogenicity of the final list of candidates for the highest-scoring neo-linked n-mers based on detectable neo-links from patient-derived mass spectrometry and RNA sequencing data. K. A heatmap showing the final HLA-A binding of glioma samples across all spatial mappings. 02:01 Presentation of new intratumoral heterogeneity of connections.

[0018] Figure 5A-5J T-cell receptors (TCRs) specifically respond to neoantigens derived from new junctions. A. CD8 derived from healthy donor PBMCs. + Overview of the workflow for identifying neolinked neoantigen-reactive TCRs through in vitro sensitization (IVS) of T cells followed by 10x V(D)J single-cell RNA sequencing (scRNA-seq). B. Reactive CD8+ after IVS with neoantigen. + IFNγ ELISA of T cell populations. C. 10x V(D)J IFNG characteristics of highly proliferating TCR clones co-cultured with T2 cells loaded with neoantigen (colored), bait antigen (dark gray), or no antigen (dark gray). Donor 3 CD8 + T cells via NJ RPL22 IVS processing (left), donor 4 CD8 + T cells via NJ RPL22 IVS processing (medium), and donor 4 CD8 + T cells via NJ GNAS IVS treatment (right) resulted in the identification of reactive TCR clonal types. D. In donor 3 CD8+ T cells via NJ RPL22 IVS processing (left), donor 4 CD8 + T cells via NJ RPL22 IVS processing (medium), and donor 4 CD8 + T cells via NJ GNAS IVS treatment (right) shows the clonogenic frequency of all identified TCR clones. E. A procedure for validating the specificity of TCR clonogenicities for novel ligation-derived antigen candidates, utilizing the TCR-transduced triple reporter Jurkat76 system followed by flow cytometry analysis. FG.NJ RPL22 -Derivatives (Part 1) and NJ GNAS -Derived (below) neoantigen-specific TCR transduced triple reporter Jurkat76 cells (F) and PBMC-derived CD8 + T cells (G) can be activated in a dose-dependent manner by neoantigen-loaded T2 cells. TCR-transduced cells were co-cultured with T2 cells loaded with the highest dose of bait peptide (right). TCR activation of the triple reporter Jurkat76 transduced by AP-1-mCherry was measured by flow cytometry. PBMC-derived CD8+ cells were targeted with CD107a and CD137 antibodies. + T cells were stained, and the surface expression of TCR co-activation markers was analyzed by flow cytometry. H.NJ RPL22 -Derivatives (Part 1) and NJ GNAS -Derived (below) neoantigen-specific TCR transduced CD8 + IFNγ ELISA using T cells co-cultured with dose-dependent neoantigen-loaded (left) and bait peptide-loaded T2 cells (right). I.NJ RPL22 -Derivatives (Part 1) and NJ GNAS -Derived (bottom) neoantigen-specific TCR-transduced triple reporter Jurkat76 cells were co-cultured with unloaded T2 cells (left), 0.1 μM neoantigen-loaded T2 cells (middle), or 0.1 μM neoantigen-loaded T2 cells treated with HLA-A2 blocking antibody (right). Cells were stained with CD3 antibody, and TCR activation was assessed by AP-1-mCherry activity. J.NJ RPL22 -Derivatives (Part 1) and NJ GNAS -Derived (bottom) neoantigen-specific TCR-transduced triple reporter Jurkat76 cells co-cultured with alanine-substituted neoantigen-loaded T2 cells, neoantigen-loaded T2 cells, or unloaded T2 cells for alanine scan mutagenesis. Flow cytometry analysis was performed to assess TCR activity by NFAT-GFP (left), NFκB-CFP (middle), and AP-1-mCherry (right) activities.

[0019] Figure 6A-6G Neolinked neoantigens are endogenously processed and presented via MHC-I to induce neoantigen-specific TCR activation and tumor-specific killing. A. Overview of the workflow for validating endogenous protease cleavage of neoantigen candidates and subsequent MHC-I binding and presentation using two methods. The mRNA encoding the full-length gene carrying the neolinked mutation and HLA-A02:01 was electroporated into HLA-deficient antigen-presenting cells (APCs; COS7 or K562). APCs were first co-cultured with Jurkat76 cells transduced with a neolinked neoantigen-specific TCR, and TCR activation was validated by flow cytometry analysis of TCR readouts. IP-MS / MS was performed on transfected APCs to validate the MHC-I-binding peptide for HLA-A02:01. The peptide bound to 0201 was sequenced. BC.NJ GNAS -Derivatives (B) and NJ RPL22 - Jurkat76 cells transduced with a neoantigen-specific TCR derived from (C) were co-cultured with transfected COS7 cells. The cultured cells were divided into the untransfected group (left) and the transfected group (right) containing the neoantigen n-mer sequence and HLA-A. 02:01 mRNA genome (in the middle), or transfected with a peptide encoding the full-length (FL) mutant peptide and HLA-A 02:01 mRNA genome (right). TCR activation of the triple reporter Jurkat76 transduced by NFAT-GFP was measured by flow cytometry. D. By HLA-A 02:01 NJ detected by IP-MS / MS after being pulled down GNAS -Derivatives (Part 2) and NJ RPL22 -Mass spectrum of the derived (top) neoantigen n-mer. E.NJ GNAS -Derivatives (above) and NJ RPL22 -Derived (below) neoantigen-specific TCR-transduced Jurkat76 cells were co-cultured with glioma cell lines, and TCR activation was assessed by flow cytometry analysis of NFAT-GFP expression. F.NJ RPL22 -Derivative (left; colored), NJ GNAS -Derived (right; colored) neoantigen-specific TCR-transduced or non-transduced (gray) CD8 +T cells were co-cultured with GBM115 tumor cells. Cell index indicated the adhesion status of tumor cells on the xCELLigence plate platform; a decreased cell index indicated tumor death. The E:T ratio was measured at a 1:1 (top) or 2:1 (bottom) ratio. G.GBM115 tumor cells were blocked with HLA-A2 antibody (yellow) or an allotype control (purple), or loaded with 1 nM of the corresponding neoantigen peptide (blue). NJ RPL22 -Derivatives (above) and NJ GNAS -Derived (below) neoantigen-specific TCR transduced CD8 + T cells were co-cultured with these GBM115 tumor cells to show whether tumor-specific killing is mediated by HLA-A2 presentation.

[0020] Figure 7A-7LGlioma-specific disease subtypes show different levels of neoconnection expression. AB. Density plots and box plots depict the total number of putative neoconnections expressed in IDH1mut cases (orange) and IDH1wt cases (green) in the TCGA GBM / LGG sample (A) and the GBM / LGG dataset with internal spatial mapping (B). CD. Histograms and box plots depict the total number of putative neoconnections expressed in IDH1wt (blue), astrocytoma (yellow), and oligodendroglioma cases (red) in the TCGA GBM / LGG sample (C) and the GBM / LGG dataset with internal spatial mapping (D). EF. Volcano plots illustrate the sets of genes that were significantly upregulated (blue) and downregulated (red) when comparing IDH1mutO cases with IDH1wt cases (left), IDH1mutA cases with IDH1wt cases (middle), and IDH1mutO cases with IDH1mutA cases (right). The analysis examined gene sets categorized under Gene Ontology Biological Processes (GOBP, E) and Gene Ontology Cellular Components (GOCC, F). Dots labeled with text represent splicing-related gene sets. G. Box plots depict splicing-related genes detected from the GOBP and GOCC gene sets above, showing a significant (p<0.05) log2 fold increase in expression of 1.5 between IDH1mutA (yellow) and IDH1mutO (blue) cases compared to the IDH1wt case (red). H. Box plots depict splicing-related genes on chromosome 1p or chromosome 19q detected from the GOBP and GOCC gene sets above, showing a significant (p<0.05) log2 fold decrease in expression of 1.5 between IDH1mutA (yellow) and IDH1wt cases (red) compared to the IDH1mutO (blue) case. I. Pearson correlation analysis of glioma-specific neolinks relative to CELF2 (left) and SNRPD2 (right) expression in IDH1mutO (z-axis), IDH1mutA (y-axis), and IDH1wt (x-axis) cases. Neolinks with a Pearson correlation greater than or equal to 0.10 with the corresponding gene are indicated by purple dots, and neolinks with a Pearson correlation less than or equal to -0.10 with the corresponding gene are indicated by yellow dots. J. Scatter plot showing splice junctions (CPMs) with the highest mean positive correlation coefficient with CELF2 (left) or the highest mean negative correlation coefficient with SNRPD2 (right). K. NJ in LGG cell lines SF10417 (left) and SF10602 (right) treated with control siRNA or siCELF2. ACAP2 Expression of NJ in GBM115 cells treated with control siRNA or siSNRPD2. ACAP2 The expression.

[0021] Figure 8 The sequences of the neoantigen peptides identified in this study (i.e., GNAS neoantigen, RPL22 neoantigen (9-mer), and RPL22 neoantigen (10-mer)) and the sequences of the TCRs binding to these antigens (i.e., GNAS TCR and RPL22 TCR) and their CDRs (as defined by the IMGT system) are shown.

[0022] Figure 9A T-cell receptors (TCRs) specifically react with neoantigens derived from newly linked tissues. A. After two rounds of NJ... GNAS After in vitro sensitization, HLA-A 02:01 Total CD8+ T cells from healthy donors (left) and glioma patients (right) undergo NeoA testing. GNAS Dextramer staining.

[0023] Figures 10A-10B The novel antigen, derived from a new linker, is endogenously processed and presented by HLA to elicit neoantigen-specific T cells in patients and induce TCR-dependent tumor-specific killing. A. Untransduced or transduced NJ GNAS CD8+ T cells with TCR versus GBM39 cells (left) or transduced HLA-A GBM39 cells (right) were co-cultured at 02:01. The xCELLigence real-time cytotoxicity assay was used to detect the interaction between CD8+ T cells and GBM39 cells (left) or HLA-A transduced cells. 02:01 co-culture of GBM39 cells (right). B. When co-cultured with tumor cell lines, NJ GNAS -TCR-transduced (purple) CD8 + T cells or untransduced (gray) CD8 + Bar graph showing the surface expression of CD107a and CD137 on T cells.

[0024] definition

[0025] As used herein, the term "treatment" refers to obtaining a desired pharmacological and / or physiological effect and / or a treatment-related response. This effect may be preventative, i.e., complete or partial prevention of the disease or its symptoms, and / or therapeutic, i.e., partial or complete cure of the disease and / or adverse effects attributable to the disease. As used herein, "treatment" includes any treatment of a disease in mammals (especially humans) and includes: (a) preventing the disease from occurring in subjects who may be susceptible to the disease but have not yet been diagnosed with it; (b) inhibiting the disease, i.e., preventing its development; and (c) alleviating the disease, i.e., causing the remission of the disease.

[0026] "Therapeutic effective amount" or "effective amount" refers to the amount of a reagent (including biological agents, such as cells), or a combination of two reagents, sufficient to achieve such treatment against a disease when administered to a mammal or other subject. The "therapeutic effective amount" will vary depending on the reagent, the disease and its severity, and the age, weight, etc., of the subject to be treated.

[0027] The terms “individual,” “subject,” “host,” and “patient,” used interchangeably herein, refer to mammals, including but not limited to rodents (e.g., rats, mice), non-human primates, humans, canines, felines, ungulates (e.g., horses, cattle, sheep, pigs, goats), rabbits, etc. In some cases, the individual is a human. In some cases, the individual is a non-human primate. In some cases, the individual is a rodent, such as a rat or mouse. In some cases, the individual is a rabbit, such as a rabbit.

[0028] As used in this article, the term "refractory" refers to a disease or condition that does not respond to treatment. Regarding cancer, as used in this article, "refractory cancer" refers to cancer that does not respond to treatment. Refractory cancer may be resistant at the start of treatment or may become resistant during treatment. Refractory cancer is also known as drug-resistant cancer.

[0029] The terms “T-cell receptor” and “TCR” are used interchangeably and generally refer to a molecule found on the surface of T cells or T lymphocytes that is responsible for recognizing antigen fragments as peptides that bind to the major histocompatibility complex (MHC) molecule. The TCR complex is a disulfide-linked membrane-anchored heterodimeric protein, typically composed of highly variable alpha (α) and beta (β) chains, expressed as part of a complex with the CD3 chain molecule. Many natural TCRs exist as heterodimers in αβ or γδ form. The complete endogenous TCR complex in its heterodimer αβ form comprises eight chains: an α chain (referred to herein as TCRα or TCR alpha), a β chain (referred herein as TCRβ or TCR beta), a δ chain, a γ chain, two ε chains, and two ζ chains. In some cases, TCR is often referred to by reference only to the TCRα and TCRβ chains; however, since assembled TCR complexes can associate with endogenous δ, γ, ε and / or ζ chains, it will be readily understood by those skilled in the art that references to TCRs present in the cell membrane can include appropriate references to fully or partially assembled TCR complexes.

[0030] Recombinant or engineered individual T-cell receptor (TCR) chains and TCR complexes have been developed. In a therapeutic context, the reference to the use of TCRs may refer to individual recombinant TCR chains. Therefore, engineered TCRs may include individually modified TCRα or modified TCRβ chains, as well as single-chain TCRs comprising modified and / or unmodified TCRα and TCRβ chains linked into a single polypeptide via adaptor peptides.

[0031] The terms MHC and HLA are synonymous.

[0032] The term "binding" refers to a direct association between two molecules due to interactions such as covalent, electrostatic, hydrophobic, and ionic and / or hydrogen bonding (including interactions such as salt bridges and water bridges). Non-specific binding refers to a binding with a specific bond strength of less than about 10-1. -7 The affinity of M, for example, has an affinity of 10 -6 M, 10 -5 M, 10 -4 The combination of affinity for M, etc.

[0033] The terms “domain” and “motif”, used interchangeably herein, refer to structured domains that have one or more specific functions and unstructured segments of peptides that, although unstructured, retain one or more specific functions. For example, a structured domain may include, but is not limited to, multiple or discontinuous amino acids or portions thereof in a folded peptide that constitute a three-dimensional structure that contributes to the specific function of the peptide. In other cases, a domain may include unstructured peptide segments comprising multiple two or more amino acids or portions thereof that retain the unfolded or disordered nature of the peptide’s specific function. This definition also includes domains that may be disordered or unstructured but become structured or ordered upon association with a target or binding partner. For example, non-limiting examples of endogenous unstructured domains and domains of endogenous unstructured proteins are described in Dyson & Wright. Nature Reviews Molecular Cell Biology 6:197-208.

[0034] As used generally in this article, the terms "synthetic," "chimeric," and "engineered" refer to non-naturally occurring, artificially derived polypeptides or polypeptide-encoded nucleic acids. Synthetic polypeptides and / or nucleic acids can be assembled de novo from basic subunits (including, for example, a single amino acid, a single nucleotide, etc.) or can be derived from pre-existing polypeptides or polynucleotides (whether naturally or artificially derived), for example, through recombinant methods. Chimeric and engineered polypeptides or polypeptide-encoded nucleic acids are typically constructed by combining, linking, or fusing two or more different polypeptides or polypeptide-encoded nucleic acids or polypeptide domains or polypeptide domain-encoded nucleic acids. Chimeric and engineered polypeptides or polypeptide-encoded nucleic acids include portions in which two or more linked polypeptide or nucleic acid "parts" are derived from different proteins (or nucleic acids encoding different proteins) and portions in which the linked parts comprise different regions of the same protein (or nucleic acid encoding a protein) but are linked in a manner not found in nature.

[0035] As used herein, the term "recombinant" describes nucleic acid molecules, such as genomes, cDNA, viruses, and semi-synthetic and / or synthetically derived polynucleotides, which, due to their origin or manipulation, are not associated with all or part of the polynucleotide sequence to which they are associated in nature. Regarding proteins or peptides, the term "recombinant" means a peptide produced by the expression of a recombinant polynucleotide. Regarding host cells or viruses, the term "recombinant" means a host cell or virus to which a recombinant polynucleotide has been introduced. This document also uses "recombinant" to refer to materials (e.g., cells, nucleic acids, proteins, or vectors) that have been modified by the introduction of a heterologous material (e.g., cells, nucleic acids, proteins, or vectors).

[0036] The term "operably linked" refers to juxtaposition, where the components described so far are in a relationship that allows them to function in the intended manner. For example, if a promoter affects the transcription or expression of a coding sequence, then the promoter is operably linked to the coding sequence. Operablely linked nucleic acid sequences can be adjacent but do not necessarily have to be adjacent. For example, in some cases, the coding sequence operably linked to the promoter may be adjacent to the promoter. In some cases, the coding sequence operably linked to the promoter may be separated by one or more spacer sequences (including coding and non-coding sequences). Additionally, in some cases, more than two sequences can be operably linked, including but not limited to cases where, for example, two or more coding sequences are operably linked to a single promoter.

[0037] The terms “polynucleotide” and “nucleic acid”, used interchangeably in this document, refer to polymers of nucleotides of any length, whether ribonucleotides or deoxyribonucleotides. Therefore, the term includes, but is not limited to, single-stranded, double-stranded, or multi-stranded DNA or RNA, genomic DNA, cDNA, DNA-RNA hybrids, or polymers containing purine and pyrimidine bases or other natural, chemically or biochemically modified, non-natural, or derivatized nucleotide bases.

[0038] The terms “polypeptide,” “peptide,” and “protein,” used interchangeably herein, refer to a polymer of amino acids of any length, which may include genetically encoded and non-genetically encoded amino acids, chemically or biochemically modified or derived amino acids, and polypeptides having a modified peptide backbone. The term includes fusion proteins, including but not limited to fusion proteins having heterologous amino acid sequences, fusion proteins having heterologous and homologous leader sequences, fusion proteins with or without an N-terminal methionine residue; immunolabeled proteins, etc.

[0039] A "vector" or "expression vector" is a replicon, such as a plasmid, bacteriophage, virus, or kinase, to which another DNA fragment (i.e., an "insertion") can attach to the replicon in order to replicate the attached fragment in the cell.

[0040] As used in this article, the term "heterologous" refers to a nucleotide or polypeptide sequence that is not present in naturally occurring (e.g., naturally occurring) nucleic acids or proteins. Heterologous nucleic acids or polypeptides can originate from a species different from the organism or cell in which the nucleic acid or polypeptide is present or expressed. Therefore, heterologous nucleic acids or polypeptides typically have a different evolutionary origin compared to the cell or organism in which they reside.

[0041] As used herein, "therapeuticly effective amount" refers to an amount of a therapeutic agent (e.g., sensitized T-cell infusion, peptide or polynucleotide vaccine, etc.) sufficient to treat or manage a disease or condition. A therapeutically effective amount may refer to an amount of a therapeutic agent sufficient to delay or minimize the onset of a disease (e.g., delay or minimize the spread of cancer), or an amount that effectively increases or decreases signaling from a receptor of interest. A therapeutically effective amount may also refer to an amount of a therapeutic agent that provides therapeutic benefit in the treatment or management of a disease. Furthermore, the therapeutically effective amount of the therapeutic agents of this invention refers to the amount of a single therapeutic agent or a therapeutic agent in combination with other therapies that provides therapeutic benefit in the treatment or management of a disease.

[0042] As used herein, the term "dosing regimen" refers to a set of unit doses (usually more than one) administered individually to a subject, typically separated by time periods. In some embodiments, a given therapeutic agent has a recommended dosing regimen, which may involve one or more doses. In some embodiments, a dosing regimen includes multiple doses, each separated from the others by time periods of equal length; in some embodiments, a dosing regimen includes multiple doses and at least two distinct time periods separating the individual doses. In some embodiments, all doses within a dosing regimen are amounts of the same unit dose. In some embodiments, the different doses within a dosing regimen are different amounts. In some embodiments, a dosing regimen includes a first dose of an amount of a first dose, followed by an additional dose of one or more amounts of a second dose, the amount of which differs from the amount of the first dose. In some embodiments, a dosing regimen includes a first dose of an amount of a first dose, followed by an additional dose of one or more amounts of a second dose, the amount of which is the same as the amount of the first dose. In some embodiments, when administered in a relevant population, the dosing regimen is associated with a desired or beneficial outcome (i.e., a therapeutic dosing regimen).

[0043] As used herein, the terms “cancer” (or “cancerous”) or “tumor” are used to refer to cells with autonomous growth capacity (e.g., an abnormal state or condition characterized by rapidly proliferating cell growth). Proliferative and neoplastic disease states can be classified as pathological (e.g., characterizing or constituting a disease state) or non-pathological (e.g., as deviations from normal but unrelated to a disease state). These terms are intended to encompass all types of cancerous growth or carcinogenic processes, metastatic tissue, or malignant transformation of cells, tissues, or organs, regardless of their histopathological type or stage of invasion. Pathological proliferative cells occur in disease states characterized by malignant tumor growth. Examples of non-pathological proliferative cells include cell proliferation associated with wound healing. The terms “cancer” or “tumor” are also used to refer to malignant tumors of various organ systems, including those affecting the lungs, breasts, thyroid glands, lymph nodes and lymphatic tissues, gastrointestinal organs and genitourinary tracts, as well as adenocarcinoma, which is generally considered to include most colon cancers, renal cell carcinomas, prostate cancers and / or testicular tumors, non-small cell lung cancers, small bowel cancers and esophageal cancers.

[0044] The term "cancer" is generally accepted in this field and refers to malignant tumors of epithelial or endocrine tissues, including respiratory cancers, gastrointestinal cancers, genitourinary cancers, testicular cancers, breast cancers, prostate cancers, endocrine cancers, and melanomas. "Adenocarcinoma" refers to cancer originating from glandular tissue or cancer in which tumor cells form identifiable glandular structures.

[0045] Exemplary cancer types include, but are not limited to, AML, ALL, CML, adrenocortical carcinoma, anal cancer, aplastic anemia, bile duct cancer, bladder cancer, bone cancer, bone metastases, brain cancer, central nervous system (CNS) cancer, peripheral nervous system (PNS) cancer, breast cancer, cervical cancer, childhood non-Hodgkin lymphoma, colorectal cancer, endometrial cancer, esophageal cancer, Ewing's family cancers (e.g., Ewing's sarcoma), eye cancer, gallbladder cancer, gastrointestinal carcinoid tumors, gastrointestinal stromal tumors, gestational trophoblastic disease, Hodgkin lymphoma, Kaposi's sarcoma, kidney cancer, laryngeal and hypopharyngeal cancer, liver cancer, lung cancer, lung carcinoid tumors, non-Hodgkin lymphoma, and male cancer. Breast cancer, malignant mesothelioma, multiple myeloma, myelodysplastic syndrome, myeloproliferative disorders, nasal and paranasal sinus cancer, nasopharyngeal carcinoma, neuroblastoma, oral and oropharyngeal cancer, osteosarcoma, ovarian cancer, pancreatic cancer, penile cancer, pituitary tumor, prostate cancer, retinoblastoma, rhabdomyosarcoma, salivary gland cancer, sarcoma, cutaneous melanoma, non-melanoma skin cancer, gastric cancer, testicular cancer, thymic cancer, thyroid cancer, uterine cancer (e.g., uterine sarcoma), transitional cell carcinoma, vaginal cancer, vulvar cancer, mesothelioma, squamous cell carcinoma or epidermoid carcinoma, bronchial adenoma, choriocarcinoma, head and neck cancer, teratoma, or Waldenstrom's macroglobulinemia.

[0046] Before further describing the invention, it should be understood that the invention is not limited to the specific embodiments described herein, and of course, these embodiments can be varied. It should also be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting, as the scope of the invention will be limited only by the appended claims.

[0047] When numerical ranges are provided, it should be understood that, unless the context explicitly specifies otherwise, every intermediate value between the upper and lower limits of the range up to one-tenth of the lower limit unit, as well as any other value or intermediate value within the range, is included within the scope of this invention. The upper and lower limits of these smaller ranges may be independently included within that smaller range and are also included within the scope of this invention, subject to any explicitly excluded boundaries within the range. When the range includes one or two boundaries, the range excluding any one or both of those inclusion boundaries is also included within the scope of this invention.

[0048] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. While any similar or equivalent methods and materials described herein may be used to practice or test the invention, preferred methods and materials are described here. All publications mentioned herein are incorporated by reference to disclose and describe the methods and / or materials cited in their respective publications.

[0049] It must be noted that, as used herein and in the appended claims, the singular forms “a,” “an,” and “the” include plural references unless the context clearly specifies otherwise. Thus, for example, a reference to “a cell” includes multiple such cells, a reference to “the cell” includes one or more cells and their equivalents known to those skilled in the art, and so on. It should also be noted that claims can be drafted to exclude any optional elements. Therefore, this statement is intended as a basis for using exclusive terms such as “solely,” “only,” or negative limitations in the description of claim elements.

[0050] It should be understood that, for clarity, certain features of the invention described in the context of different embodiments may also be provided in combination in a single embodiment. Conversely, for brevity, various features of the invention described in the context of a single embodiment may also be provided individually or in any suitable sub-combination. All combinations of embodiments relating to the invention are specifically covered by the invention and disclosed herein, as if each combination were individually and explicitly disclosed herein. Furthermore, all sub-combinations of various embodiments and their elements are also specifically covered by the invention and disclosed herein, as if each such sub-combination were individually and explicitly disclosed herein.

[0051] The publications discussed herein are provided only because of their publication prior to the filing date of this application. Nothing herein should be construed as an admission that the invention is not entitled to any prior invention prior to such publications. Furthermore, the publication dates provided may differ from the actual publication dates, which may require independent verification. Invention Details

[0053] A separated peptide is provided, comprising the amino acid sequence of SLLLPSFHL (SEQ ID NO: 1, referred to herein as the “GNAS” neoantigen) or IMDAANFFL (SEQ ID NO: 2, referred herein as the “RPL22” neoantigen). The length of the peptide may vary depending on how the peptide will be used. In any embodiment, the peptide may be 9 to 100 amino acids long (e.g., 9 to 40 amino acids). In any embodiment, the peptide may be relatively short (e.g., 9, 10, or 11 amino acids long). In any embodiment, the peptide may be relatively long, for example, 15-50 or 18 to 40 amino acids long. In some embodiments, the peptide may consist of SEQ ID NO: 1, 2, or 3.

[0054] In any embodiment, the peptide disclosed herein may be flanked by additional amino acid residues, provided that the peptide retains its TCR inducibility. Such a peptide may be less than about 40 amino acids, for example less than about 20 amino acids, for example less than about 15 amino acids. The amino acid sequence flanking the peptide consisting of the amino acid sequences of SEQ ID NO: 1 and 2 is not limited and may consist of any kind of amino acids, as long as it does not impede TCR recognition. The amino acid sequence may be modified by substituting one or more amino acids. Those skilled in the art will recognize that a single addition or substitution of an amino acid sequence, which alters a single amino acid or a small percentage of amino acids, results in the preservation of the properties of the original amino acid side chain; this is therefore referred to as a “conservative substitution” or “conservative modification,” in which the alteration of the protein results in a protein with similar function.

[0055] In addition to the sequence modifications described above, peptides can be linked to other substances, provided they retain their TCR-binding activity. Available substances include peptides, lipids, sugars and glycans, acetyl groups, natural and synthetic polymers, etc. Peptides can contain modifications such as glycosylation, side-chain oxidation, or phosphorylation; as long as the modifications do not impair the biological activity of the peptide as described herein. Such modifications can be made to impart additional functions (e.g., targeting and delivery functions) or to stabilize the peptide.

[0056] For example, to increase the in vivo stability of peptides, the introduction of particularly useful D-amino acids, amino acid mimics, or non-natural amino acids is known in the art; this concept can also be applied to the peptides of the present invention. The stability of peptides 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).

[0057] The peptides disclosed herein can be prepared using well-known techniques. For example, peptides can be prepared by recombinant DNA technology or chemical synthesis. The peptides disclosed herein can be synthesized alone or as longer polypeptides (e.g., two or more peptides or polypeptides and non-peptides) containing two or more peptides. The peptides can be isolated, i.e. purified to be substantially free of other naturally occurring host cell proteins and their fragments, for example, at least about 70%, 80%, or 90% purification.

[0058] Compositions comprising peptides and pharmaceutically acceptable excipients are also provided, wherein in some embodiments, the composition may further comprise an adjuvant having an immunostimulatory effect.

[0059] In some embodiments, the peptide may be complexed with a major histocompatibility complex (MHC) (i.e., it may be in an MHC-peptide complex). Therefore, in some embodiments, the MHC-peptide complex may comprise a class I MHC and a peptide. In other embodiments, the composition may comprise a class I MHC and a peptide. If the MHC is soluble (this can be generated by expressing an MHC without its transmembrane domain), the complex may be in solution. In other embodiments, the MHC complex may be on a cell surface. Cells expressing such a major histocompatibility complex (MHC) complex on their surface are thus provided. In some embodiments, the cell may be an antigen-presenting cell, such as a professional antigen-presenting cell, such as a dendritic cell. As is apparent, in some embodiments, the MHC and peptide may be recombinantly expressed. In other embodiments, the MHC may be recombinantly expressed and the peptide may be exogenously added. In other embodiments, the peptide may be exogenously added to cells that naturally express MHC. In any embodiment, the cell may be in vitro or in vivo. In embodiments where the peptide is added exogenously, the peptide may be relatively short (e.g., 9, 10, or 11 amino acids long). In these embodiments, the peptide can bind directly to MHC. In other embodiments, a longer peptide (e.g., a peptide with an amino acid range of, for example, 15-50) can be added. In these embodiments, the peptide can be processed by the cell before being bound to MHC. In any embodiment, the MHC can be HLA-A. 02:01 Class I MHC binding domain (this is the most common domain). However, other MHC binding domains should bind to this peptide, so this is not necessary.

[0060] A T-cell receptor is also provided. In some embodiments, this T-cell receptor may be a first (or "GNAS") T-cell receptor that recognizes class I MHCs (especially those with HLA-A receptors) that bind to GNAS neoantigens. 02:01 A complex of class I MHCs (specifically those with HLA-A binding domains). In these embodiments, the T cell receptor may comprise an α chain having the CDR sequences of SEQ ID NO: 5, 6, and 7 and a β chain having the CDR sequences of SEQ ID NO: 9, 10, and 11. In any embodiment, the first T cell receptor may have (a) an α chain variable domain having an amino acid sequence that is at least 90% identical (e.g., at least 95% or at least 98% identical) to SEQ ID NO: 4 and a β chain variable domain having an amino acid sequence that is at least 90% identical (e.g., at least 95% or at least 98% identical) to SEQ ID NO: 8. In other embodiments, the T cell receptor may be a second (or "RPL22") T cell receptor that recognizes class I MHCs (specifically those with HLA-A binding domains) that bind to the RPL22 neoantigen. 02:01 A complex of class I MHC binding domains. In these embodiments, the T cell receptor may comprise an α chain having the CDR sequences of SEQ ID NO: 13, 14, and 14 and a β chain having the CDR sequences of SEQ ID NO: 17, 18, and 19. In any embodiment, the second T cell receptor may have (a) an α chain variable domain having an amino acid sequence that is at least 90% identical (e.g., at least 95% or at least 98% identical) to SEQ ID NO: 12 and a β chain variable domain having an amino acid sequence that is at least 90% identical (e.g., at least 95% or at least 98% identical) to SEQ ID NO: 16. It should be noted that the binding between the TCR and the MHC complex is primarily mediated by the CDR3 domains of the α and β chains. These sequences are shown in Figure 8 middle.

[0061] It also provides engineered immune cells that express the first or second T cell receptor. In some cases, the immune cell is a mammalian cell, particularly a human, mouse, or non-human primate cell. Suitable mammalian immune cells include primary cells and immortalized cell lines. Suitable mammalian cell lines include human cell lines, non-human primate cell lines, rodent (e.g., mouse, rat) cell lines, etc. In some instances, the cell is not an immortalized cell line but a cell obtained from an individual (e.g., a primary cell). For example, in some cases, the cell is an immune cell, immune cell progenitor cell, or immune stem cell obtained from an individual. As an example, the cell is a lymphoid cell, such as a lymphocyte, or its progenitor cell, obtained from an individual. As another example, the cell is a killer cell, or its progenitor cell, obtained from an individual.

[0062] Such cells include, for example, lymphoid cells, namely lymphocytes (T cells, B cells, natural killer (NK) cells), and myeloid-derived cells (neutrophils, eosinophils, basophils, monocytes, macrophages, dendritic cells). "T cells" include all types of immune cells that express CD3, including helper T cells (CD4+ cells) and cytotoxic T cells (CD8+ cells). "Cytotoxic cells" include CD8+ T cells, natural killer (NK) cells, and neutrophils, which can mediate cytotoxic responses.

[0063] These cells can be generated by any convenient method. Nucleic acids encoding one or more components of the test circuit can be stably or transiently introduced into the test immune cells, including cases where the test nucleic acid is only temporarily present, maintained extrachromosomally, or integrated into the host genome. The introduction of the test nucleic acid and / or the genetic modification of the test immune cells can be performed in vivo, in vitro, or ex vivo.

[0064] It also provides RNA or DNA encoding peptides.

[0065] In some cases, the test nucleic acid is introduced and / or genetically modified in vitro. For example, primary T lymphocytes, stem cells, or NK cells are obtained from an individual; and the cells obtained from the individual are modified to express the components of the present invention. In some embodiments, heterologous immune cells may be used.

[0066] Given the known genetic code, the sequence encoding any of the proteins of this invention can be readily determined. In some embodiments, the coding sequence may be codon-optimized for expression in mammalian (e.g., human or mouse) cells, and strategies for this are well known (see, for example, Mauro et al., Trends Mol. Med. 2014 20: 604-613 and Bell et al., Human Gene Therapy Methods 27: 6). As understood, the coding sequence can be operatively linked to a promoter, which may be inducible, tissue-specific, or constitutive. In some embodiments, the promoter may be activated by a cellularly heterologous engineered transcription factor, such as Gal4-, LexA-, Tet-, Lac-, dCas9-, zinc finger-, and TALE-based transcription factors. In some embodiments, the peptide may be synthesized synthetically using a peptide synthesizer or the like.

[0067] A promoter can be a constitutively active promoter (i.e., a promoter that is constitutively active / “on”), an inducible promoter (i.e., whose state, active / “on” or inactive / “off”, is controlled by external stimuli, such as a specific temperature, the presence of a compound or protein), a spatially restricted promoter (i.e., transcriptional regulatory elements, enhancers, etc.) (e.g., tissue-specific promoters, cell-type-specific promoters, etc.), or a time-restricted promoter (i.e., the promoter is “on” or “off” during a specific stage of embryonic development or a specific stage of a biological process, such as the hair follicle cycle in mice).

[0068] For expression in eukaryotic cells, suitable promoters include, but are not limited to, light chain and / or heavy chain immunoglobulin gene promoters and enhancer elements; cytomegalovirus immediate early promoters; herpes simplex virus thymidine kinase promoters; SV40 early and late promoters; promoters present in long terminal repeat sequences of retroviruses; mouse metallothionein-I promoters; and various known tissue-specific promoters.

[0069] Suitable reversible promoters, including reversibly inducible promoters, are known in the art. Such reversible promoters can be isolated and derived from many organisms, such as eukaryotes and prokaryotes. Modifying reversible promoters derived from a first organism for use in a second organism, such as a first prokaryote and a second eukaryote, or vice versa, is well known in the art. Such reversible promoters and systems based on such reversible promoters but also including additional regulatory proteins include, but are not limited to, alcohol-regulated promoters (e.g., promoters of alcohol dehydrogenase I (alcA) gene, promoters responsive to alcohol transactivator protein (AlcR), etc.), tetracycline-regulated promoters (e.g., promoter systems including TetActivator, TetON, TetOFF, etc.), steroid-regulated promoters (e.g., rat glucocorticoid receptor promoter system, human estrogen receptor promoter system, retinoic acid promoter system, thyroid promoter system, ecdysone promoter system, mifepristone promoter system, etc.), metal-regulated promoters (e.g., metallothionein promoter system, etc.), pathogen-associated regulatory promoters (e.g., salicylic acid-regulated promoters, ethylene-regulated promoters, benzothiadiazole-regulated promoters, etc.), temperature-regulated promoters (e.g., heat shock-inducible promoters (e.g., HSP-70, HSP-90, soybean heat shock promoter, etc.), light-regulated promoters, synthesis-inducible promoters, etc.).

[0070] Suitable inducible promoters include any inducible promoters described herein or known to those skilled in the art. Examples of inducible promoters include, but are not limited to, chemical / biochemical and physical regulatory promoters, such as alcohol-regulated promoters, tetracycline-regulated promoters (e.g., anhydrous tetracycline (aTc) responsive promoters and other tetracycline-responsive promoter systems, including tetracycline repressor protein (tetR), tetracycline operon sequence (tetO), and tetracycline transactivation fusion protein (tTA)), steroid-regulated promoters (e.g., promoters based on rat glucocorticoid receptor, human estrogen receptor, moth ecdysone receptor, and promoters from the steroid / retinoic acid / thyroid receptor superfamily), metal-regulated promoters (e.g., promoters derived from yeast, mouse, and human metallothionein (proteins that bind to and chelate metal ions) genes), pathogen-associated regulatory promoters (e.g., induced by salicylic acid, ethylene, or benzothiadiazole (BTH)), temperature / heat-induced promoters (e.g., heat shock promoters), and light-regulated promoters (e.g., light-responsive promoters in plant cells).

[0071] In some cases, the promoter is a CD8 cell-specific promoter, a CD4 cell-specific promoter, a neutrophil-specific promoter, or an NK cell-specific promoter. For example, the CD4 gene promoter can be used; see, for example, Salmon et al. (1993) Proc. Natl. Acad. Sci. USA 90: 7739; and Marodon et al. (2003) Blood 101:3416. As another example, the CD8 gene promoter can be used. NK cell-specific expression can be achieved by using the Ncr1(p46) promoter; see, for example, Eckelhart et al. (2011) Blood 117:1565.

[0072] In some cases, the promoter is a cardiomyocyte-specific promoter. In some cases, the promoter is a smooth muscle cell-specific promoter. In some cases, the promoter is a neuron-specific promoter. In some cases, the promoter is an adipocyte-specific promoter. Other cell type-specific promoters are known in the art and are suitable for use herein.

[0073] Suitable expression vectors include, but are not limited to, viral vectors (e.g., vaccinia virus-based; poliovirus-based; adenovirus-based viral vectors (see, for example, Li et al., Invest Opthalmol Vis Sci 35:2543 2549, 1994; Borras et al., Gene Ther 6:515 524, 1999; Li and Davidson, PNAS 92:7700 7704, 1995; Sakamoto et al., Hum Gene Ther 5:1088 1097, 1999; WO 94 / 12649, WO 93 / 03769; WO93 / 19191; WO 94 / 28938; WO 95 / 11984 and WO 95 / 00655); adeno-associated viruses (see, for example, Ali et al., Hum Gene Ther 9:81 86, 1998, ...). Flannery et al., PNAS 94:6916 6921, 1997; Bennett et al., Invest Opthalmol Vis Sci 38:2857 2863, 1997; Jomary et al., Gene Ther 4:683 690, 1997; Rolling et al., Hum Gene Ther 10:641 648, 1999; Ali et al., Hum MolGenet 5:591 594, 1996; Srivastava, Samulski et al., J. Vir. (1989) 63:3822-3828 in WO 93 / 09239; Mendelson et al., Virol. (1988) 166:154-165; and Flotte et al., PNAS (1993). 90:10613-10617); SV40; herpes simplex virus; human immunodeficiency virus (see, for example, Miyoshi et al., PNAS 94:10319 23, 1997; Takahashi et al., J Virol 73:7812 7816, 1999); retroviral vectors (e.g., murine leukemia virus, spleen necrosis virus, and vectors derived from retroviruses such as Rous sarcoma virus, Harvey sarcoma virus, avian leukemia virus, lentivirus, human immunodeficiency virus, myeloproliferative sarcoma virus, and mammary tumor virus); etc. In some cases, the vector is a lentiviral vector. Transposon-mediated vectors, such as piggyback and sleeping beauty vectors, are also suitable.

[0074] The peptide and T-cell receptor can be used to sensitize T cells in vitro and in vivo, and for various vaccination / treatment methods, etc. Specifically, if the peptide is combined with an antigen-presenting cell in vitro, the cell can be used to select and expand immune cells that recognize the MHC complex in vivo or in vitro. The method may include contacting an antigen-presenting cell (such as dendritic cells) having an MHC complex containing the peptide on its surface with a population of T cells, and culturing the T cells with the T cells under conditions in which T cells recognizing the MHC-peptide complex are activated and / or expanded.

[0075] In some embodiments, the method can induce an immune response. The method may include administering an effective dose of a peptide or a nucleic acid (DNA or RNA) encoding the peptide, or the compositions or cells described above, to an individual. These embodiments may include administering the peptide to an individual. Again, in these embodiments, the peptide may be relatively short (e.g., 9, 10, or 11 amino acids in length). In these embodiments, the peptide may bind directly to MHC, for example, in antigen-presenting cells. In other embodiments, longer peptides (e.g., peptides with a length of, for example, 15-50 amino acids) may be added, in which case they may be treated prior to presentation by MHC. In these embodiments, the peptide, the APC presenting the peptide, and the T cells recognizing the MHC-peptide complex containing the peptide can be used as readily available cancer vaccines because neoantigens are present in so many cancers. In alternative embodiments, RNA may be administered.

[0076] In some embodiments, the method may be a treatment method. In these embodiments, the method may include administering the peptides, compositions, or cells described above to an individual in need. In these embodiments, the individual may have glioblastoma, low-grade glioma, lung adenocarcinoma, lung squamous cell carcinoma, mesothelioma, hepatocellular carcinoma, gastric adenocarcinoma, clear cell renal carcinoma, papillary renal carcinoma, chromophobe renal carcinoma, colonic adenocarcinoma, or prostate adenocarcinoma.

[0077] In some embodiments, the method is used to treat cancer (e.g., reduce tumor cell growth, promote tumor cell death) by administering to an individual a peptide disclosed herein or a polynucleotide encoding the peptide, or a T cell modified as described above. In a related aspect, isolated sensitized T cells that have already been sensitized with the peptide disclosed herein can be used. In another aspect, antigen-presenting cells are provided, which comprise a complex formed between an HLA antigen and the peptide disclosed herein. In some embodiments, the antigen-presenting cells are isolated.

[0078] If desired, pharmaceutical preparations containing peptides or polynucleotides encoding peptides may optionally include other therapeutic substances as active ingredients, provided that such substances do not inhibit the TCR-stimulatory effect of the peptide of interest. For example, formulations may include anti-inflammatory agents, analgesics, chemotherapeutic agents, etc. In addition to including other therapeutic substances in the drug itself, the drug may be administered sequentially or simultaneously with one or more other pharmacological agents. The amounts of the drug and pharmacological agents depend on, for example, the type of pharmacological agent used, the disease being treated, and the plan and route of administration.

[0079] If necessary, the composition can be administered directly as a pharmaceutical preparation, formulated using conventional methods. In such cases, in addition to peptides, it may appropriately include carriers, excipients, etc., commonly used in pharmaceuticals, without particular limitation. Examples of such carriers include sterile water, physiological saline, phosphate buffer, culture medium, etc. Furthermore, the pharmaceutical preparation may contain stabilizers, suspending agents, preservatives, surfactants, etc., as needed. The pharmaceutical preparation can be used to treat and / or prevent cancer.

[0080] Peptides can be prepared into combinations. Using standard techniques, peptides can be in mixtures or conjugated together.

[0081] The compositions may optionally contain adjuvants that enable the effective establishment of cellular immunity, or they may be administered together with other active ingredients and may be administered by formulation into granules. An adjuvant is a compound that enhances the immune response against an immunologically active protein when administered co-administered (or sequentially) with that protein. Applicable adjuvants include those described in the literature. Exemplary adjuvants include, but are not limited to, aluminum phosphate, aluminum hydroxide, alum, cholera toxin, salmonella toxin, etc.

[0082] In addition, liposome formulations, particulate formulations in which peptides are bound to beads a few micrometers in diameter, and formulations in which lipids are bound to peptides are readily available. Alternatively, intracellular vesicles called exosomes are provided, which present a complex formed between the peptide and the HLA antigen on their surface. Exosomes can be used as vaccines, similar to peptides.

[0083] In some embodiments, the compositions of the present invention may comprise components sensitized to T lymphocytes. Lipids have been identified as agents capable of sensitizing CTLs in vivo against viral antigens. For example, palmitic acid residues may be linked to the ε- and α-amino groups of lysine residues, and then linked to a peptide disclosed herein. The lipotropic peptide can then be administered directly in microparticle or particulate form, incorporated into liposomes, or emulsified in an adjuvant. As another example of lipid sensitization for CTL responses, E. coli lipoproteins, such as tripalmitoyl-S-glycerol cysteyl lysyl serine (P3CSS), can be used to sensitize CTLs when covalently linked to a suitable peptide (see, for example, Deres et al., Nature 342: 561, 1989).

[0084] The administration method can be oral, intradermal, subcutaneous, intravenous, etc., and systemic or local application near the target site is useful. Administration can be performed as a single dose or through multiple doses for enhancement. The dosage of the peptide can be appropriately adjusted according to the disease to be treated, the patient's age, weight, method of administration, etc., and is typically from 0.001 mg to 1000 mg, for example, 0.1 mg to 10 mg, and can be administered from once every few days to once every few months. Those skilled in the art can appropriately select the appropriate dosage.

[0085] The pharmaceutical agents disclosed herein may also comprise nucleic acids encoding the disclosed peptides in an expressible form. The term "in an expressible form" means that, when introduced into cells, the polynucleotide will be expressed in vivo as a polypeptide with stimulating antitumor immunity. In one embodiment, the nucleic acid sequence of the polynucleotide of interest includes regulatory elements necessary for the expression of the polynucleotide in target cells. The polynucleotide may be configured to stably insert into the genome of the target cell (see, for example, Thomas KR and Capecchi MR, Cell 51:503-12, 1987, for describing homologous recombination cassette vectors). See, for example, Wolff et al., Science 247:1465-8, 1990; U.S. Patent Nos. 5,580,859; 5,589,466; 5,804,566; 5,739,118; 5,736,524; 5,679,647; and WO 98 / 04720. Examples of DNA-based delivery technologies include “naked DNA”, facilitated (bupivacaine, polymer, peptide-mediated) delivery, cationic lipid complexes, and particle-mediated (“gene gun”) or pressure-mediated delivery (see, for example, U.S. Patent No. 5,922,687).

[0086] Antigen-presenting cells (APCs) that present complexes formed between MHC and peptides on their surface are also provided. APCs are obtained by contacting cells with peptides or nucleotides encoding said peptides, and can be prepared from subjects targeted for treatment and / or prevention, and can be administered alone or in combination with other drugs, including peptides, exosomes, or cytotoxic T cells, as a vaccine. APCs are not limited to any type of cell and include dendritic cells (DCs), Langerhans cells, macrophages, B cells, and activated T cells, all of which are known to present proteinaceous antigens on their cell surfaces for recognition by lymphocytes. DCs are specifically used as APCs because they are the most representative APCs with the strongest CTL-inducing activity.

[0087] For example, APCs can be obtained by inducing dendritic cells from peripheral blood mononuclear cells and then contacting them with peptides in vitro, ex vivo, or in vivo (stimulation). When the peptide is administered to a subject, APCs with peptides immobilized thereon are stimulated in the subject's body. "Inducing APCs" involves contacting cells with the peptide or nucleotides encoding the peptide (stimulation) to present the complex formed between the HLA antigen and the peptide on the cell surface. Alternatively, after immobilizing the peptide to APCs, the APCs can be administered to the subject as a vaccine. For example, ex vivo administration may include the following steps: a: collecting APCs from the subject, and b: contacting the APCs from step a with the peptide. APCs obtained through step b can be administered to the subject as a vaccine.

[0088] Such APCs can be prepared by a method that includes the step of in vitro transfer of a gene containing a polynucleotide encoding a peptide into the APC. The introduced gene can be in the form of DNA or RNA. There are no particular limitations on the introduction method, and various methods conventionally performed in the field can be used, such as lipid transfection, electroporation, and calcium phosphate methods.

[0089] T cells stimulated by any of the peptides disclosed herein can be used as vaccines similar to the peptides. Therefore, this invention provides isolated T cells stimulated by any of the peptides of this invention. Such T cells can be obtained by (1) administering the peptide to a subject or (2) contacting the peptide in vitro with APCs and CD8-positive cells or peripheral blood mononuclear leukocytes derived from the subject (stimulation). T cells already stimulated by APCs presenting the peptide can be derived from a subject as a target for treatment and / or prevention and can be administered alone or in combination with other drugs, including peptides or exosomes, to achieve a modulating effect. The obtained T cells specifically antagonize the target cells presenting the peptide, for example, the same peptide used for sensitization. Target cells can be endogenously expressed cells or transfected cells, and cells presenting the peptide on their cell surface due to stimulation by these peptides can also be targets of attack. In some embodiments, the T cells are contacted in vitro with the peptide, i.e., the T cells are then transferred to the recipient.

[0090] For purposes of disclosure, effector cells may include autologous or allogeneic immune cells that have cytolytic activity against target cells (including, but not limited to, tumor cells). Effector cells can be obtained by engineering peripheral blood lymphocytes (PBLs) in vitro and then culturing them in combination with increased activation of cytokines and / or antigens. Cells may optionally be separated from unwanted cells before culture, before administration, or both. It is believed that cell-mediated cytolysis of target cells by immune effector cells is mediated by localized, directed exocytosis of cytoplasmic granules that penetrate the cell membrane of the bound target cells.

[0091] Cytotoxic T lymphocytes (CTLs) that respond to tumor cells are specific effector cells for adoptive immunotherapy and are of great significance for engineering by sensitization with the peptides disclosed herein or for engineering TCRs expressed herein. The induction and expansion of CTLs are antigen-specific and MHC-restricted.

[0092] T cells collected from subjects can be isolated from the cell mixture using techniques for enriching the desired cells, or they can be engineered and cultured without isolation. They can be dispersed or suspended using appropriate solutions. Such solutions are typically balanced salt solutions, such as physiological saline, PBS, Hank's balanced salt solution, etc., conveniently supplemented with fetal bovine serum or other naturally occurring factors, bound to low concentrations of acceptable buffers, typically 5–25 mM. Convenient buffers include HEPES, phosphate buffer, lactate buffer, etc.

[0093] Affinity separation techniques can include magnetic separation, the use of antibody-coated magnetic beads, affinity chromatography, cytotoxic agents linked to or used in combination with monoclonal antibodies, such as complement and cytotoxins, and antibody "plating" onto a solid matrix (e.g., a plate), or other convenient techniques. Techniques providing precise separation include fluorescence-activated cell sorters, which can have varying degrees of complexity, such as multiple color channels, low-angle and obtuse-angle light scattering detection channels, impedance channels, etc. Cells can be selected for dead cells by employing dyes associated with dead cells (e.g., propidium iodide). Any technique that does not unduly impair the viability of the selected cells can be used. Affinity reagents can be specific receptors or ligands against the aforementioned cell surface molecules. In addition to antibody reagents, peptide-MHC antigen and T-cell receptor pairs; peptide ligands and receptors; effector molecules and receptor molecules, etc., can also be used.

[0094] The collected and optionally enriched cell populations can be used immediately for genetic modification, or they can be frozen and stored at liquid nitrogen temperature, thawed, and reused. Cells are typically stored in 10% DMSO, 50% FCS, and 40% RPMI 1640 medium.

[0095] Engineered cells can be infused into a subject via any convenient route of administration in any physiologically acceptable medium, typically intravascular, although they can also be introduced via other routes where the cells can find suitable sites for growth. Typically, at least 1 × 10⁻⁶ cells will be administered. 6 Cells / kg, at least 1×10 7 Cells / kg, at least 1×10 8 Cells / kg, at least 1×10 9 Cells / kg, at least 1×10 10 The number of cells per kg or more is typically limited by the number of T cells obtained during collection.

[0096] Filtering methods

[0097] Methods for screening HLA-binding peptides containing novel links are provided. As described herein, novel links are cancer-specific aberrant splicing events. Peptides containing novel links can be understood as translations of transcripts originating from aberrant splicing. Screening methods involve obtaining transcriptome sequence data from tumor tissue samples and normal tissue samples. In some embodiments, transcriptome data can be obtained from multiple tumor tissue samples and multiple normal tissue samples. In some cases, multiple tumor tissue samples may be derived from a single tumor, e.g., from a sliced ​​tumor. In some cases, transcriptome sequence data can be obtained from multiple tumor tissue samples of the same type (e.g., glioblastoma) and multiple normal tissue samples of the same type (e.g., brain tissue). In some cases, transcriptome data can be obtained from multiple tumor tissue samples of different types and multiple normal tissue types. Obtaining transcriptome sequence data from multiple tumor and normal tissue samples may be advantageous when, for example, screening for peptides containing novel links that are conserved among individuals across tumor types or across tumor types. Methods, kits, and reagents for sequencing transcriptomes are known in the art and readily available. Alternatively, transcriptome sequence data repositories for both tumor tissue samples and normal tissue samples are known in the art and readily available. For example, transcriptome sequence data of tumor tissue samples can be obtained from the Cancer Genome Atlas (TCGA) or the Mayo Clinic Brain Tumor Patient Derived Xenograft National Resource. As another example, transcriptome sequence data of normal tissue samples can be obtained from the Genotype Tissue Expression (GTEx) repository. Methods for screening HLA-binding peptides containing novel linkages may also include identifying one or more transcripts containing novel linkages in the tumor tissue sample. In some cases, identifying one or more transcripts containing novel linkages in the tumor tissue sample includes quantifying detected splice linkages in protein-coding transcripts expressed in the tumor tissue sample or in multiple tumor tissue samples (i.e., quantifying sequencing reads containing detected splice linkages), and quantifying detected splice linkages in protein-coding transcripts expressed in normal tissue samples or in multiple normal tissue samples. Methods for detecting splice linkage events (e.g., STAR alignments, IRFinders) are implemented in algorithms well-known in the art and readily available. Identifying one or more transcripts containing newly joined splice junctions in a tumor tissue sample may also include identifying unannotated splice junctions as part of a set of detected splice junctions that are not present in the annotated set of classical splice junctions. Datasets of classical splice junction annotations are well-known in the art and readily available for use with a selected reference genome. For example, classical splice junction annotations are available from the GENCODE database for the GRCh37 human reference genome.Identifying the presence of one or more newly linked transcripts in tumor tissue samples may also include defining the splice frequency of unannotated splice junctions. In some cases, the splice frequency of unannotated splice junctions may be defined as:

[0098]

[0099] In some embodiments, the novel linker is identified as an unannotated splice linker having a splice frequency of at least 0.1 in a tumor tissue sample or in multiple tumor tissue samples, and not detected in normal tissue samples, or not detected in less than 1% of multiple normal tissue samples. The method for screening HLA-binding peptides containing novel links may further include inputting sequences of one or more transcripts containing novel links into a computer-readable medium and generating a computer-simulated library of possible multimeric peptide sequences from the sequences of one or more transcripts containing novel links. In some embodiments, generating the computer-simulated library of possible multimeric peptide sequences includes translating the transcripts containing novel links into their corresponding amino acid sequences via computer simulation. In some embodiments, generating the computer-simulated library of possible multimeric peptide sequences further includes, for each amino acid sequence, partitioning the amino acid sequence into all possible multimeric peptide sequences of a specified length. In some embodiments, the length of the possible multimeric peptide sequences can be in the range of 8-40 amino acids. For example, in some cases, the length of the multimeric peptide sequences can be in the range of 8-9 amino acids, 8-10 amino acids, 8-11 amino acids, 8-12 amino acids, 8-15 amino acids, 8-20 amino acids, 8-25 amino acids, 8-30 amino acids, or 8-40 amino acids. In some embodiments, the computer-simulated library for generating the possible multimeric peptide sequences also includes each multimeric peptide sequence detectable in a reference proteome dataset discarded from normal tissue samples. Reference proteome datasets are known in the art and readily available. Suitable reference proteomes for use in the methods of the present invention may include, but are not limited to, UniProt proteome ID#UP000005640. The method for screening HLA-binding peptides containing newly linked peptides may also include, for each peptide in the computer-simulated multimeric peptide library, generating a first MHC I presentation score using an antigen treatment advantage algorithm and a second MHC I presentation score using a pan-HLA-A binding affinity algorithm. In some cases, it can target HLA-A 01:01, HLA-A 02:01, HLA-A 03:01, HLA-A 11:01 and HLA-A 24:02 Allele assessment of HLA-A binding affinity. In some embodiments, an antigen treatment advantage algorithm can be implemented using MHCFlurry 2.0 software. In some embodiments, a pan-HLA-A binding affinity algorithm can be implemented using HLAthena software. Methods for screening HLA-binding peptides containing newly linked peptides may further include identifying one or more HLA-binding peptides determined by a first MHC-I presentation score and a second MHC-I presentation score. Identifying one or more HLA-binding peptides may include sorting multimeric peptides in a library according to the first MHC-I presentation score and the second MHC-I presentation score. In some embodiments, identifying one or more HLA-binding peptides may further include identifying multimeric peptides in the top 10% of the first MHC-I presentation score and the second MHC-I presentation score as one or more HLA-binding peptides. In some cases, multimeric peptides in the top 1% of the first MHC-I presentation score and the second MHC-I presentation score may be identified as one or more HLA-binding peptides. Figure 1 An exemplary embodiment of a method for screening HLA-binding peptides containing novel linkages is illustrated schematically.

[0100] Methods for screening T cell receptors (TCRs) that recognize and bind to novel linker epitopes are provided. In some embodiments, the method for screening TCRs includes introducing an HLA-binding peptide containing the novel linker into a population of antigen-presenting cells (APCs) to generate a population of APCs loaded with the HLA-binding peptide (i.e., presenting the HLA-binding peptide on the cell surface in combination with MHC-I). In some embodiments, the APC population may be transformed or transduced to recombinantly express a polypeptide containing the novel linker peptide, which is further treated by the APCs and presented on the cell surface in combination with MHC-I. In some embodiments, the APCs may be professional APCs (e.g., dendritic cells). In some cases, the APCs may be deficient in endogenous antigen processing (e.g., T2 cells). The method for screening TCRs may also include contacting a population of T cells with a population of APCs loaded with the HLA-binding peptide. The T cells may be any suitable T cells. In some embodiments, the T cells are naive CD8+ T cells. In some embodiments, the subsequent harvesting of T cells and contacting the T cell population with a population of APCs loaded with HLA-binding peptides can be iterated once or multiple times. In some embodiments, the method for screening TCRs may further include selecting T cells that increase cytokine production. In some embodiments, cytokines may include IFNγ and TNFα. The method for screening TCRs may also include paired single-cell RNA and V(D)J sequencing of T cells (i.e., TCR sequencing). In some embodiments, the T cells may be selected to increase cytokine production. In some embodiments, the sequencing method includes stimulating T cells by contacting them with a population of APCs loaded with HLA-binding peptides prior to harvesting the T cells for sequencing. Methods, reagents, and kits for single-cell RNA and V(D)J sequencing are known in the art and readily available, and are described, for example, in Hwang, Byungjin, Ji Hyun Lee, and Duhee Bang. "Single-cell RNA sequencing technologies and bioinformatics pipelines." Experimental & molecular medicine 50.8 (2018): 1-14, and Pai, Joy A., and Ansuman T. Satpathy. "High-throughput and single-cell T cell receptor sequencing technologies." Nature methods 18.8 (2021): 881-892, the disclosures of which are incorporated herein by reference.Methods for screening TCRs may also include identifying the TCR sequence of reactive T cells, such as by single-cell RNA sequencing. In some embodiments, reactive T cells may be identified by increased cytokine expression in T cells relative to a control. In some embodiments, reactive T cells may be identified by increased expression of IFNG, TNF, and / or GZMB in T cells relative to a control. In some embodiments, the control may include T cells contacted with APCs loaded with bait peptides (i.e., non-antigenic peptides) prior to sequencing. In some embodiments, the control may include T cells contacted with APCs not loaded with any peptides prior to sequencing. Figure 1 An exemplary implementation of a method for screening, identifying, and incorporating new connective epitopes is illustrated in the diagram.

[0101] Example

[0102] The following embodiments are presented to provide a complete disclosure and illustration of how to make and use the invention to those skilled in the art, and are not intended to limit the scope of the invention as the inventors believe, nor are they intended to represent that the following experiments are all or only experiments performed. Efforts have been made to ensure the accuracy of the figures used (e.g., quantities, temperatures, etc.), but some experimental errors and biases should be taken into account. Unless otherwise stated, parts refer to parts by weight, molecular weight refers to weight-average molecular weight, temperature is in degrees Celsius, and pressure is at or near atmospheric pressure. Standard abbreviations may be used, such as bp, base pair; kb, kilobase; pl, picoliter; s or sec, second; min, minute; h or hr, hour; aa, amino acid; kb, kilobase; bp, base pair; nt, nucleotide; im, intramuscular; ip, intraperitoneal; sc, subcutaneous, etc.

[0103] Example 1

[0104] Despite their promising prospects, T-cell-mediated immunotherapy is largely limited by the extent to which cancer-specific targets (neoantigens) are expressed across the entire tumor landscape. Splicing aberrations (neoconnections) in cancer can represent potential sources of both public (shared among patients) and tumor-wide (expressed in all intratumoral samples) neoantigens, thus prompting the development of novel procedures for identifying these targets across multiple cancers. Analysis characterized several public, tumor-wide neoconnections and identified neoantigen-specific CD8+ T-cell clones targeting splicing mutations derived from GNAS and RPL22, the former being observed to be tumor-wide. TCR-engineered CD8+ T cells targeting these mutations exhibited neoantigen-specific responses and tumor-specific killing. Upregulated neoconnection levels, associated with dysregulated splicing-related gene expression, were observed in the IDH1-mutant glioma subtype compared to its wild-type variant. Notably, knockdown of the chromosomal 1p / 19q co-deletion-related splicing genes resulted in a significant increase in neoconnection expression. Overall, a new set of connections was revealed that were conserved both among patients and within tumors, demonstrating the development potential of novel T-cell-based immunotherapies that address intratumoral heterogeneity.

[0105] Materials and methods

[0106] Data Download: Batch RNA sequencing data for samples of glioblastoma (GBM; n=167), low-grade glioma (LGG; n=516), lung adenocarcinoma (LUAD, n=517), lung squamous cell carcinoma (LUSC, n=501), mesothelioma (MESO, n=516), hepatocellular carcinoma (LIHC, n=371), gastric adenocarcinoma (STAD, n=415), clear cell renal carcinoma (KIRC; n=533), papillary renal carcinoma (KIRP; n=290), chromophobe renal carcinoma (KICH, n=66), colon adenocarcinoma (COAD; n=458), and prostate adenocarcinoma (PRAD; n=497) were downloaded from TCGA in FASTQ format. The download of intratumoral multi-region sampling sequencing data was detailed in the preceding sections. Similarly, batch RNA sequencing data for 9166 normal tissue samples were downloaded in FASTQ format from the Genotype-Tissue Expression Database (GTEx) repository. We received bulk RNA sequencing data from 66 patient-derived GBM cell lines from the Mayo Clinic's National Resource Center for Brain Tumor Xenografts. We also downloaded proteomics data from 100 GBM samples from the Clinical Proteomics Tumor Analysis Collaboration (CPTAC).

[0107] RNA sequencing alignment: All downloaded RNA sequencing datasets were individually aligned using a STAR-based alignment pipeline. A genome index containing unannotated links was constructed using the STAR software (version 2.7.7a) through the initial alignment pass of the input data. Complete set of command line parameters: --runThreadN 1 \ --outFilterMultimapScoreRange1 \ --outFilterMultimapNmax 20 \ --outFilterMismatchNmax 10 \ --alignIntronMax 500000 \ --alignMatesGapMax 1000000 \ --sjdbScore 2 \ --alignSJDBoverhangMin 1 \ --genomeLoad NoSharedMemory \ --limitBAMsortRAM80000000000 \ --readFilesCommand gunzip -c \ --outFilterMatchNminOverLread0.33 \ --outFilterScoreMinOverLread 0.33 \ --sjdbOverhang 100 \ --outSAMstrandField intronMotif \ --outSAMattributes NH HI NM MD AS XS \ --limitSjdbInsertNsj 2000000 \ --outSAMunmapped None \ --outSAMtype BAMSortedByCoordinate \ --outSAMheaderHD @HD VN1.4 \ --twopassMode Basic \ --outSAMmultNmax 1 \, and use the GRCH37 STAR index file for comparison.

[0108] TCGA Sample Selection and Gene Expression Quantification: TCGA tumor samples with an absolute tumor purity greater than 0.60 were retained for downstream computer simulation analysis. (Aran et al., 2015; Ceccarelli et al., 2016) Non-mitochondrial, protein-coding transcripts defined by the Ensembl Homo sapiens GRCH37.87 Gene Annotation Gene Transfer Format (GTF) file were selected, and protein-coding transcript isoforms in TCGA RNA sequencing data were selected and retained using this curated set. Transcript-level expression data (log2[RSEM-TPM+0.001]) for all TCGA samples were downloaded from the UCSCXena Toil pipeline and converted to standard TPM values. Protein-coding transcript isoforms with a median TPM ≥ 10 were retained for downstream analysis. In the case of TCGA gliomas, subsequent expression data in TPM were subsets categorized into six disease types: all cases (n=429), GBM cases (n=115), LGG cases (n=314), IDH1-WT cases (n=166), IDH1-MUT astrocytoma cases (n=140), and IDH1-MUT oligodendroglioma (n=123). Protein-coding transcript isoforms with a median TPM ≥10 in at least one of the six disease types were retained for further analysis.

[0109] Characterization of Public New Connections: For the counting of public cancer-specific splicing events, a custom R script was designed to detect and quantify unannotated cancer-specific splicing events found in each corresponding patient cohort. Variable splicing events were quantified in the corresponding sj.out.tab files within the detected connection counts from the output files derived from the STAR alignment in the previous step. Splicing events detected from the GRCh37.87 GTF sj.out.tab files were removed to define unannotated splicing connections. Unannotated splicing connections overlapping with non-mitochondrial protein-coding genes identified in the previous step were retained for further analysis. All splicing connections with fewer than 10 target splice reads (counts) or fewer than 20 total splice reads (depth) across the entire cohort were removed. Similar to previous studies, splicing frequencies were calculated as the sum of the total number of target splice reads divided by the sum of splice reads from both target and canonical connections. Public splicing connections were defined as connections presumed to be expressed with the above criteria in at least 10% of the study patient cohorts, and these connections were retained for further analysis. To characterize cancer-specific splicing events, also known as new connections, all connections that were presumed to be expressed with the same parameters in more than 1% of normal GTEx samples were removed.

[0110] Detection of cancer-specific intron retention events

[0111] Intron splicing events were detected and characterized using IRFinder v1.2.3. RNA sequencing data from TCGA (GBM / LGG) and GTEx (CNS) were aligned to GRCh37 (hg19) and imported into the software for the detection of intron retention events. Analysis based on a generalized linear model (GLM) was used to assess differential intron retention. The intron retention ratio was calculated as the sum of (intron reads / (intron reads, normal splice reads)). Significant changes in intron retention were defined as (1) not less than 10% in both directions and (2) an adjusted p-value less than 0.05. The PSR of intron retention events in TCGA or GTEx was defined as the number of cases meeting these criteria divided by the total number of cases in the cohort. Possible cancer-specific intron retention neoconnections were characterized as intron retention events with a TCGA PSR ≥ 0.10 and a GTEx PSR < 0.01.

[0112] Transcriptome validation of newly expressed connections

[0113] Detection of novel connections expressed in patient-derived GBM / LGG cell lines RNA sequencing data derived from the GBM PDX cell line were downloaded from the Mayo Clinic Brain Tumor Patient-Derived Xenograft National Resource Center. Patient-derived LGG cell lines were generated from surgically resected specimens from the Neurosurgery Brain Tumor Center at the University of California, San Francisco. The RNA sequencing data from the GBM and LGG cell lines were compared and processed as described above. Common neo-sponge connections with a splice join count >0 per million (CPM) were considered detectable in the cell line-derived RNA sequencing data. Detection of novel connections expressed in multi-regional cases: In a spatially mapped glioma case cohort, approximately ten or more anatomical biopsy samples at maximum distances were collected from each patient, allowing for assessment of intratumoral genetic heterogeneity via RNA sequencing, whole-exome sequencing, etc. RNA sequencing data collected from each multi-region sample were processed and aligned as described above. Presumed novel connections previously characterized in TCGA were searched. Common novel connections with CPM > 0 were considered detectable. Common novel connections with presumed expression (≥10 splice reads) in two or more mapped samples within the same case were considered spatially conserved novel connections. Novel connections detected in all multi-region samples within the same tumor were considered tumor-wide novel connections.

[0114] Proteomic validation of expressed novel linker-derived peptides

[0115] From the putative neolinks detected in the above-described process, a database of all plausible peptides derived from all neolinks was generated. To detect neolink-derived peptides in GBM cases, .RAW files of GBM and LGG MS data stored in the Clinical Proteomics Tumor Analysis Consortium (CPTAC, n=99), Bader et al. (n=99), Lam et al. (n=92), and Yanovich-Arad et al. (n=84) were analyzed. MaxQuant (v1.6.17.0) was used to identify trypsin sequences from the corresponding CPTAC MS datasets. Predicted neolink-derived peptides, decoy sequences, and the human reference proteome (Uniprot proteome ID#UP000005640) were input as FASTA files into MaxQuant, and trypsin sequences from the input files were matched against the CPTAC database. Cancer-specific peptides spanning neolink-derived protein sequences were considered CPTAC-confirmed. The relative detection levels of neolink-derived peptides and normal tissue-derived peptides were assessed using their log2 (peak intensity). In addition to the default settings, the following commands and parameters used in MaxQuant for MS analysis were modified and used: Digestion mode = trypsin / P; Maximum omission = 3; Minimum peptide length = 5; Minimum peptide length for nonspecific search = 5.

[0116] Peptide treatment and prediction of MHC-I binding and presentation

[0117] Cancer-specific transcripts with relevant novel linkages were computer-simulated translated into their corresponding amino acid sequences. Libraries of all possible multimers ranging from 8 to 11 amino acids in length were then generated, and cancer-specific sequences were selected by removing those detectable in normal tissue peptide isoforms in a reference human proteome dataset (Uniprot proteome ID#UP000005640). All cancer-specific multimers and their upstream and downstream flanking sequences (maximum flanking length 30 amino acids) were independently analyzed and ranked using MHCFlurry 2.0 and HLAthena MSiC. In both cases, HLA-A was targeted. 01:01, HLA-A 02:01, HLA-A 03:01, HLA-A 11:01 and HLA-A 24:02 HLA:MHC-I binding affinity was assessed. In the evaluation of antigen binding and presentation for corresponding HLA haplotypes in HLAthena, alleles were assigned by rank with a threshold of 0.1. Peptide aggregation analysis was performed using up to 30 flanking amino acids from each of the N- and C-termini as background, without logarithmic transformation of expression. The baseline MHCFlurry 2.0 model was used, including a peptide:MHC-I binding affinity (BA) predictor and an antigen processing (AP) predictor. Overall, multimer:HLA presentation scores were characterized in both MHCFlurry 2.0 and HLAthena using the mhcflurry_presentation_score and the MSiC_HLA score, respectively. To select high-binding complexes, a list of the top 10 percentile multimer:HLA complexes from both prediction algorithms was compiled.

[0118] Cell culture

[0119] GBM PDX cell culture: Patient-derived xenograft (PDX) glioblastoma cell lines, GBM34, GBM43, GBM108, GBM115, GBM118, GBM102, GBM137, GBM148, GBM164, and GBM195, were obtained from the Mayo Clinic's National Resource Center for Brain Tumor Xenografts. The xenograft lines were cultured according to recommended conditions in previous literature and passaged a maximum of 20 times before reverting to earlier passages. Cells were cultured in Dulbecco Modified Eagle Medium (DMEM): F12 supplemented with 10% fetal bovine serum and 1% penicillin and streptomycin (P / S). Prior to use, the cells were cultured at 4°C with DPBS (containing calcium and magnesium) and 10% laminin (Gibco... TM Overnight cell culture plates (Cat. #23017015) GBM / LGG from primary patients Cell culture:Primary patient-derived wild-type IDH1 GBM (SF7996), mutant IDH1 astrocytoma (SF10602), and mutant IDH1 oligodendroglioma (SF10417) cell lines were previously generated from isolated glioma biopsies and cultured as described above. Cells were cultured in serum-free glioma neural stem cell (GNS) culture medium containing Neurocult NS-A (STEMCELL Technologies Cat. #05751), supplemented with N-2 supplement (Invitrogen Cat. #17502048), B-27 supplement (vitamin A-free) (Invitrogen Cat. #12587010), 1% P / S, 1% glutamine, and 1% sodium pyrophosphate. Before immediate use in culture, supplement GNS medium with 20 ng / mL EGF (Peprotech Cat. #AF-100-15), bFGF (Peprotech Cat. #AF-100-18B), and PDGF-AA (Peprotech Cat. #AF-100-13A). Similar to the GBM PDX cell line, incubate overnight at 4°C with DPBS (containing calcium and magnesium) and 10% laminin (Gibco) before use. TM Cat. #23017015) cell culture plate. Jurkat76 cell culture: Jurkat76 cells were used as TCR α- and β-negative human T cell derivatives that allow for the non-competitive introduction of exogenous TCRs. CD8+ Jurkat76 cells were cultured in RPMI supplemented with 10% fetal bovine serum and 1% P / S. T2 cell culture: This study used T2 cells to monitor the response of immune cells to exogenous antigens of interest in a non-competitive environment. T2 cells lack peptide transporters (TAPs) involved in antigen processing; therefore, these cells were induced with exogenously administered peptides to enable them to interact with class I MHC molecules (especially HLA-A). 0102) Association and presentation. T2 cells were cultured in IMDM medium supplemented with 20% FBS. COS7 and K562 cells nourish: The COS7 (ATCC Cat. #CRL-1651) and K562 (ATCC Cat. #CCL-243) cell lines were used as primate and artificial antigen-presenting cell (aAPC) models, respectively. These cell lines do not express HLA molecules, which allows for the introduction of HLA alleles of interest. COS7 cells were cultured in DMEM medium supplemented with 10% FBS and 1% P / S. K562 cells were cultured in IMDM medium supplemented with 10% FBS and 1% P / S. THP-1 cell culture:Immunoreactivity against neoantigens presented by dendritic cells was studied using THP-1 cells (ATCC Cat.#TIB-202). THP-1 cells were cultured in RPMI-1640 supplemented with 10% FBS.

[0120] Sensitization of healthy PBMCs from in vitro donors

[0121] Purchase HLA-A from StemExpress 02:01:01 Positive PBMCs, in fresh or frozen form. Immediately administer approximately 1×10 9 Fresh PBMCs (StemExpress Cat. #LE001F) were allocated proportionally to 3 × 10⁶ cells. 8 Aliquots of cells were aliquoted and cryopreserved in liquid nitrogen, with one aliquot used in downstream IVS. Approximately 3 × 10⁻⁶ cells were aliquoted from each vial. 8 One vial of cryopreserved PBMCs (StemExpress Cat. #PBMNC300C) per cell was used for each IVS procedure. The PBMCs were thawed with 1:1000 Benzonase:RPMI (Sigma Aldrich Cat. #E8263). Following the manufacturer's instructions, CD14+ populations were isolated from the PBMCs using CD14+ Miltenyi microbeads (Miltenyi Biotec Cat. #130-050-201). The CD14- flow-through was cryopreserved for 6 days and then used for naive CD8+ T cell isolation. The isolated CD14+ cells were cultured at 5 × 10⁶ cells per well. 5The cells were seeded at a density of 1000 U / mL in untreated 24-well plates and cultured for 3 days in CellGenix GMP DC medium (CellGenix Cat#20801-0500) supplemented with 1% human serum (Sigma Aldrich Cat#H6914) and 1% P / S. On day 3, the cultured CD14+ population was stimulated with 1000 U / mL recombinant human IL-4 (Peprotech Cat. #200-04) and GM-CSF (Peprotech Cat. #300-03). On day 5, the monocytes in the CD14+ population were stimulated with 1000 U / mL recombinant human IL-4 and GM-CSF and matured into dendritic cells (DCs) by introducing 250 ng / mL LPS (Sigma Aldrich Cat. #L6529). On day 6, naive CD8+ T cells were isolated from the thawed CD14- population using the EasySep Human Naive CD8+ T Cell Isolation Kit (STEMCELL Technologies Cat. #19258) according to the manufacturer's instructions. The isolated naive CD8+ T cells were cultured at 5 × 10⁶ cells per well. 5 Cells were seeded at a density of 1000 cells / well in 48-well plates in X-Vivo 15 medium (Lonza Cat. #04-418Q) supplemented with 5% human serum, 1% P / S, and 10 ng / mL recombinant human IL-7 (Peprotech Cat. #200-07). On day 8, adherent mature DCs were harvested from the plates with cold PBS. The collected DCs (1 × 10⁶ cells / well) were diluted with 1 μM of neoantigen peptide, influenza peptide, or no peptide. 6 DCs (cells / mL) were exogenously loaded at 37°C for 1 hour. Then, either peptide-loaded or unloaded DCs were co-cultured with naive CD8+ T cells in 48-well plates, with an optimal DC:T cell ratio of 1:4. Co-culture was maintained for 10 days in X-Vivo 15 medium supplemented with 10 ng / mL recombinant human IL-7, 10 ng / mL recombinant human IL-15 (Peprotech Cat. #200-15), and 60 ng / mL recombinant human IL-21 (Peprotech Cat. #200-21), with restimulation with IL-7 and IL-15 every 2 days. Cells were reseeded into subsequent 24-well, 12-well, and 6-well plates based on confluence. This completed the first round of IVS with the neoantigen and influenza peptide. On days 19 and 29, sensitized CD8+ T cells were reintroduced for the second and third rounds of stimulation with freshly loaded DCs, and co-culture was maintained for another 10 days until the second and third rounds of IVS were completed. Immunogenic cytokine assays were performed at the end of the second and third rounds of IVS to determine the presence of an expanded population of peptide-responsive T cells.

[0122] Mutation-specific ELISA screening

[0123] Aliquots containing CD8+ T cells from each parent IVS well were harvested and evenly divided into 96-well progeny plates, with each well containing 1×10⁶ cells. 5 T2 cells were used to stimulate triplicate progeny wells for 16 hours at an effector cell to target cell (E:T) ratio with T2 cells loaded with the neoantigen peptide of interest, bait peptide, no peptide, or no T2 cells at all. T2 cells were loaded with 1 pM to 1 μM of the neoantigen peptide of interest, bait peptide, or no peptide at 37°C for 1 hour. Influenza-responsive T cells were co-cultured with influenza peptide-loaded T2 cells as a positive control. The co-culture supernatant was collected and diluted according to the manufacturer's instructions for use in IFNγ (BD Biosciences Cat. #555142) and TNFα (BD Biosciences Cat. #555212) ELISA. ELISA readouts were performed on an Epoch microplate spectrophotometer (BioTek Instruments) using BioTek Gen5 data analysis software (version 1.11). Wells showing significant increases in IFNγ and TNFα expression levels were selected for downstream single-cell immunoassays using single-cell RNA and V(D)J sequencing.

[0124] Single-cell immunoassay profile

[0125] Once the expanded neoantigen-reactive CD8+ T cell population from IVS was identified, single-cell RNA and V(D)J sequencing were performed using the 10x Genomics platform. Prior to sequencing, CD8+ T cells from the expanded neoantigen-reactive (ELISA-selected positive) wells were harvested and co-cultured with 1 μM of the neoantigen peptide of interest, bait peptide, or peptide-free T2 cells at a 1:1 E:T ratio. One co-culture was repeated for 3 hours for single-cell RNA sequencing analysis, and another for 16 hours for IFNγ and TNFα ELISA confirmation. The final cell concentration was adjusted to approximately 1 × 10⁻⁶ cells / well. 4Cells / μL, with initial cell viability of at least 90% to maximize the likelihood of achieving the desired cell recovery target. Sequencing of independent CD8+ T cells and unloaded T2 single cultures with co-culture conditions was performed to differentiate cell types in downstream single-cell sequencing analysis. Preparation for single-cell sequencing analysis was performed using the ChromiumNext GEM Single Cell 5' Kit v2 (Dual Indexing) (10xGenomics, Cat. #CG000331). Gel beads in an emulsion (GEM) were generated by combining single-cell 5' gel beads, separator oil, and a master mixture containing cells onto a Chromium Next GEM chip K. Cell lysis and barcoded reverse transcription of RNA were performed within the corresponding GEM for all single cells. Barcoded cDNA products were recovered via GEM-RT post-cleaning and PCR amplification. cDNA quality control and quantification were performed on a Fragment Analyzer system (Agilent Technologies). 50 ng of cDNA was used to construct 5' gene expression libraries, and each sample was indexed using the Chromium i7 Sample Indexing Kit. The procedure was performed on an Illumina NovaSeq 6000 sequencer at the Institute for Human Genetics (IHG) at UCSF, with at least 20,000 read pairs for the 5' gene expression library per cell. Enrichment products were measured using the Fragment Analyzer system. 50 ng of the enriched TCR product was used for library construction. Single-cell V(D)J enriched libraries were then sequenced on the Illumina NovaSeq 6000, with at least 5,000 read pairs for the V(D)J library per cell. Cell Ranger 7.0.0 (10x Genomics Cloud Analysis) was used to preprocess raw single-cell RNA sequencing and identify V(D)J clonotypes. Annotation files 'vdj_GRCh38_alts_ensembl-3.1.0-3.1.0' and 'GRCh38-3.0.0' were used to demultiplex cell barcodes, perform read alignment, and generate a feature-barcode matrix. Only cells with usable clonoid information were retained for downstream analysis. Single-cell gene expression and corresponding V(D)J sequences of candidate T cell clonoids were analyzed on the Loupe V(D)J browser. T cells with detectable CD8A expression were specifically isolated as a CD8+ cell population and subsequently grouped according to their clonoids. To identify T cell clonoids associated with neoantigen-specific responses, expanded TCR clonoids exhibiting significantly increased levels of IFNG, TNF, and GZMB expression under T2 conditions of T cells:neoantigen loading compared to T2 conditions of T cells:decoy loading and T2 conditions of T cells:unloaded were selected.

[0126] HLA typing

[0127] OptiType 1.3.1 is used to perform HLA allelic typing of available glioma cell lines from available WES data using default parameters.

[0128] plasmids and peptides

[0129] HLA-A Both the 02:01 and novel ligation-derived gene sequences were synthesized and cloned into the pTwist Lenti SFFVPuro WPRE vector (Twist Biosciences). Full-length and truncated multimeric versions encoding wild-type and mutant GNAS and RPL22 sequences were generated. TCRα / β was synthesized and cloned into the pTwist Lenti SFFV vector (TwistBiosciences). HPLC-grade novel ligation-derived neoantigen peptide multimers (>95%) were manufactured by TC Laboratories.

[0130] Lentiviral transduction

[0131] HEK293T cells were used at a rate of 1 × 10⁻⁶ cells per well. 6 Cells were plated at a density in 6-well plates containing 2 mL of DMEM medium supplemented with 10% FBS without antibiotics. Approximately 18 to 24 hours later, or at 90% confluence, HEK293T cells were transfected with the expression constructs (see above), as well as the lentiviral packaging plasmids pMD2.G (Addgene, #12259) and psPAX2 (Addgene, #12260). TCRα / β transduction:1.0 μg TCRα / β transfer plasmid, 0.75 μg psPAX2, and 0.25 μg pMD2.G were combined with 200 μL Opti-MEM (Thermo Fischer Scientific Cat. #31985062). 6 μL Xtremegene HP was added to this mixture, and complex formation was allowed to occur at room temperature for 15 minutes. The reaction mixture was then added to the corresponding HEK293T cells. The transfection medium was replaced with fresh DMEM after 24 hours. Viral supernatant was collected after 48 hours, and functional viral titers were performed on 6-well plates inoculated with Jurkat76 cells or CD8+ T cells at 60–70% confluence. Viral transduction was performed by supplementing the viral stock solution with a final concentration of 4 μg / mL polybrene using a 3-fold serially diluted viral stock solution. The medium was replaced 24 hours after viral transduction. Transduction efficiency was assessed by measuring the surface expression of TCRα / β and CD3 in cells using fluorescence-activated cell sorting (FACS) analysis after 3–4 days. Cells exhibiting high double-positive expression of TCRα / β and CD3 were flow-sorted and maintained in downstream co-culture and immunogenicity assays. HLA and neoantigen transduction: HLA-A expression The 02:01 construct was linearized and digested with BamHI and XhoI (New England Biolabs) restriction enzymes, and purified using the Zymoclean Gel DNA Recovery Kit (Zymo Research Cat. #D4007). HLA-A was then... The 0201 sequence was linked downstream of the EF1A-core promoter and upstream of IRES in the lentiviral construct, followed by the blasticidin resistance gene. 1.0 μg of HLA-A 02:01 or a combination of neoantigen transfer plasmid, 0.75 μg psPAX2, and 0.25 μg pMD2.G with 200 μL Opti-MEM (Thermo Fischer Scientific Cat. #31985062). Add 6 μL Xtremegene HP to this mixture and allow complex formation to occur at room temperature for 15 minutes, at which point add the reaction mixture to the corresponding HEK293T cells. As described above, the neoantigen construct encodes a full-length or truncated multimeric version of the newly linked peptide. Replace the transfection medium with fresh DMEM after 24 hours. HLA-A testing is performed first before neoantigen lentiviral transduction and selection. 02:01 Lentiviral transduction and screening to simplify drug selection. Viral supernatant was collected 48 hours later, and functional viral titers were performed on 6-well plates inoculated with COS7 or K562 cells at 60-70% confluence. Viral transduction was performed using a 3-fold serially diluted viral stock solution supplemented with 4 μg / mL polybrene. The medium was replaced 24 hours after viral transduction with complete medium supplemented with blasticidin. Transduction efficiency was assessed 3-4 days later using drug screening. HLA-A was tested before assessing cell viability at each titer. The APCs transduced at 02:01 were cultured for approximately 7 days in medium treated with 10 μg / mL Blasticidin. Neoantigen-lentivirus transduction was then performed, followed by HLA-A... APCs transduced with the 02:01 and neoantigen expression constructs were cultured for approximately 7 days in medium treated with 3 μg / mL puromycin. Cell viability was then assessed under all titer conditions. After 3–4 days, transduction efficiency was assessed by measuring fluorescence-activated cell sorting (FACS) of HLA-A2 surface expression.

[0132] Dose-dependent assessment of TCR responsiveness to neoantigens

[0133] The specificity of neoantigen-reactive CD8+ T cells and TCR-transduced T cells was assessed using human IFNγ (BD Biosciences Cat. #555142), IL-2 (BD Biosciences Cat. #555190), and TNFα ELISA (BD Biosciences Cat. #555212). TCR recognition of exogenously introduced neoantigen peptides presented by MHC-I molecules was assessed by co-culturing T cells with peptide-loaded T2 cells. The neoantigen peptide of interest, decoy peptide, or peptide-free T2 cells were cultured at 37°C for 1 hour at concentrations between 1 pM and 1 μM. Influenza-reactive T cells were co-cultured with influenza peptide-loaded T2 cells as a positive control. Cells were cultured at 1 x 10-1 cells per cell type. 5T cells and T2 cells were co-cultured in 200 μL of medium in 96-well round-bottom plates for 16 hours. The supernatant was collected and diluted for cytokine assays. The supernatant was collected and diluted for cytokine release assays according to the manufacturer's instructions. ELISA assays were performed using BioTek Gen5 data analysis software on an Epoch microplate spectrophotometer. To characterize dose-dependent activation of the TCR in transduced triple reporter Jurkat76 cells, flow cytometry was performed to assess the expression levels of NFAT-GFP, NFκB-CFP, and AP-1-mCherry after 16 hours of co-culture. Similarly, the responsiveness of TCR-transduced CD8+ T cells was evaluated by flow cytometry after staining with anti-CD107a (BioLegend, Cat#328620) and anti-CD137 antibody (4-1BB; Biolegend Cat#309804).

[0134] Synthesis of in vitro transcribed (IVT) mRNA

[0135] All constructs were subcloned into pcDNA3.1 (Invitrogen, 2520855) and linearized using the XhoI restriction enzyme, in which plasmid DNA templates were transcribed downstream of the phage T7 promoter sequence. 1 μg and 0.5 μg of template were used for long (>0.5 kb) and short (<0.5 kb) transcripts, respectively. The reaction was assembled at room temperature using the mMESSAGE mMACHINE T7 transcription kit (Invitrogen, 2582905) according to the manufacturer's instructions, and incubated at 37°C for 1 h for long transcripts and 16 h for short transcripts. Following DNase treatment, poly(A) tailing was performed for 1 h according to the HiScribe T7 ARCA manual (NEB, E2060S). Subsequently, the synthesized mRNA was purified by LiCl precipitation using 70% DEPC-based ethanol. The synthesized mRNA was heat-shocked (70°C, 5 min) with formaldehyde loading dye to verify quality by gel electrophoresis.

[0136] HLA-A 02:01. Transfection of truncated neoantigens and full-length neoligated mRNAs encoding mRNAs

[0137] According to the manufacturer's instructions, IVT-synthesized mRNA was transfected into COS7 and K562 cells via electroporation using a 100 μL Neon transfection system kit (Invitrogen, MPK10096). 100 μL of Neon... TM Washing and resuspending with resuspension buffer 1x10 6COS7 and K562 cells. 5 μg of HLA-A2 and 5 μg of candidate mRNA (truncated neoantigen sequence or full-length neolinked sequence) were added to the cell solution. Electroporation was performed on a Neon NxT electroporation system (Invitrogen, NEON1). Electroporation of COS7 cells was performed using the following optimized conditions: pulse voltage 1200 V, width 30 ms, and 2 pulses. Electroporation of K562 cells was performed using the following optimized conditions: pulse voltage 1450 V, width 10 ms, and 3 pulses. Transfected cells were immediately transferred to warm, antibiotic-free RPMI. Aliquots of transfected cells were retained for HLA-A2 expression validation by staining with an HLA-A2 monoclonal antibody (BB7.2, Thermo Scientific, 17-9876-42) and subsequent flow cytometry analysis.

[0138] Evaluation of TCR specificity for neoantigens processed endogenously and presented by MHC-I

[0139] By HLA-A 02:01 / Neoantigen-transfected COS7 or K562 cells were co-cultured with TCR-transduced T cells for endogenous processing and characterization of neoantigens presented by surface MHC-I. Similarly, 1x10 cells were used for each cell type. 5 T cells and COS7 / K562 cells were co-cultured in 200 μL of medium in 96-well plates for 16 hours. Supernatants were collected and diluted according to the manufacturer's instructions for cytokine release assays, and cytokine release levels were assessed using an Epoch microplate spectrophotometer and BioTek Gen5 data analysis software. In all cytokine release assays, the maximum cytokine release per well was determined by adding 0.2 μL of cell activation mixture (without brevidin A) (BioLegendCat. #423302) to each 100 μL cell solution. Cytokine release was determined by an E:T ratio of 1:1 (1 x 10⁶ cells per well in a 96-well plate). 5 TCR-transduced triple reporter Jurkat76 cells were co-cultured with glioma cells to evaluate the endogenous processing and presentation of neoantigens in glioma cell lines. Flow cytometry analysis was performed to assess the expression levels of NFAT-GFP, NFκB-CFP, and AP-1-mCherry after 16 hours of co-culture.

[0140] HLA-IP and LC-MS / MS

[0141] 10 μg of HLA-A encoded DNA was transfected using the Neon transfection system (100 μL pipette tip, settings: 1,050 V / 10 ms / 2 pulses). 02:01 Each mRNA containing the allele and the full-length coding sequence of the mutated GNAS or RPL22 was co-electroplated into COS-7 cells. Electroporation was performed at 20 × 10⁻⁶ cells per condition. 6Cells were plated overnight in six-well non-TC plates. Cells were harvested by incubation at 37°C with 1 mM EDTA (Millipore Sigma) for 10 min. For immunoprecipitation, cells were lysed at 4°C in 8 mL of 1% CHAPS (Millipore Sigma) for 1 h, followed by centrifugation of the lysate at 20,000 g for 1 h at 4°C, and collection of the supernatant. For affinity-based immunopurification of HLA-I ligands, 40 mg of cyanogen bromide-activated Sepharose 4B (Millipore Sigma) was activated with 1 mM hydrochloric acid (Millipore Sigma) for 30 min. Subsequently, 1 mg of W6 / 32 antibody (Bio X Cell) was conjugated to Sepharose in the presence of binding buffer (150 mM sodium chloride, 50 mM sodium bicarbonate, pH 8.3; sodium chloride) at room temperature for 2 h. Sepharose was blocked with glycine for 1 h and washed three times with PBS. The supernatant of cell lysates was run on an affinity column overnight at 4°C using a peristaltic pump at a flow rate of 6 mL / min. The HLA complex and binding peptide were eluted five times from the column using 1% TFA. The peptide and HLA-I complex were separated using a C18 column (Sep-Pak C18 1 cc Vac Cartridge, 50 mg adsorbent per column, 37–55 μm particle size, Waters). The C18 column was pretreated with 80% ACN (Millipore Sigma) in 0.1% TFA and equilibrated with two washes in 0.1% TFA. The sample was loaded, washed twice with 0.1% TFA, and eluted in 30%, 40%, and 50% acetonitrile in 0.1% TFA, 300 μL each time. All three fractions were combined, dried by vacuum centrifugation, and stored at -80°C until further processing. The HLA-I ligand was separated by solid-phase extraction using a homemade C18 microcolumn. Samples were analyzed using high-resolution / high-precision LC-MS / MS (Lumos Fusion, ThermoFisher Scientific). The operating resolutions for MS and MS / MS were 60,000 and 30,000, respectively. Only charge states 1, 2, and 3 were allowed. A 1.6 Thomson spectral density was selected as the isolation window, and the collision energy was set to 30%. For MS / MS, the maximum injection time was 100 ms, and the automatic gain control was 50,000. MS data were processed using FragPipe. The protein FDR was set to 0.01. For all samples, methionine oxidation, phosphorylation of serine, threonine, and tyrosine, and N-terminal acetylation were set as variable modifications. Review proteins including UniProt Cercopithecus aethiops (supplemented with human HLA-A) were analyzed. The samples were searched using databases containing allele sequences (02:01), mutRPL22 and mutGNAS, and common pollutants.

[0142] Characteristics of CD8+ T cell-mediated antitumor response

[0143] To determine whether TCR-transduced T cells could produce an antitumor response, TCR-transduced J76 or CD8+ T cells were co-cultured with patient-derived GBM or LGG cell lines. CD8+ T cells were isolated from healthy donor-derived PBMCs using the EasySep™ Human CD8+ T Cell Isolation Kit (STEMCELL Technologies, Cat. #17953). Then, cells were isolated in 25 μL / 1 × 10⁻⁶ cells. 6 The concentration of each cell was used to activate CD8+ T cells with Dynabeads™ Human T-Activator CD3 / CD28 (Thermo Scientific, Cat. #11161D) for T cell expansion and activation. Supplementation with IL-7 (30 μL / 1 × 10⁻⁶ cells) was also employed. 6 CD8+ T cells were cultured in a medium containing 100 cells for 7 days, replenished every 2 days. The CD8+ T cells were then transduced using the neoantigen-specific TCR and a heterozygous mouse TCR constant region via the transduction procedure described above. This additional step eliminated the possibility of TCR α- and β-chain mismatches and allowed us to assess TCR transduction efficiency by staining with an anti-mouse TCR constant region antibody (clone H57-597; BioLegendCat. #109208). Highly transduced CD8+ T cells were isolated by flow cytometry sorting using cells strongly stained with anti-CD3 and anti-mouse TCR constant region antibodies. The sorted transduced CD8+ T cells were expanded for 7 days prior to use in co-culture assays. Kill assays were performed using an xCELLigence RTCA S16 real-time cell analyzer. Tumor cells were cultured for 48 hours in a medium pretreated with 100 ng / mL IFNγ (Peprotech, Cat. #300-02) and washed twice with PBS before inoculation. 1×10 plates were laid in each well of a 96-well E-plate (Agilent). 4 Tumor cells were incubated, and impedance was read continuously for 16 hours during incubation. TCR-transduced CD8+ T cells were introduced into each well at an E:T ratio of 1:1 or 2:1, and tumor-specific killing was measured by changes in cell index over 24–48 hours.

[0144] Identification of MHC-I-restricted CD8+ T cell-mediated reactivity to neoantigens

[0145] MHC-I-restricted reactivity was evaluated by introducing anti-MHC-I antibodies to disrupt TCR and HLA:peptide interactions. In the dose-dependent immunogenicity assay, the concentration in each well of a 96-well plate was 1 × 10⁻⁶. 5 T2 cells containing tumor cells were washed twice with PBS and incubated for 30 min with either a blocking anti-MHC-I antibody (50 μg / well; clone W6 / 32, Bio X Cell, Cat. #BE0079) or an allotype control (50 μg / well; Bio X Cell, Cat. #BE0085) in a total volume of 100 μL. Without further washing, T cells were added to reach a final volume of 200 μL. For the tumor killing assay, tumor cells were added to each well of a 96-well E-plate for initial seeding in a total volume of 50 μL. Thirty minutes before adding T cells, either the anti-MHC-I antibody or the allotype control (50 μg / well) was added to each well to reach a total volume of 100 μL. T cells were added to each well to reach a final volume of 200 μL, and impedance was measured over the next 24–48 hours.

[0146] FACS analysis and antibodies

[0147] TCR-transduced cell lines were stained with anti-human TCRα / β (clone IP26, BioLegend Cat. #306717) and anti-human CD3 antibody (clone HIT3a, BioLegend Cat. #300307) to assess the surface expression level of transduced TCRs. CD8+ T cells were stained with anti-CD107a (BioLegend, Cat #328620) and anti-CD137 antibody (4-1BB; Biolegend Cat #309804) to assess CD8+ T cell degranulation and TCR activation, respectively. Zombie Green antibody was used. TM Cell viability was assessed using a fixation viability kit (BioLegend, Cat. #423111). APCs and patient-derived glioma cell lines were stained with HLA-A2 monoclonal antibody (clone BB7.2, Thermo Fisher Scientific Cat. #17-9876-42). Approximately 1 × 10⁻¹⁰ saturates per 100 μL of FACS buffer (PBS supplemented with 1% BSA (Sigma Aldrich Cat. #L6529)) as directed by the manufacturer. 6 One cell was incubated with one test volume of antibody for 20 minutes. The stained cells were washed once with FACS buffer and then resuspended to 4 × 10⁶ cells / 100 μL FACS buffer. 5The concentration of cells was then determined. The cells were then analyzed using an Attune NxT flow cytometer (Thermo Fischer Scientific).

[0148] Gene set enrichment analysis

[0149] Differential gene expression was performed and quantified using DESeq2, based on TCGA, GTEx, and UCSF GBM / LGG RNA sequencing. Only genes with an absolute fold change >2 as called by DESeq2 and a Benjamini-Hochberg adjusted p-value <0.05 were considered differentially expressed. Pre-ranked gene set enrichment analysis (GSEA) was performed by ranking genes by the product of their fold change sign and -log10 (adjusted p-value). Disease subtype-specific differential gene analysis: GSEA comparisons were performed between IDH1 mutant subtypes (wild-type IDH1 and mutant IDH1) and glioma disease subtypes (wild-type IDH1 glioblastoma, mutant IDH1 astrocytoma, and mutant IDH1 oligodendroglioma). Gene sets related to splicing were selected based on keyword searches, and those with adjusted p-values ​​<0.05 when comparing two groups were considered differentially enriched. Unbiased hierarchical clustering of differentially enriched gene sets allowed for characterization of subgroup-specific upregulated genes. Genes upregulated in mutant IDH1 were selected by demonstrating a 1.5-fold (p<0.05) log2 increase in expression of splicing-related genes compared to their wild-type counterparts in mutant IDH1 cases. The selection of splicing genes affected by oligodendroglioma-specific chromosome 1p / 19q loss was determined by a log2-fold decrease in expression of chromosome 1p / 19q-related genes in IDH-O cases compared to both IDH-A and IDH-wt cases (p<0.05). To determine the correlation factors between each of the identified common neojoints and each splicing gene of interest, Pearson correlation analysis was performed on each neojoint and splicing-related gene pair. Newly linked load-specific differential gene analysis: TCGA LGG and GBM samples were ranked based on the putative total number of new connections expressed per sample. Within each disease subtype, high (NJ) levels were ranked. HI ) and low new connection load (NJ LO The samples were characterized as the upper and lower 0.10 percentiles of the ranked samples, respectively. In each disease subgroup, NJ... HI and NJ LO GSEA was performed between samples. Gene sets with unidirectional fold changes and adjusted p-values ​​<0.05 were considered enriched gene sets associated with neoconnection load. Gene sets associated with splicing were selected based on keyword searches. Leading marginal genes shared across all disease subgroups within the same gene set were defined as enriched genes associated with neoconnection load.

[0150] 1. Vaubel, R.A., Tian, S., Remonde, D., Schroeder, M.A., Mladek, A.C., Kitange, G.J., Caron, A., Kollmeyer, T.M., Grove, R., Peng, S., et al. (2020). Genomic and Phenotypic Characterization of a Broad Panel of Patient-Derived Xenografts Reflects the Diversity of Glioblastoma. Clin. Cancer Res. 26, 1094–1104.

[0151] 2. Edwards, N.J., Oberti, M., Thangudu, R.R., Cai, S., McGarvey, P.B., Jacob, S., Madhavan, S., and Ketchum, K.A. (2015). The CPTAC Data Portal: A Resource for Cancer Proteomics Research. J. Proteome Res. 14, 2707–2713.

[0152] 3. Kahles, A., Lehmann, K.-V., Toussaint, N.C., Hüser, M., Stark, S.G., Sachsenberg, T., Stegle, O., Kohlbacher, O., Sander, C., Cancer Genome Atlas Research Network, et al. (2018). Comprehensive Analysis of Alternative Splicing Across Tumors from 8,705 Patients. Cancer Cell 34, 211-224.e6.

[0153] 4. Jones, LE, Hilz, S., Grimmer, MR, Mazor, T., Najac, C.,Mukherjee, J., McKinney, A., Chow, T., Pieper, RO, Ronen, SM. Patient-derived cells from recurrent tumors that model the evolutionof IDH-mutant glioma. Neuroncol Adv 2, vdaa088.

[0154] 5. Stathias , V. , Jermakowicz , AM , Maloof , ME , Forlin , M. ,Walters , W. , Suter , RK , Durante , MA , Williams , SL , Harbor , JW ,Volmar , C.-H. (2018). Drug and disease signature integrationidentifies synergistic combinations in glioblastoma. Nat. Commun. 9 , 5315 .

[0155] 6. Jermakowicz, AM, Rybin, MJ, Suter, RK, Sarkaria, JN,Zeier, Z., Feng, Y., and Ayad, NG (2021). The novel BET inhibitor UM-002reduces glioblastoma cell proliferation and invasion. Sci. Rep. 11 , 23370 .

[0156] 7. Zhang, Y., Liu, Y., Moxley, K.M., Golden-Mason, L., Hughes, M.G., Liu, T., Heemskerk, M.H.M., Rosen, H.R., and Nishimura, M.I. (2010). Transduction of human T cells with a novel T-cell receptor confers anti-HCV reactivity. PLoS Pathog. 6, e1001018.

[0157] 8. Bossi, G., Gerry, A.B., Paston, S.J., Sutton, D.H., Hassan, N.J., and Jakobsen, B.K. (2013). Examining the presentation of tumor-associated antigens on peptide-pulsed T2 cells. Oncoimmunology 2, e26840.

[0158] 9. Hosken, N.A., and Bevan, M.J. (1990). Defective presentation of endogenous antigen by a cell line expressing class I molecules. Science 248, 367–370.

[0159] 10. Chandran, S.S., Ma, J., Klatt, M.G., Dündar, F., Bandlamudi, C., Razavi, P., Wen, H.Y., Weigelt, B., Zumbo, P., Fu, S.N., et al. (2022). Immunogenicity and therapeutic targeting of a public neoantigen derived from mutated PIK3CA. Nat. Med. 28, 946–957.

[0160] 11. Britten, C.M., Meyer, R.G., Kreer, T., Drexler, I., Wölfel, T.,and Herr, W. (2002). The use of HLA-A 0201-transfected K562 as standardantigen-presenting cells for CD8(+) T lymphocytes in IFN-gamma ELISPOTassays. J. Immunol. Methods 259, 95–110.

[0161] 12. Suhoski, M.M., Golovina, T.N., Aqui, N.A., Tai, V.C., Varela-Rohena, A., Milone, M.C., Carroll, R.G., Riley, J.L., and June, C.H. (2007).Engineering artificial antigen-presenting cells to express a diverse array ofco-stimulatory molecules. Mol. Ther. 15, 981–988.

[0162] 13. Love, M.I., Huber, W., and Anders, S. (2014). Moderatedestimation of fold change and dispersion for RNA-seq data with DESeq2. GenomeBiol. 15, 550.

[0163] 14. Matsushima, S., Ajiro, M., Iida, K., Chamoto, K., Honjo, T., andHagiwara, M. (2022). Chemical induction of splice-neoantigens attenuatestumor growth in a preclinical model of colorectal cancer. Sci. Transl. Med.14, eabn6056.

[0164] 15. Subramanian, A., Tamayo, P., Mootha, VK, Mukherjee, S., Ebert, BL, Gillette, MA, Paulovich, A., Pomeroy, SL, Golub, TR, Lander, ES, et al. (2005). Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc. Natl. Acad. Sci. USA 102, 15545–15550.

[0165] 16. Blum, A., Wang, P., and Zenklusen, JC (2018). SnapShot: TCGA-Analyzed Tumors. Cell 173, 530.

[0166] result

[0167] Characterization of public, intratumoral conserved new connections

[0168] To investigate the expression of new connections, RNA sequencing (RNA-seq) data from the Cancer Genome Atlas (TCGA) were first used to identify unannotated splice reads in various cancer types. Figure 2A This analysis was applied to primary tumor samples with corresponding spatial mappings for tumor RNA-seq data. Figure 2B These include sequencing datasets from glioblastoma (GBM; n=167), low-grade glioma (LGG; n=516), lung adenocarcinoma (LUAD, n=517), lung squamous cell carcinoma (LUSC, n=501), mesothelioma (MESO, n=516), hepatocellular carcinoma (LIHC, n=371), gastric adenocarcinoma (STAD, n=415), kidney clear cell carcinoma (KIRC; n=533), kidney papillary cell carcinoma (KIRP; n=290), kidney chromophobe carcinoma (KICH, n=66), colon adenocarcinoma (COAD; n=458), and prostate adenocarcinoma (PRAD; n=497). To minimize contamination by healthy tissue cells, the procedure used only samples with a measured tumor purity of 0.60 or higher. Figure 2CProtein-coding, non-annotated joins were identified by selecting splicing events not characterized in GENCODE (data not shown). The positive sample rate (PSR) of joins represents the percentage of samples in the cohort expressing splicing events, with a join read frequency of at least 10% compared to classic spliced ​​joins. A filter for common neo-joins was added by focusing on robustly expressed neo-joins shared among patients, which exhibited elevated inter-patient PSRs within their tumor cohorts. TCGA ≥10% Figure 2D Following the previous new ligation nomenclature, splicing events expressed at less than 1% PSR (n=9166; PSR) were selected from normal tissue RNA sequencing data from the Genotype-Tissue Expression (GTEx) project. GTEx <1% (data not shown). Therefore, an average of 373 new connections and 202 common new connections were identified across all tumor types in TCGA. Figure 2E The generality of expression for all new connections is similar. Figure 2F Further characterization of common neoconnections across tumor types indicated that alterations to the 3' and 5' splice sites were predominant. Figure 2G Furthermore, the proportion of newly generated frameshifted connections remained consistent across all tumor types. Figure 2H While the expression of presumed public neoconnections was mostly disease-specific (not shown), hierarchical clustering revealed subgroups of shared neoconnections consistently expressed across multiple cancer types. Figure 2I These results indicate a significant conservation of neoconnection expression among patients, and highlight a subset of patient-conserved neoconnections expressed across multiple cancer types.

[0169] New conserved intratumoral connections were identified through multisite analysis.

[0170] Immunotherapy targeting a single TSA may be insufficient to eradicate the tumor, leading to recolonization by clones lacking a targeted neoantigen. In the case of targeted T-cell therapy, this underscores the importance of focusing on multiple neoantigens shared throughout the malignant compartment to prevent immune escape due to antigen loss. Neoconnections have previously been shown to generate immunogenic antigens, and for this purpose, ITHs (internal tract antigens) of common neoconnections are explored by locating neoconnection reads in multiple samples within the same tumor. Figure 3A RNA sequencing data from multiple intratumoral tumor samples from the prostate, liver, colon, stomach, kidney, and lung were analyzed to investigate whether common neoconnections persist across tumor landscapes in various cancer types (data not shown). Analysis of multi-site intratumoral datasets revealed subsets of common neoconnections expressed in multiple samples across various tumor landscapes, some of which were present in all biopsies and considered to be tumor-wide. Figure 3BExpression of common neoconnections was identified in multiple samples from the same tumor, indicating that a subset of splicing events is stably present in the tumor landscape. In datasets containing multiple cases with multi-site sampling, conserved neoconnections in the common space were readily expressed across all cases. Figure 3C Targeting these novel intratumoral connections, which are conserved in multiple patients, would provide a promising strategy for overcoming ITH in various cancer types.

[0171] Studies using single-cell RNA sequencing data from IDH1 wild-type and IDH1 mutant gliomas further demonstrate the notorious heterogeneity of CNS tumors, revealing the existence of numerous subclones with branching evolutionary trajectories. While publicly available multi-site RNA sequencing provides data from a small number of intratumoral biopsies, it is difficult to conclude whether a small sample is sufficient to fully reproduce the genomic landscape of a tumor, especially in highly heterogeneous tumors such as gliomas. Therefore, an internal dataset was built that significantly expands the number of intratumoral biopsy analyses across all glioma subtypes. Based on previous datasets, biopsies of approximately 10 maximum distance spatial mappings across 56 LGG and GBM tumors were evaluated. Figure 3D This was used for whole-exome sequencing and RNA sequencing. Similar to multi-region studies of other tumor types, the expression of novel connections was detected within tumors across multiple patients. Hierarchical clustering revealed subsets of novel connections associated with mutant IDH1 subtypes, wild-type IDH1 subtypes, or highly expressed in both subtypes. Figure 3E When the study initially identified putative expression (connection reads ≥10 reads) of 789 total neoconnections in both LGG and GBMTCGA cases, 774 (98.10%) neoconnections were confirmed to be detectable in more than one region within at least one tumor, and 547 (69.33%) neoconnections were detectable in 10% of the glioma subtype cohort. Figure 3F This indicates that most common neoconnections identified from the TCGA LGG / GBM analysis were consistently expressed across multiple tumor regions in the comprehensive patient database. Neoconnections present in multiple tumors but not all regions were defined as intratumorally conserved neoconnections (icNJs). Neoconnections detectable in all biopsies within a single tumor were termed tumor-wide neoconnections (twNJs): significantly, in at least one case, 280 (34.49%) characterized neoconnections were robustly expressed tumor-wide, half of which (n = 141, 17.87% of the total) were found tumor-wide in two or more cases. Figure 3G These findings highlight new categories of novel connections appearing in the tumor landscape across multiple patients, establishing a promising library of new targets for cancer therapies.

[0172] Public new connections are detectable at both RNA and peptide levels in patient-derived tumor samples.

[0173] My next attempt was to validate whether the expression of common neolinks and their protein derivatives could be detected in patient-derived transcriptomics and proteomics data. The remainder of the analysis and workflow focused on investigating glioma-specific common neolinks. RNA-seq data from patient-derived GBM (n=66; obtained from the Mayo Clinic Brain Tumor PDX National Resource) and LGG (n=2) cell lines were investigated. RNA-seq data from patient-derived primary astrocytoma and oligodendroglioma LGG cell lines were generated in previous studies. In total, 510 and 767 previously characterized common neolinks were detected in the GBM and LGG disease types, respectively. Figure 4A (4B). However, fragmented reads generated by short-read RNA sequencing can lead to difficulties in accurately identifying complete mRNA structures carrying splicing aberrations. Therefore, primers were designed to span selected new links and their flanking exons, and amplicon sequencing was performed on the PCR amplification products to detect continuous reads carrying the selected new links.

[0174] To evaluate the extent to which neolinks were translated into proteins, mass spectrometry (MS) data from glioma patients (n=447) derived from publicly available MS datasets were analyzed. The analysis confirmed the expression of 49 unique neolink-specific peptide sequences, which mapped back to 75 (9.5%) unique common neolinks. Figure 4C Sequence-specific searches in MS data and subsequent analysis of the resulting MS profiles confirmed that these detected peptide sequences spanned aberrant splicing regions (data not shown). When both glioma-specific neo-links were considered in conjunction with RNA sequencing and MS validation, the presence of 44 (5.6%) common neo-links expressed across samples from all patient sources was confirmed. Figure 4D Overall, these findings demonstrate the reproducibility of public new connections across multiple transcriptomics and proteomics platforms.

[0175] Novel linker encoding predictions across the entire tumor range were processed and novel antigens presented by MHC-I were analyzed.

[0176] Given that some novel links are translated and presented as targetable neoantigens, this study investigated whether characterized common novel links can generate peptides loaded onto MHC-I after proteasome treatment. A database of novel link-derived proteins was generated by computer simulation translating novel link-derived sequences from TCGA. The process iteratively traversed all possible n-mers of 8 to 11 amino acids. Figure 4EThose peptide sequences not present in normal tissue sequences within the reference human proteome dataset (n = 67,364; UniProt proteome ID #UP000005640) will be defined as tumor-specific n-mers. Past studies have typically considered MHC binding affinity as a major determinant of peptide-MHC presentation and immune response. However, presentation predictions based solely on binding affinity data often overlook factors beyond HLA binding affinity, such as endogenous peptide processing steps.

[0177] To address these oversights and follow the stepwise biological process of MHC binding following proteasome treatment, two independent prediction algorithms were incorporated into the pipeline to identify novel epitope sequences that could be treated and presented: MHCflurry 2.0 and HLAthena (data not shown). Each prediction algorithm demonstrated strong capabilities in predicting peptide treatment motifs or pan-MHC binding affinity. The antigen treatment predictor of MHCflurry 2.0 simulates allele-independent sequence characteristics and is favorable for neoantigen scoring if their sequences (combining flanking amino acids and n-mers) are consistent with established proteasome cleavage motifs. Meanwhile, the prediction algorithm of HLAthena uses a large peptidomimetics dataset modeled with 95 cell lines expressing a single class I HLA allele, allowing for independent HLA binding motif characterization. To identify putative common neoantigen candidates detectable across various patient profiles, the binding potential of n-mer candidates to the most commonly expressed human leukocyte antigen-A (HLA-A) haplotypes in a broad population was examined. Therefore, this analytical procedure investigated HLA-A 01:01, HLA-A 02:01, HLA-A 03:01, HLA-A 11:01 and HLA-A 24:02 Presentation probability of neoantigen candidates. To select high-binding targets, emphasis was placed on the top 10% of n-mer candidates scored in both the HLAthena and MHCFlurry 2.0 algorithms. Figure 4F-4G Candidates expected to generate these scores (n=832) were readily processed and presented in gliomas and subsequently retained for downstream analysis (data not shown). Mapping these top candidates back to novel links of their origin identified 315 links encoding neoantigens (NEJ; 39.92% of the initially characterized common neolinks) that yielded cancer-specific peptide sequences containing high-scoring n-mer candidates. Although a significant number of high-scoring n-mer candidates originated from frameshift and variable 3' splice site deletions (data not shown), presentation scores remained relatively consistent between NEJ-generated n-mer candidates with or without frameshift mutations or any particular type of alternative splicing. Figure 4H-4I To further narrow down the NEJ candidate list, 315 NEJs were cross-referenced with 44 novel connections previously characterized as shared across transcriptomics and proteomics platforms. Figure 4D Of the 44 new connections, 20 were characterized as NEJs by this analysis (data not shown), and due to the high prevalence of HLA alleles in the population and their extensive characterization in neoantigen studies, the focus was on generating connections with HLA-A. 02:01 Downstream analysis of NEJ (n=8) candidates with strong binding ( Figure 4J When characterizing the ITH of these eight new connections on the spatially mapped dataset, it can be seen that most of these NEJs exhibit high intratumoral conservation, particularly the new connections located within GNAS (NJs). GNAS ()( Figure 4K These findings suggest that new, conserved connections in the public sphere may generate subtypes that can be presented by MHC-I, thus becoming targets for immunotherapy.

[0178] NEJ-specific T-cell receptors can be isolated from neoantigen-reactive CD8+ T cells.

[0179] To determine whether NEJ-derived neoantigens could be recognized by T cells, in vitro sensitization (IVS) was performed to stimulate and identify neoantigen-responsive CD8+ T cell populations from healthy donor-derived peripheral monocytes (PBMCs) by increasing IFNγ levels stimulated by the neoantigen. Figure 5A Initial analysis focused on 4 out of 8 high-scoring NEJ candidates (NJ). GNAS NJ S100A6 NJ RPL22 NJ TCF12 They were identified as generating HLA-A 02:01 Polymers with high binding affinity ( Figure 4J-4K Due to the high diversity of TCR repertoires, it is theoretically expected that any individual could carry a naive T cell repertoire targeting any MHC: peptide complex. Following this logic, targeting HLA-A... 02:01+ Neoantigen-loaded autologous monocyte-derived dendritic cells (moDCs) collected from healthy donors (n=5) were used for in vitro saturation (IVS) of initial CD8+ T cells to ultimately obtain TCR gene sequences conferred specificity against these neoantigens. Throughout three 10-day cycles, T cells were sensitized with NEJ-derived neoantigen multimers, influenza multimer positive controls, or mature dendritic cells without peptide loading. To determine whether the neoantigen-responsive CD8+ T cell subset had expanded at the final time point, sensitized CD8+ T cells were cultured with T2 cells loaded with exogenously applied neoantigen peptides or influenza peptides. T2 cells lack peptide transporters (TAPs) involved in antigen processing, and therefore are often used as a model to detect T cell recognition of exogenous antigens in a non-competitive manner. Subsequent IFNγ and TNFα ELISA screening under the corresponding T2:CD8+ conditions revealed neoantigen-specific immunogenicity for two of the four public NEJ-derived neoantigens: NeoA RPL22 and NeoA GNAS ( Figure 5B These results indicate that a NEJ-specific CD8+ T cell subset exists in a healthy donor T cell pool and can be expanded in vitro.

[0180] To characterize and identify TCR sequences reactive to NEJ-derived multimers, researchers repeatedly performed peptide-loaded T2:CD8+ T cell co-culture assays on RPL22A3LIF-reactive and GNASA3LFS-reactive CD8+ T cell populations, and combined single-cell V(D)J sequencing and RNA sequencing for all co-culture conditions. Additionally, IFNγ and TNFα ELISA assays were performed on the supernatant under the same co-culture conditions to confirm the immunogenicity of each neoantigen against the CD8+ T cell population (data not shown). Neoantigen-specific TCR clonal types were associated with significantly elevated IFNG, TNFA, and GZMB signatures when co-cultured with APCs treated with the corresponding neoantigens, but not when co-cultured with bait peptide-loaded or unloaded APCs. Using this method, seven NeoA... RPL22 -Specific TCR, 2 from donor 3 (TCR R3.7 and TCR R3.9 ) and 5 from donor 4 (TCR R4.5 TCR R4.6 TCR R4.7 TCR R4.9 and TCR R4.11 ), and one NeoA from donor 4. GNAS -Specific TCR (TCR) G4.1 ()( Figure 5C Although only one NeoA was characterized.GNAS - A specific TCR clone, but this clone is the most proliferating TCR clone, with the TCR pool expanding to over 4% in the CD8+ T cell population. Figure 5D These findings indicate that the neoantigen-specific CD8+ subgroup of donor-derived PBMCs amplifies upon recognition of MHC-I-presented multimers, thereby allowing for accurate determination of neoantigen-specific TCR sequences.

[0181] NEJ-specific TCRs recognize NEJ-derived neoantigens only in an MHC-I restricted manner.

[0182] To further establish the identification of TCR R3.9 Reactivity and TCR G4.1 - Specificity of reactive T cell clones, transduced transgenic CD8-expressing TCR-deleted triple reporter (TR) Jurkat76 cells and PBMC-derived CD8 cells using a lentiviral (LV) vector encoding TCR α- and β-chains. + T cells. Jurkat76 cells are TCR α-negative and β-negative human T cell derivatives that allow for the non-competitive introduction of exogenous TCRs. The Jurkat76 cells used in this study expressed the transgenic CD8. TR Jurkat76 cells have previously been transduced with a transcriptional reporter construct, enabling NFAT, NF-κB, and AP-1 response elements to drive the expression of eGFP, CFP, and mCherry, respectively. Therefore, TCR activation can be assessed by simultaneously quantifying fluorescent protein expression levels ( Figure 5E TCR-transduced TR Jurkat76 cells were co-cultured with T2 cells loaded with different concentrations of neoantigens to demonstrate a dose-dependent response. Figure 5F Both TCRs showed neoantigen recognition even at low levels of peptide presentation (1 nM), indicating high immunogenicity of the neoantigen target and high affinity of the corresponding TCR. The antigen specificity of these responses was supported by negligible TCR activation in the presence of maximal bait peptide presentation (1 μM). TCR-transduced PBMC-derived CD8+ T cells exhibited similar dose-dependent neoantigen-specific behavior. Figure 5G , 5H TCR-transduced CD8+ T cells were stained for CD137 and CD107a surface expression, and T cell stimulation and proliferation were assessed by flow cytometry. T cell activation was observed at neoantigen-peptide loading concentrations as low as 1 pM. Figure 5G Similarly, IFNγ and TNFα expression levels measured by ELISA indicated that both TCRs were highly specific, as evidenced by their ECGs. 50 As shown in the range of 0.01 to 0.1 nM ( Figure 5H To confirm that T cell activation is mediated by MHC-I-restricted neoantigen presentation, neoantigen-loaded T2 cells were treated with an HLA-blocking antibody before co-culturing with TCR-transduced TR Jurkat76 cells. As expected, the anti-HLA-I antibody completely blocked activation. Figure 5I Finally, both TCRs were subjected to alanine scanning mutagenesis to determine the presence of other known human proteins that share the peptide motifs recognized by these TCRs. (In NeoA) RPL22 and NeoA GNAS A single alanine mutation was introduced at each position within the n-mer epitope. Triple reporter Jurkat76 cells transduced with TCR were co-cultured with the neoantigen isoform containing the substituted residues, and key residues were characterized by reduced TCR activation. Figure 5J Peptides bind to HLA-A The change in stability at 02:01 indicates that the substituted residues are related to HLA-A. 02:01 Binding or TCR recognition is important. Comparison of key residue sequences with a normal human proteome library (UniProt proteome ID#UP000005640) demonstrated the absence of known human proteins sharing the same set of key amino acid residues for TCR recognition, indicating that the TCR specifically recognizes cancer-specific regions of the neoantigen sequence. These results reveal a TCR that recognizes public NEJ-derived neoantigens with robust sensitivity and specificity, and provide potential immunotherapeutic approaches for targeting such novel neoantigen libraries using TCR-engineered T cells.

[0183] Neoantigens derived from NEJ are produced through endogenous translation, processing, and MHC-I presentation.

[0184] Next, it was verified that the newly derived transcripts were efficiently translated, processed by the proteasome, and presented on the cell surface by MHC-I. The presentation of NEJ-derived neoantigens was evaluated using two separate methods: functional TCR-recognition and HLA immunoprecipitation (HLA-IP), followed by liquid chromatography-MS / MS (LC-MS / MS). Figure 6A Using HLA-A encoding. The mRNA of the 02:01 and full-length mutant transcripts was transfected into COS7 or K562 cells. COS7 and K562 cells lack HLA expression, thus allowing the introduction of non-competitive transgenes of any HLA of interest.

[0185] To demonstrate that neo-linked expression at the transcriptional level can ultimately lead to a neoantigen-mediated immune response, COS7 cells transfected with the same HLA and full-length mutant transcripts were co-cultured with TCR-transduced TR Jurkat76 cells. COS7 cells were incubated for 24 hours after mRNA electroporation to achieve optimal expression before the introduction of TCR-transduced TR Jurkats76. Both transfected COS7 cell lines induced TCR... R3.9 and TCR G4.1 - Neoantigen-specific activation in transduced TR Jurkat76 cells indicated that the identified common NEJ underwent a major biological step of protease cleavage and MHC-I binding, followed by MHC-I restriction presentation and recognition (Fig. 6B-C). HLA-I ligands were purified from COS7 cells transfected with HLA and mutant transcripts using affinity-column-based immunopurification. MS analysis showed that the same NeoA... GNAS The multimer candidate was identified as being in HLA-A 02:01 and full length NJ GNAS - High confidence and high abundance of HLA-A2-binding peptide in transfected COS7 cells. Similarly, in HLA-A 02:01 and NJ RPL22 - In transduced COS7 cells, two NeoA... RPL22 All polymers were detected with high confidence, among which NeoA scored the highest. RPL22 The 9-mer was identified at high abundance. Figure 6D These results are consistent with earlier computer predictions of proteasome processing and HLA binding (data not shown).

[0186] TCR-transduced CD8 + T cell-mediated cytotoxic effects against glioma cells expressing NEJ-derived neoantigens.

[0187] Based on the impressive sensitivity of neoantigen-specific TCRs ( Figure 5F , 5G It is expected that the intrinsic level of common NEJ expression in tumor cells is sufficient to trigger a neoantigen-specific immune response. To test this hypothesis, TCRs were... R3.9 -and TCR G4.1 - Transduced triple reporter Jurkat76 cells with NJ expression previously identified in RNA sequencing data RPL22 or NJ GNAS IFNγ-pretreated HLA-A2+ LGG and GBM cell lines were co-cultured (data not shown). As expected, flow cytometry analysis after co-culture showed that both TCR-transduced CD8+ T cell lines produced tumor-specific immune responses against the studied GBM cell line. Figure 6E To assess whether this same neoantigen recognition was sufficient to induce tumor-specific killing, the researchers used TCR... R3.9 and TCR G4.1 The hybrid form was transduced into PBMC-derived CD8+ T cells. The mouse constant region was replaced with the NeoA target. RPL22 Or NeoA GNAS A specific constant region of the TCR was identified to allow for precise pairing of the transgenic α and β chains. High transduction efficiency populations of murine constant region Ab-stained CD8+ T cells were screened by flow cytometry. The efficacy of NJ targeting endogenously expressed TCR was evaluated by co-culturing TCR-transduced CD8+ T cells with the glioma cell line GBM115. RPL22 and NJ GNAS The tumor-specific cytotoxicity was observed, exhibiting a consistent proliferation and seeding profile. As a positive cytotoxicity control, the GBM115 cell line loaded with the neoantigen peptide was used to define maximum cell killing. At a 1:1 effector:target ratio, TCR... R3.9 and TCR G4.1 - Transduced CD8+ T cells effectively demonstrated cytotoxic killing of glioma cells at 37.18% and 61.52% of rates, as detected using an impedance-based cell growth cytotoxicity assay (xCELLigence). Figure 6F Increasing the effector:target ratio to 2:1 increased the potency of both clones, respectively, by TCR. R3.9 and TCR G4.1 - Transduced CD8+ T cells killed 42.97% and 82.74% of tumor cells, respectively. Figure 6F To confirm that tumor-specific killing is initiated by TCR recognition of the MHC: peptide complex, HLA-I blocking antibodies were introduced into the co-culture conditions. Compared with the isotype control, the presence of anti-HLA-I antibodies prevented killing (…). Figure 6G The conclusion is that NEJs are endogenously processed and presented in tumor cells at levels sufficient to initiate neoantigen-specific CD8+ T cell tumor-specific cytotoxicity.

[0188] Disease subtype-specific factors drive differences in new connection expression

[0189] The next interest was to investigate the differences in putative neoconnection expression across all glioma subtypes. Both TCGA and the spatially mapped RNA-seq platform of this invention revealed that, compared to their wild-type variants, the mean expression of putative neoconnections was significantly greater in mutant IDH1 gliomas. Figure 7A , 7BFurther noteworthy is the significant differentiation in neoconnection expression when comparing wild-type IDH1 (IDH1wt), mutant IDH1 astrocytoma (IDH1mut-A), and mutant IDH1 oligodendroglioma (IDH1mut-O) glioma subtypes. While IDH1mut-A gliomas showed significantly higher average neoconnection expression compared to IDH1wt gliomas, IDH1mut-O neoconnection expression far exceeded that of the other disease subtypes. Figure 7C , 7D Paired Pearson correlation analysis to explore the prevalence of neoconnection expression in conjunction with somatic mutations in commonly mutated splicing factors revealed the co-occurrence of mutations in several splicing factors with IDH1 mutations (data not shown). FUBP1 mutations have previously been reported to be prevalent in the IDH1mut-O glioma subtype. However, hierarchical clustering of neoconnections revealed no significant trend in neoconnection expression with mutational status of FUBP1, SF3A1, or NIPBL (data not shown). Previous studies have reported that dysregulation of a single splicing factor can lead to aberrant splicing, and for this purpose, aberrant expression levels of splicing-related genes associated with neoconnection generation were explored. To investigate the possible drivers of observed differences in neoconnection expression among glioma subtypes, differential gene expression analysis (DESeq2) was performed on RNA sequencing data from all three disease subtypes of TCGA. Gene set enrichment analysis (GSEA) was subsequently performed on the DESeq2 results, and differentially expressed gene sets were investigated (data not shown). GSEA highlighted gene ontology processes (GOBP) in the process of gene ontology biology (GOB). Figure 7E ) and Gene Ontology Cellular Components (GOCC) database ( Figure 7F In the study, splicing-related gene sets were significantly upregulated in mutated IDH1 cases compared to their wild-type counterparts. When sorted based on expression increases from new connections, highly expressed splicing-related genes clustered together in most of the mutated IDH1 disease subtypes (data not shown).

[0190] To further investigate specific splicing-related genes that can lead to increased neoconnection expression in mutated IDH1 gliomas ( Figure 7G-7H ), selected GOBP splicing-related genes (n=24) that showed a 1.5-fold increase in expression compared to wild-type in mutant IDH1 cases with statistical significance (p<0.05). Figure 7GIt is noteworthy that, as previously reported, CELF2 and ELAVL4 generate splicing aberrations when overexpressed. For this purpose, a correlation analysis was performed between CELF2 expression and the expression of all 789 common neo-connections identified through the procedure. A greater proportion of neo-connections were identified across all glioma subtypes, with expression typically increasing with the expression of these two splicing-related genes (mean Pearson correlation coefficient > 0.10). Figure 7I Of the 789 new connections, 359 (45.5%) showed increased expression with increasing CELF2 expression, while 81 (10.3%) tended to show decreased expression with increasing CELF2 levels. New connections with the highest correlation associated with two splicing-related genes (data not shown) were selected, confirming that the expression of these new connections was significantly enhanced in mutant IDH1 cases. Figure 7J Then, siRNA-mediated CELF2 knockdown was performed in the mutant IDH1 cell lines SF10417 and SF10602 (data not shown), and decreased expression of related neoconnections was observed in both cases. Figure 7K These findings suggest that the generation of new connections is associated with dysregulation of splicing regulatory genes specific to disease subtypes, and that regulating these genes can effectively lead to changes in the level of new connections.

[0191] When the GOBP splicing-related gene set (data not shown) was reanalyzed, subclusters of genes were significantly downregulated in IDH1mut-O cases. Most of the genes found within these clusters were located on chromosome 1p or chromosome 19q. A distinctive diagnostic feature of IDH1mut-O glioma is the co-deletion of chromosome 1p and chromosome 19q. Therefore, the observed consistent reduction in expression of chromosome 1p-based splicing-related genes is consistent. To assess whether the downregulation of these genes could lead to the characteristic increase in putative neoconnection expression observed in the IDH1mut-O subtype, GOBP splicing-related genes (n=26) with statistically significant (p<0.05) 1.5-fold reduced expression compared to IDH1mut-A and IDH1wt cases in IDH1mut-O cases were selected. Figure 7HAmong these splicing genes, previous reports have shown that disrupting appropriate SNRPD2 expression can lead to aberrations in splicing. Correlation analysis between SNRPD2 expression and the expression of 789 neojoints across all glioma subtypes supports the hypothesis that reduced SNRPD2 expression leads to higher neojoint expression. Of the 789 neojoints, 93 (11.8%) tended to show increased expression with increasing CELF2 levels, while 385 (48.8%) showed increased expression with decreased SNRPD2 expression. Notably, siRNA knockdown of SNRPD2 in the GBM115 cell line (data not shown) resulted in a significant increase in the expression level of the associated neojoints across three replicates. Figure 7L This indicates that the loss of expression of splicing-related genes due to the co-deletion of chromosomes 1p and 19q can lead to the upregulation of specific neo-joints. These results suggest that specific components of the splicing mechanism can be targeted to potentially increase the expression levels of neo-joints and ultimately increase neoantigen expression.

[0192] discuss

[0193] While numerous neoantigen discovery processes have confirmed the responsiveness of immune cells to tumor-specific antigens, few studies have considered the ITH of their targets, and to date, no studies have investigated the ITH of neoantigens derived from RNA splicing aberrations. These results are the first to identify novel, conserved intratumoral links that generate tumor-specific peptides to be presented on HLA-I molecules, and demonstrate that recognition of these neoantigens can lead to tumor-specific killing. In validating NJ... GNAS Immunogenicity and TCR G4.1 With specificity, it successfully identified common whole-tumor NEJs that were robustly expressed within the patient's tumor. Figure 4K When characterizing neoantigens, their persistent presence throughout the tumor landscape is often overlooked. This study highlights a novel subset of NEJs that can be identified across the entire tumor range through multiple biopsy sites. Incorporating the intratumoral conservation of neoantigen expression into the computational pipeline offers advantages in addressing the ITH challenges across multiple cancer types. RPL22 and NeoA GNAS The endogenous processing and presentation further demonstrate that the novel computerized process accurately predicts tumor-specific peptide candidates that are appropriately processed by the proteasome and strongly bound to MHC-I.

[0194] Although only 8 high-resolution HLA-A molecules were studied The immunogenicity of four of the novel antigens presented at 02:01 was tested, but the immunogenicity of the candidate NEJ (NJ) against a specific TCR clone was successfully demonstrated. RPL22 and NJ GNASOther cancers in pan-cancer studies have also been identified as new connections. Notably, specific pan-cancer studies with large cohorts of samples from multiple sites have shown that NJ... RPL22 NJ was expressed in multiple samples within the same tumor. GNAS It is expressed across the entire tumor spectrum in various tumors. Figure 3C The higher expression level of the classic GNAS allele compared to RPL22 appears to be the primary reason for the detection of NJ in all studies. GNAS The main reason for the high probability (data not shown). This supports TCR. G4.1 The phenomenon of greater immunogenicity and tumor-specific killing ( Figure 7F -H), because NeoA GNAS The generation and presentation frequency of these genes is higher. These results indicate that the public NEJs (especially NJs) identified in this study are more likely to be generated and presented. GNAS It can serve as a target for cancer types other than glioma.

[0195] This study investigated the dysregulation of splicing-related genes associated with IDH1 mutations in gliomas. Mutations in IDH1 and IDH2 are prevalent in various cancer types beyond gliomas, including acute myeloid leukemia (AML), chondroma, chondrosarcoma, undifferentiated sinus carcinoma, and angioimmunoblastic T-cell lymphoma. Further analysis could determine whether similar splicing-related gene dysregulations exist in these cancers and whether their aberrant expression leads to a similar level of increased neo-connection levels. Similarly, this question involves mutations in splicing factors frequently found in tumors. Although analysis of mutated FUBP1 showed only a moderate alteration in neo-connection expression (data not shown), splicing factor mutations are expected to contribute to the generation of splicing aberrations across various cancers.

[0196] Overall, this study highlights that RNA splicing abnormalities are a poorly understood, robust source of public and intratumorally conserved TSAs that can be recognized by the immune system. The ability to target these tumor-wide neoantigens with engineered T cells enables robust therapies to address the ITH challenges faced by many neoantigen-based immunotherapies. Ultimately, the results from this study can enable the design of effective vaccine compositions containing identified tumor-wide neoantigen targets, as well as engineered T cell forms that target intratumorally conserved splice-derived antigens across multiple cancer types.

[0197] Example 2

[0198] This embodiment includes some information from the above embodiments, but also includes additional information.

[0199] result

[0200] Characterization of public, pan-cancer new connections

[0201] To investigate novel linker expression, RNA sequencing (RNA-seq) data from the Cancer Genome Atlas (TCGA) was first used to identify unannotated linker reads across cancer types (data not shown). For ITH evaluation, tumor samples from 12 tumor types with multiple spatial mappings were included (data not shown). Samples with inferred tumor purity ≥60% were included only when identifying protein-coding, unannotated links (data not shown) (data not shown). The positive sample rate (PSR) of links represents the percentage of samples expressing novel links within the cohort, with a linker read frequency ≥1% relative to classic spliced ​​links. This allowed us to define common novel links as those exhibiting elevated PSRs in each TCGA tumor cohort (PSR ≥ 1%). TCGA Those ≥10% (data not shown). Following the new linker nomenclature, cancer-specific splicing events were selected as those in the Genotype-Tissue Expression Database (GTEx) project (n=9166; PSR). GTEx Those with PSR <1% in normal tissues (data not shown). An average of 94 common neoconnections were identified for each TCGA tumor type (data not shown), which were expressed at similar frequencies (data not shown). Further characterization of common neoconnections across tumor types demonstrated a variable distribution of splicing types (data not shown) and a consistent proportion of splicing events producing frameshifts (data not shown). Subsets of these neoconnections were detected in recent splicing studies (data not shown). Unbiased hierarchical clustering revealed that cases from the same tumor type tended to cluster together, suggesting similar neoconnection expression profiles in patients with the same cancer type. Some subsets of neoconnections were expressed in multiple tumor types (data not shown). Therefore, neoconnection expression can be conserved across multiple cancer types, suggesting potential pan-cancer immunotherapy targets arising from aberrant splicing.

[0202] The new connections exhibit intratumoral heterogeneity.

[0203] Immunotherapy targeting a single TSA may be insufficient to eradicate tumors due to the growth of cancer cells that do not possess TSA. For targeted T-cell therapy, this underscores the importance of focusing on multiple neoantigens shared throughout the tumor to prevent immune evasion through antigenic evolution. Neoconnections can generate immunogenic antigens. To this end, ITHs of common neoconnections were explored by filtering neoconnection reads from multiple samples from the same tumor (data not shown). RNA-seq data from multiple intratumoral samples from the prostate, liver, colon, stomach, kidney, and lung cancer were analyzed to investigate the presence of common neoconnections throughout the tumor landscape (data not shown). This analysis revealed common neoconnections expressed in multiple intratumoral samples (data not shown), and spatially conserved common neoconnections were identified across all cases in a dataset with multi-site sampling, further revealing ideal neoconnection targets (data not shown).

[0204] Among the cancer types analyzed in this study, gliomas are notoriously characterized by larger ITHs, which further complicates immunotherapy. It is important to determine the required number of samples to adequately represent the tumor's transcriptomic landscape. Therefore, the number of intratumoral biopsies was increased across the three major glioma subtypes. Spatial mapping samples from approximately 10 tumors with maximum distances (data not shown) from 51 tumors with exome and RNA-seq were evaluated, and the expression of neo-connections within the tumor was detected across multiple patients. Iterating from one sample to ten samples, the number of universally expressed neo-connections was negatively correlated with the number of samples (data not shown). These findings highlight the important need to sample multiple biopsies for each tumor type to more confidently characterize neo-connections across the entire tumor.

[0205] Hierarchical clustering of large intratumoral datasets revealed a subset of novel connections that tended to be associated with the mutated isocitrate dehydrogenase (IDHmut) or wild-type (IDHwt) subtypes (data not shown). Interestingly, IDHmut gliomas expressed a significantly greater number of whole-tumor novel connections compared to IDHwt gliomas. While the proportion of whole-tumor novel connections across patients was significantly lower than that expressed in other subclones (data not shown), at least one whole-tumor novel connection was detected in 45 (88.2%) patients (data not shown), with 13 (25.5%) patients expressing more than 50 whole-tumor novel connections (data not shown). Although 774 (98.1%) of the LGG and GBM novel connections characterized by TCGA were detectable in multiple regions of at least one tumor in the cohort, 37 (4.7%) of novel connections were found across more than 10% of all samples in the study cohort (data not shown). Therefore, most common novel connections derived from TCGA LGG / GBM analyses were concentrated in multiple tumor regions in multisampled tumor data, but not whole-tumor. This finding suggests that combining new connections may be a reasonable approach to covering the entire tumor landscape.

[0206] Neoconnections were characterized as conserved in both time and space at metastasis and recurrence. Analysis of RNA-seq data from common cutaneous melanoma (SKCM) revealed 13 (9.6%) neoconnections detected across metastatic sites in at least one patient (data not shown). When matched primary / metastatic pairs were examined in TCGA, 43.8% to 72.6% of neoconnections identified in the primary tumor in COAD, PRAD, and SKCM cancers were observed to persist in metastasis (data not shown). Similarly, when matched primary / recurrence pairs were examined across TCGA COAD, GBM, LGG, LIHC, and LUAD cancers, an average of 36.4% of neoconnections were conserved at recurrence (data not shown). In the glioma dataset, 79.2% and 82.3% of neoconnections were conserved in gliomas at recurrence after temozolomide treatment (data not shown). Overall, these findings demonstrate that neoconnections can persist both spatially and temporally.

[0207] Disease subtype-specific factors drive differential expression of new connections

[0208] Noting the subtype-specific expression of neoconnections across glioma subtypes in the TCGA and spatial mapping datasets (data not shown), we investigated the dysregulation of splicing mechanisms that may contribute to these patterns. While previous studies have reported a potential role for IDH mutations in splicing abnormalities, this study reveals further complexity. Both the TCGA and spatial mapping datasets showed a significantly larger total number of common neoconnections per case in IDH mutants compared to IDH wild-type gliomas (data not shown). Although IDHmut-A gliomas exhibited a significantly higher average neoconnection expression level compared to IDHwt gliomas, IDHmut-O neoconnection expression was even greater (data not shown). Paired Pearson correlation analysis was performed to investigate whether neoconnection expression was associated with somatic mutations in commonly mutated RNA splicing factors (data not shown). High correlations were found between FUBP1, SF3A1, or NIPBL mutations and IDH mutations. FUBP1 mutations were prevalent in IDHmut-O gliomas. However, hierarchical clustering of neoconnections revealed no significant trend in neoconnection expression with FUBP1, SF3A1, or NIPBL mutation status (data not shown). Dysregulation of a single splicing factor can lead to aberrant splicing. Based on these findings, the correlation between aberrant expression levels of splicing-related genes and neoconnection generation was investigated. To investigate the drivers of neoconnection expression differences in possible glioma subtypes, differentially expressed gene sets from three glioma subtypes from TCGA were evaluated (data not shown). Gene set enrichment analysis highlighted splicing-related gene sets that were significantly upregulated in mutated IDH cases compared to their wild-type counterparts, spanning the Gene Ontology Biological Processes (GOBP) (data not shown) and Gene Ontology Cellular Components (GOCC) databases (data not shown). When sorted based on neoconnection expression, splicing-related genes highly expressed in both mutated IDH tumor subtypes largely clustered together (data not shown).

[0209] To further investigate specific splicing-related genes (data not shown) that may lead to increased neoconnection expression in IDHmut gliomas, GOBP splicing-related genes (n=24) with statistically significant (p<0.05) expression increases of 1.5-fold compared to wild-type in IDHmut cases were selected. Notably, splicing aberrations have been previously reported when CELF2 is overexpressed. Correlation analysis of CELF2 expression against all 789 common neoconnections identified a higher percentage of neoconnections whose expression generally increased with increasing CELF2 expression across all glioma subtypes (mean Pearson correlation coefficient >0.10) (data not shown). Of the 789 neoconnections, 359 (45.5%) increased with increasing CELF2 expression, while 81 (10.3%) tended to decrease with increasing CELF2 levels. Neolinks most highly associated with splicing-related genes were selected, and siRNA-mediated CELF2 knockdown was performed in two patient-derived IDH-mutant cell lines, SF10417 and SF10602 (data not shown), with a trend toward decreased neolink expression observed in both lines (data not shown). 244 neolinks significantly upregulated in IDH mutants compared to wild-type glioma cases were characterized (log2 change >1.5, p <0.05), some of which were detected in other TCGA IDHmut cancer types (data not shown). RNA sequencing analysis showed that, compared to untreated controls, CELF2 knockdown resulted in a 19 (8.6%) and 28 (12.7%) decrease in IDHmut-related neolink expression in SF10417 and SF10602, respectively (data not shown), and candidate IDHmut neolink expression correlated with increasing expression of IDHmut-related splicing-related genes (data not shown). These findings suggest that the prevalence of new connections is regulated by altered expression of RNA-binding proteins in tumor subtypes, and that regulation of these genes can lead to changes in the level of new connections.

[0210] When the GOBP splicing-related gene set was re-examined (data not shown), subclusters of genes that were significantly downregulated in IDHmut-O cases were identified. Most of the genes found in these clusters were located on chromosome 1p or 19q, and their co-deletion is a unique diagnostic feature of IDHmut-O gliomas. To assess whether the downregulation of these genes would contribute to the characteristic increase in putative neoconnection expression seen in the IDHmut-O subtype, GOBP splicing-related genes (n=26) with statistically significant (p<0.05) 1.5-fold reduced expression in IDH1mut-O cases compared to IDH1mut-A and IDH1wt cases were selected (data not shown). Among these splicing genes, disruption of normal SNRPD2 and SF3A3 expression has previously been reported to lead to splicing aberrations. Correlation analysis of SNRPD2 and SF3A3 expression with expression of 789 neoconnections across all glioma subtypes supported the hypothesis that reduced SNRPD2 and SF3A3 expression may contribute to greater neoconnection expression (data not shown). Of the 789 new connections, 385 (48.8%) increased with decreasing SNRPD2 expression, while 93 (11.8%) tended to increase with increasing CELF2 levels. Similarly, with increasing SF3A3 expression levels, 178 (22.6%) new connections tended to increase expression, while 127 (16.1%) tended to decrease expression. Notably, siRNA knockdown of SNRPD2 or SF3A3 in the GBM115 cell line (data not shown) resulted in a significant increase in the expression levels of their associated new connections (data not shown). 52 IDHmut-O-related new connections were also characterized as those significantly upregulated in mutant IDH-O glioma cases compared to their mutant IDH-A and wild-type counterparts (log2-fold change >1.5, p <0.05). In GBM115 cells treated with SF3A3 or SNRPD2 siRNA, increased expression of 7 (13.5%) and 4 (7.7%) IDHmut-O-related neoconnections was observed (data not shown). While previous studies have linked splicing factor mutations to neoconnection in cancer, these results reveal a previously undescribed mechanism demonstrating that reduced wild-type splicing factor expression consistently generates neoconnections. These findings suggest that a common altered component of RNA splicing mechanisms in glioma is mechanistically associated with increased neoconnection expression.

[0211] Finally, this analysis was extended to the remaining TCGA cancer types used in this study to identify tumor subtypes with significantly dysregulated neoconnection expression. While neoconnection expression was relatively consistent across SKCM, KIRP, KICH, and PRAD cancers (data not shown), iCluster 3 and iCluster 6 subtypes in TCGA LIHC and LUAD showed significantly differential neoconnection expression compared to other iCluster subtypes (data not shown). Gene set enrichment analysis of the six LUAD iCluster subtypes revealed reduced expression of splicing-related gene pathways. Notably, this set of 23 splicing-related genes was consistently downregulated in LUAD iCluster 6 compared to the other five iCluster subtypes. Together, these results suggest that, in addition to splicing factor mutations, dysregulated expression of typical splicing-related genes can lead to disease-specific neoconnection generation.

[0212] Public new connections are detectable at both RNA and peptide levels in patient-derived tumor samples.

[0213] The next step was to validate whether the expression of common neoconnections and their protein products could be detected in cell line transcriptomic and tumor tissue proteomic data. For this analysis, the focus was on common neoconnections expressed in gliomas, as these tumors exhibit high levels of ITH and poor clinical outcomes. First, xenograft (PDX) (n = 66) and LGG (n = 2) cell lines derived from GBM patients were investigated. Overall, the expression of 767 (97.2%) and 510 (64.6%) characterized common neoconnections was measured in GBM and LGG, respectively (data not shown). Batch RNA-seq had limited reads across each splicing aberration. Therefore, we designed primers spanning subsets of neoconnections and their flanking exons, performed deep amplicon sequencing, and confirmed the mRNA expression of reads spanning neoconnections in glioma cell lines (data not shown).

[0214] To test whether the neolinks were translated into proteins, mass spectrometry (MS) data from glioma patients (n = 447) were analyzed using publicly available MS datasets. This analysis detected the expression of novel peptides mapping back to 302 (38.3%) unique common neolinks (data not shown). These peptide sequences were confirmed to span aberrant splicing regions after sequence-specific searches in the MS data and subsequent analysis of the resulting MS spectra (data not shown). Interestingly, 41.7% of the detected peptides mapped back to neolinks leading to frameshifts (data not shown), indicating that frameshift-inducing splicing aberrations can lead to detectable levels of translated peptides. Peptidommic analysis determined that the transcripts encoded by the neolinks were actively translated into detectable levels of protein products. When RNA-seq and MS confirmation of glioma-specific neolinks were considered, the presence of 192 (24.3%) common neolinks expressed across all patient-derived samples was validated (data not shown). These findings demonstrate the existence of recurrent common neolinks that generate tumor-specific peptides.

[0215] Novel tumor-wide connectivity encoding predicts novel epitopes processed and presented by HLA.

[0216] Considering that a subset of newly translated links could be presented as targetable new epitopes, we investigated whether 789 characterized publicly disclosed new links could generate peptides loaded onto human leukocyte antigen (HLA) class I after proteasome processing. All correctly translated new link-derived sequences from TCGA were computer-simulated to generate a dataset of new link-derived proteins. All possible 8–11 amino acid n-mers (data not shown) were iterated, with tumor-specific n-mers defined as those not present in the UniProt reference normal human tissue proteome dataset.

[0217] Predicting class I HLA-presented peptides requires integrating key aspects of the antigen presentation mechanism, including peptide processing and HLA binding. To this end, two independent prediction algorithms were introduced to identify novel epitope sequences that may be processed and presented: MHCflurry 2.0 and HLAthena (data not shown). To rank candidate common novel peptides, the binding potential of n-mer candidates to the most prevalent HLA-A type expressed in a broad population was evaluated. These analyses investigated HLA-A in 36 epidemiologically dominant HLA-A alleles (data not shown). 01:01, HLA-A 02:01, HLA-A 03:01, HLA-A 11:01 and HLA-A 24:02 Presentation probability of neoantigen candidates. These alleles are commonly expressed by the majority of the global population. To select high-binding targets, focus was placed on the top 1% of n-mer candidates (data not shown) that scored highest in both algorithms. Candidate n-mers (n=832) that generated these scores were predicted to be processed and presented in gliomas and retained in downstream analyses (data not shown). Mapping these best candidates to the new links from which they originated, 315 new links encoding novel peptides were identified (NEJ; 39.9% of the original characterized common new links) that yielded cancer-specific peptide sequences containing the best n-mer candidates. Although a larger number of the highest-scoring n-mer candidates generated from frameshift mutations and alternative exon 3' splice sites (data not shown), presentation scores of n-mer candidates from NEJs with or without frameshift mutations or any particular splice type remained relatively consistent (data not shown). To further narrow down the list of NEJ candidates, 315 NEJs were cross-referenced with 192 novel connections previously characterized as shared between transcriptomics and proteomics platforms (data not shown). Of the 192 novel connections, 81 were characterized as NEJs by this analysis (data not shown), many of which encode multiple strongly predicted candidates. Downstream analysis focused on 32 candidate NEJs that were predicted to be associated with HLA-A. The 02:01 binding was strong because this allele is prevalent in North American and European populations and can be benchmarked against other neoantigens (data not shown). When these 32 neolinked ITHs were studied in datasets from spatially mapped samples, high intratumoral conservation was observed for most of these NEJs, particularly the neolinks encoding two-nucleotide A3 loss located within GNAS (NJs). GNAS (Data not shown). These findings demonstrate that conserved common new connections within tumors can generate novel HLA-presented peptides.

[0218] NEJ reactive T cell receptors can be isolated from donor CD8+ T cells.

[0219] The next step was to determine whether the novel NEJ-derived peptide could drive T-cell immunogenicity. In vitro sensitization (IVS) was performed to identify neoantigen-reactive CD8+ cells from healthy donor (HD)-derived peripheral monocytes (PBMCs). + T cell population (data not shown). Initial analysis focused on predicting the production of HLA-A... A subset (n=4) of the 32 top NEJ candidates for high-affinity conjugates of HLA-A (data not shown). Therefore, for HLA-A... 02:01 +Autologous mononuclear cell-derived dendritic cells (moDCs) loaded with novel peptides, collected by HD (n=5), were subjected to in vitro sequencing (IVS) on naive CD8+ T cells to retrieve TCR gene sequences that confer specificity against these novel antigens. Subsequently, these sequences were used in the corresponding APCs:CD8+ T cells. + Under certain conditions, IFNγ ELISA assays showed neoantigen reactive immunogenicity in two of the four common NEJ-derived neoantigens (NeoAs): NeoA RPL22 and NeoA GNAS (Data not shown). NeoA GNAS This results in the loss of two nucleotides in A3, leading to a frameshift and premature stop codon. NeoA RPL22 The loss of 6 nucleotides in the A3 frame resulted in the loss of two amino acids in the α-helix (data not shown). These results also indicate that NEJ reactivity with CD8... + T cells can exist in a naturally occurring T cell pool.

[0220] To retrieve the TCR gene sequence conferring responsiveness to these neoantigens, the sequence was repeated in NeoA. RPL22 -and NeoA GNAS -Reactive CD8 + APC:CD8 loaded with peptide on T cell population + T cells were co-cultured and combined with single-cell V(D)J and RNA-seq were performed. Neoantigen-reactive TCR clonal types were associated with significantly elevated IFNG, TNFA, and GZMB transcripts in a neoantigen peptide-specific manner. Using this method, seven NeoA... RPL22 -Reactive TCR, two from donor 3 (TCR R3.7 and TCR R3.9 Five from donor 4 (TCR) R4.5 TCR R4.6 TCR R4.7 TCR R4.9 and TCR R4.11 ), and a NeoA from donor 4 GNAS -Reactive TCR (TCR G4.1 (Data not shown). Although only one NeoA is characterized. GNAS - A reactive TCR clone, but this same clone is the most proliferative TCR clone, expanding to CD8. + More than 4% of the TCR repertoire in the T cell population (data not shown). Neoantigen-reactive CD8 + The expansion of T cell clones indicates the strong immunogenicity of these two neoantigens.

[0221] NEJ reactive TCRs recognize NEJ-derived neoantigens in an HLA-restricted manner.

[0222] To further test and identify the TCR R3.9 and TCR G4.1 - Specificity of reactive T cell clones was achieved by transducing TCR-deleted triple reporter (TR) Jurkat76 cells with a lentiviral vector encoding recovered TCR α- and β-chains, expressing CD8a / b heterodimer (Jurkat76 / CD8) or PBMC-derived CD8a / β-chain. + T cells. TR Jurkat76 / CD8 cells possess NFAT, NF-κB, and AP-1 response elements that drive the expression of eGFP, CFP, and mCherry, respectively (data not shown). TCR-transduced TR Jurkat76 cells co-cultured with T2 cells loaded with different concentrations of neoantigen peptides showed dose-dependent responsiveness (data not shown). Both TCRs showed neoantigen recognition at nM levels, indicating relatively high functional affinity for the corresponding TCRs. The antigen specificity of these receptors is supported by negligible TCR activation in the presence of supraphysiological levels of control peptides (1 μM). TCR-transduced PBMC-derived CD8 + T cells exhibited similar dose-dependent neoantigen-specific behavior (data not shown). Targeting the surface-expressed T cell activation and degranulation markers CD137 and CD107a against TCR-transduced CD8... + T cells were stained to quantify markers of T cell activation and effector function, respectively. T cell activation was observed at concentrations as low as 1 pM of the neoantigen-peptide (data not shown). Similarly, IFNγ and TNFα expression levels, measured by ELISA, indicated that both TCRs were potent, as indicated by their half-maximal effective peptide concentration (EC50). 50 The values ​​were shown to be between 0.01 and 0.1 nM (data not shown). Treatment of neoantigen-loaded T2 cells with HLA-blocking antibodies prior to co-culturing with TCR-transduced TR Jurkat76 cells confirmed that neopeptide T cell activation was HLA-dependent (data not shown).

[0223] Next, alanine scanning mutagenesis was performed to determine whether both NEJ-responsive TCRs could recognize peptides derived from off-target normal human proteins. Triple reporter Jurkat76 / CD8 cells transduced with the TCRs were cultured against neoantigen isoforms with residue substitutions, and key residues were defined by those leading to reduced TCR activation (data not shown). The altered recognition of the variant peptides indicated that the substituted residues were crucial for TCR recognition. Comparison of the peptide recognition motifs of each TCR with a normal human proteome library (UniProt proteome ID#UP000005640) showed that no known human proteins shared the key residues required for TCR recognition. Together, these results reveal TCRs that recognize only NEJ-derived common neoantigens with robust sensitivity and highlight the potential for immunotherapies using TCR-engineered T cells to target this novel class of shared neoantigens.

[0224] Finally, by analyzing data from three known NEJs... GNAS HLA-A expression 02:01 An immunosurveillance study was conducted on archived PBMC samples from glioma patients to determine whether circulating NEJ-reactive CD8 could be detected in glioma patients. + T cell population (data not shown). NEJ data for these PBMC samples. GNAS IVS resulted in an immune response detected in one of three glioma patients, against unrelated HLA-A antibodies. The 02-restricted 9-mer neoantigen dextramer control showed no immunogenicity. Figure 9A These findings further support the immunogenicity and potential clinical applications of targeting NEJ-derived neoantigens.

[0225] NEJ-derived public neoantigens are endogenously processed and presented by HLA.

[0226] Next, we tested whether the novel linker-derived transcript generated peptides that were functionally presented by HLA and recognized by reactive TCRs. Presentation of the NEJ-derived neoantigen was assessed using two methods: functional TCR recognition and HLA immunoprecipitation, followed by liquid chromatography-MS / MS (data not shown). To determine whether neolinker transcript expression led to immune recognition, COS7 cells transfected with HLA-A2 and full-length mutant transcripts were compared with TCR-transduced TR Jurkat76 or CD8. + T cell co-culture. TCR R3.9 and TCR G4.1 - Transduced TR Jurkat76 and CD8 +T cells responded to COS7 cells transfected with their respective neoantigens, demonstrating endogenous processing and presentation of the common NEJ (data not shown). HLA-I ligand immunopurification based on affinity column was then performed on COS7 cells co-transfected with HLA / mutant transcripts. MS analysis identified the same NeoA with high confidence. GNAS The peptide is a highly abundant HLA-A2-binding peptide. Similarly, it is associated with HLA-A... 02:01 and NJ RPL22 Two NeoA variants were detected with high confidence on co-transfected COS7 cells. RPL22 NeoA, a novel peptide with high ratings RPL22 The 9-mer was identified at a high relative abundance (data not shown). Furthermore, HLA-A was detectable in the unmodified GBM cell line (GBM115). NeoA (restricted at 02:01) GNAS Peptides (data not shown). This finding indicates that the physiological levels of newly linked expression in tumor cells are sufficient to generate NEJ-derived neoantigens. Together, these experimental observations confirm computer simulation predictions of proteasome processing and HLA binding (data not shown).

[0227] TCR-transduced CD8 + T cell-mediated cytotoxicity against glioma cells expressing NEJ-derived common neoantigens.

[0228] Based on the sensitivity of neoantigen-specific TCRs (data not shown) and the endogenous presentation of NEJ-derived neoantigens by tumor cells (data not shown), it is expected that the physiological level of common NEJ expression in tumor cells will trigger a neoantigen-specific cytotoxic T cell response. The TCR-transduced CD8+ response was evaluated. + T cells respond to endogenous expression of NJ RPL22 and NJ GNAS HLA-A 02:01 + Cytotoxicity of tumor cells. As a positive control, tumor cells loaded with neoantigen peptides were used to define maximum cell killing. At an effector:target ratio of 1:1, TCR... R3.9 and TCR G4.1 - Transduced CD8 + T cells mediated TCR-dependent cytotoxicity against GBM115 cells (data not shown). G4.1 - Transduced CD8 +T cells mediated comparable levels of tumor killing against a second glioblastoma cell line, GBM102, and two melanoma cell lines, RPMI-7951 and WM-266-4 (data not shown). The addition of an HLA-I blocking antibody partially blocked the killing compared to the isotype control, confirming that tumor cell killing is initiated by TCR recognition of the HLA:peptide complex (data not shown). G4.1 - Transduced CD8 + T cell expression and HLA-A2 negative expression NJ GNAS Co-culture of the GBM cell line (Mayo PDX GBM39) revealed HLA-A2-dependent cytotoxicity. Figure 10A These results indicate that the recognition and killing of tumor cells expressing NEJ are mediated by HLA-dependent neoantigen presentation. This is in contrast to non-transduced CD8. + T cells, CD8 transducers of TCR, co-cultured with tumor cells + Increased surface expression of CD107a and CD137 on T cells further confirms neoantigen-specific T cell activation. Figure 10B Together, these data indicate that NEJ has sufficient capacity to make the neoantigen specific to CD8. + T cells achieve tumor-specific cytotoxicity through endogenous processing and presentation.

[0229] discuss

[0230] Our analysis of multi-site sample cohorts showed that NJ was expressed in multiple samples within the same tumor. RPL22 Most notably, NJ GNAS It is expressed across the entire tumor spectrum in various tumor types, including glioma, mesothelioma, prostate cancer, and hepatocellular carcinoma (data not shown). The discovery of a targetable, tumor-wide neoantigen in GBM provides a novel potential therapeutic strategy for this devastating disease. The higher expression levels of the typical GNAS allele compared to RPL22 may contribute to the detection of NJ in all analyses. GNAS The popularity of TCR. G4.1 Observations of greater immunogenicity and tumor-specific killing (data not shown) due to the generation and presentation of NeoA GNAS The frequency is higher. Indeed, NJ can be detected in the loop. GNAS Specific CD8 + T cell population, which originates from tumors exhibiting NJ GNAS HLA-A expression 02:01 + Glioma patients (data not shown).

[0231] The study also investigated the dysregulation of splicing-related genes associated with increased neoconnection expression in IDH-mutant gliomas. Mutations in IDH and IDH2 are prevalent in other cancers, including acute myeloid leukemia (AML), cholangiocarcinoma, chondrosarcoma, undifferentiated sinus carcinoma, and angioimmunoblastic T-cell lymphoma. This study demonstrates the dysregulation of splicing factor expression in different disease types, and that these abnormalities lead to significant changes in neoconnection generation. In the case of IDH-mutant oligodendrogliomas, SNRPD2 expression was reduced due to the characteristic co-deletion of chromosomes 1p and 19q, and targeted knockdown increased neoconnection expression. This suggests that components of the RNA splicing mechanism are mechanistically associated with neoconnection generation. In future studies, inhibiting these components could potentially enhance the expression of targetable NEJ-derived neoantigens.

[0232] method

[0233] Data and code availability

[0234] Spatially mapped biopsy RNA sequencing data will be stored. HLA-IP and LC-MS / MS data will be publicly available from the date of publication. Additionally, single-cell V(D)J sequencing data for identified TCRs will be publicly available from the date of publication. Accession numbers are listed in the key resources table. All original code used to identify broadly public new connections in tumors is stored on GitHub (https: / / github.com / dakwok / SSNIP) and has been publicly available from the date of publication. Any further information required for reanalysis of the data reported in this paper may be obtained from the principal contact upon request.

[0235] Human clinical dataset

[0236] RNA sequencing data from the following studies were used in intratumoral multi-region sampling cohorts for various cancer types:

[0237] 1. This paper describes multi-regional sampling for glioblastoma and low-grade glioma.

[0238] 2. Yang et al. (Genome Medicine, 2022), for multi-region sampling of hepatocellular carcinoma.

[0239] 3. Joung et al. (PLoS One, 2016) used multi-region sampling for hepatocellular carcinoma, gastric adenocarcinoma, renal cell carcinoma and colon adenocarcinoma.

[0240] 4. Ku et al. (Briefings in Bioinformatics, 2021), for multi-region sampling in prostate cancer.

[0241] 5. Meiller et al. (Genome Medicine, 2021), for multi-region sampling of mesothelioma.

[0242] 6. Bakir et al. (Nature, 2023), for multi-region sampling in non-small cell lung cancer.

[0243] If FASTQ files are available, the novel ligand prediction pipeline is immediately used to analyze the expression of novel ligands within multi-region samples. If RNA sequencing data is only available in BAM format, the sequencing files are converted to FASTQ format using Picard software (version 2.7.7a). Novel ligand prediction is detailed in the Methods section.

[0244] Data download

[0245] Data Download: Batch RNA sequencing data for samples of glioblastoma (GBM; n=167), low-grade glioma (LGG; n=516), lung adenocarcinoma (LUAD, n=517), lung squamous cell carcinoma (LUSC, n=501), mesothelioma (MESO, n=516), hepatocellular carcinoma (LIHC, n=371), gastric adenocarcinoma (STAD, n=415), clear cell renal carcinoma (KIRC; n=533), papillary renal carcinoma (KIRP; n=290), chromophobe renal carcinoma (KICH, n=66), colon adenocarcinoma (COAD; n=458), and prostate adenocarcinoma (PRAD; n=497) were downloaded from TCGA in FASTQ format. The download of intratumoral multi-region sampling sequencing data was detailed in the preceding sections. Similarly, batch RNA sequencing data for 9166 normal tissue samples were downloaded in FASTQ format from the Genotype-Tissue Expression Database (GTEx) repository. We received bulk RNA sequencing data from 66 patient-derived GBM cell lines from the Mayo Clinic's National Resource Center for Brain Tumor Xenografts. We also downloaded proteomics data from 100 GBM samples from the Clinical Proteomics Tumor Analysis Collaboration (CPTAC).

[0246] RNA sequencing alignment

[0247] All downloaded RNA sequencing datasets were individually aligned using a STAR-based alignment pipeline. A genome index containing unannotated links was constructed using the STAR software (version 2.7.7a) through the initial alignment pass of the input data. Complete set of command line parameters: --runThreadN 1 \ --outFilterMultimapScoreRange 1 \ --outFilterMultimapNmax 20 \ --outFilterMismatchNmax 10 \ --alignIntronMax500000 \ --alignMatesGapMax 1000000 \ --sjdbScore 2 \ --alignSJDBoverhangMin1 \ --genomeLoad NoSharedMemory \ --limitBAMsortRAM 80000000000 \ --readFilesCommand gunzip -c \ --outFilterMatchNminOverLread 0.33 \ --outFilterScoreMinOverLread 0.33 \ --sjdbOverhang 100 \ --outSAMstrandFieldintronMotif \ --outSAMattributes NH HI NM MD AS XS \ --limitSjdbInsertNsj2000000 \ --outSAMunmapped None \ --outSAMtype BAM SortedByCoordinate \ --outSAMheaderHD @HD VN1.4 \ --twopassMode Basic \ --outSAMmultNmax 1 \, and compared using the GRCH37 STAR index file.

[0248] TCGA Sample Selection and Gene Expression Quantification

[0249] TCGA tumor samples with an absolute tumor purity greater than 0.60 were retained for downstream computer simulation analysis. (Aran et al., 2015; Ceccarelli et al., 2016) Non-mitochondrial, protein-coding transcripts defined by the Ensembl Homo sapiens GRCH37.87 gene annotation gene transfer format (GTF) file were selected, and protein-coding transcript isoforms in TCGA RNA sequencing data were selected and retained using this curated set. Transcript-level expression data (log2[RSEM-TPM+0.001]) of all TCGA samples were downloaded from the UCSC Xena Toil pipeline and converted to standard TPM values. Protein-coding transcript isoforms with a median TPM ≥ 10 were retained for downstream analysis. In the case of TCGA gliomas, subsequent expression data in TPM were subsets categorized into six disease types: all cases (n=429), GBM cases (n=115), LGG cases (n=314), IDH1-WT cases (n=166), IDH1-MUT astrocytoma cases (n=140), and IDH1-MUT oligodendroglioma (n=123). Protein-coding transcript isoforms with a median TPM ≥10 in at least one of the six disease types were retained for further analysis.

[0250] Representation of public new connections

[0251] For the counting of public cancer-specific splicing events, a custom R script was designed to detect and quantify unannotated cancer-specific splicing events found in each corresponding patient cohort. Variable splicing events were quantified in the corresponding sj.out.tab file within the detected connection counts from the output file derived from the STAR alignment in the previous step. Splicing events detected from the GRCh37.87 GTF sj.out.tab (GENCODE v33) file were removed to define unannotated splicing connections. Unannotated splicing connections overlapping with non-mitochondrial protein-coding genes identified in the previous step were retained for further analysis. All splicing connections with fewer than 10 target splice reads (counts) or fewer than 20 total splice reads (depth) across the entire cohort were removed. Similar to previous studies, splice frequency was calculated as the sum of the total number of target splice reads divided by the sum of splice reads from both target and canonical connections. Splicing connections with a read frequency greater than 1% were retained for downstream analysis. Common splice junctions were defined as junctions presumed to be expressed with the aforementioned total read count, read depth, and read frequency criteria in at least 10% of the study patient cohort, and these junctions were preserved for further analysis. To characterize cancer-specific splicing events, also known as neo-junctions, all junctions presumed to be expressed with the same parameters in more than 1% of the GTEx normal samples were removed.

[0252] Detection of cancer-specific intron retention events

[0253] Intron splicing events were detected and characterized using IRFinder v1.2.3. RNA sequencing data from TCGA (GBM / LGG) and GTEx (CNS) were aligned to GRCh37 (hg19) and imported into the software for the detection of intron retention events. Analysis based on a generalized linear model (GLM) was used to assess differential intron retention. The intron retention ratio was calculated as the sum of (intron reads / (intron reads, normal splice reads)). Significant changes in intron retention were defined as (1) not less than 10% in both directions and (2) an adjusted p-value less than 0.05. The PSR of intron retention events in TCGA or GTEx was defined as the number of cases meeting these criteria divided by the total number of cases in the cohort. Possible cancer-specific intron retention neoconnections were characterized as intron retention events with a TCGA PSR ≥ 0.10 and a GTEx PSR < 0.01.

[0254] Transcriptome validation of newly expressed connections

[0255] Detection of novel connections expressed in patient-derived GBM / LGG cell lines RNA sequencing data derived from the GBM PDX cell line were downloaded from the Mayo Clinic Brain Tumor Patient-Derived Xenograft National Resource Center. Patient-derived LGG cell lines were generated from surgically resected specimens from the Neurosurgery Brain Tumor Center at the University of California, San Francisco. The RNA sequencing data from the GBM and LGG cell lines were compared and processed as described above. Common neo-sponge connections with a splice join count >0 per million (CPM) were considered detectable in the cell line-derived RNA sequencing data. Detection of novel connections expressed in multi-regional cases: In a spatially mapped glioma case cohort, approximately ten or more anatomical biopsy samples at maximum distance were collected from each patient, allowing for assessment of intratumoral genetic heterogeneity via batch RNA sequencing and whole-exome sequencing. Multi-region sequencing data for various other cancer types varied in the number of sampling regions per tumor and are detailed in the relevant references (data not shown). RNA sequencing data collected from each multi-region sample were processed and aligned as described above. Presumed novel connections previously characterized in TCGA were searched. Common novel connections with CPM > 0 were considered detectable. Common novel connections with presumed expression (≥ 10 splice reads) in two or more mapped samples within the same case were considered spatially conserved novel connections. Novel connections detected in all multi-region samples within the same tumor were considered tumor-wide novel connections.

[0256] Proteomic validation of expressed novel linker-derived peptides

[0257] From the hypothetical neo-links detected in the above process, a database of all plausible peptides derived from all neo-links was generated. Transcripts encoding the neo-links were generated by mapping the link coordinates to the hg19 human genome assembly within the Ensembl annotation database (AH13964, EnsDb.Hsapiens.v75). The prediction of the neo-link-derived amino acid sequences was then performed, and the translated sequence (methionine start residue, the sequence after removing the first stop codon) was appropriately preserved for downstream n-mer iterations. To detect neo-link-derived peptides in GBM cases, .RAW files of GBM and LGG MS data stored in the Clinical Proteomics Tumor Analysis Consortium (CPTAC, n=99), Bader et al. (n=99), Lam et al. (n=92), and Yanovich-Arad et al. (n=84) were analyzed. MaxQuant (v1.6.17.0) was used to identify trypsin sequences from the corresponding MS datasets. Predicted novel linker-derived peptides, decoy sequences, and the human reference proteome (Uniprot proteome ID#UP000005640) were input as a FASTA file into MaxQuant, and trypsin sequences from the input file were matched against publicly available MS databases. Cancer-specific peptides spanning novel linker-derived protein sequences were considered MS-confirmed. The relative detection levels of novel linker-derived peptides and normal tissue-derived peptides were assessed using their log2 (peak intensity). In addition to the default settings, the following commands and parameters used in MaxQuant for MS analysis were modified and used: Digestion mode = Trypsin / P; Maximum omission = 3; Minimum peptide length = 5; Minimum peptide length for nonspecific search = 5.

[0258] Peptide processing and prediction of HLA binding and presentation

[0259] Cancer-specific transcripts with relevant novel linkages were computer-simulated translated into their corresponding amino acid sequences. Libraries of all possible peptides ranging from 8 to 11 amino acids in length were then generated, and cancer-specific sequences were selected by removing those detectable in normal tissue peptide isoforms in a reference human proteome dataset (Uniprot proteome ID#UP000005640). All cancer-specific peptides and their upstream and downstream flanking sequences (maximum flanking length 30 amino acids) were independently analyzed and ranked using MHCFlurry 2.0 and HLAthena MSiC. In both cases, HLA-A was targeted. 01:01, HLA-A 02:01, HLA-A 03:01, HLA-A 11:01 and HLA-A 24:02 HLA-I binding affinity was assessed. In HLAthena's evaluation of antigen binding and presentation for corresponding HLA haplotypes, peptides were assigned to alleles by rank with a threshold of 0.1. Peptide aggregation analysis was performed using up to 30 flanking amino acids from each of the N-terminus and C-terminus as background, without logarithmic transformation of expression. The baseline MHCFlurry 2.0 model was used, including a peptide:HLA-I binding affinity (BA) predictor and an antigen processing (AP) predictor. Overall, peptide:HLA presentation scores were characterized in both MHCFlurry 2.0 and HLAthena using the mhcflurry_presentation_score and the MSiC_HLA score. To select high-conjugation peptides, a list of the top 10 percentile peptide:HLA complexes from both prediction algorithms was compiled.

[0260] Cell culture

[0261] GBM PDX cell culture: Patient-derived xenograft (PDX) glioblastoma cell lines, GBM34, GBM43, GBM108, GBM115, GBM118, GBM102, GBM137, GBM148, GBM164, and GBM195, were obtained from the Mayo Clinic's National Resource Center for Brain Tumor Xenografts. The xenograft lines were cultured according to recommended conditions in previous literature and passaged a maximum of 20 times before reverting to earlier passages. Cells were cultured in Dulbecco modified Eagle medium (DMEM) supplemented with 10% fetal bovine serum and 1% penicillin and streptomycin (P / S). Before use, cells were cultured at 4°C with DPBS (containing calcium and magnesium) and 10% laminin (Gibco... TM Overnight cell culture plates (Cat. #23017015) Primary patient-derived GBM / LGG cell culture keep:Primary patient-derived wild-type IDH1 GBM (SF7996), mutant IDH1 astrocytoma (SF10602), and mutant IDH1 oligodendroglioma (SF10417) cell lines were previously generated from isolated glioma biopsies and cultured as described above. Cells were cultured in serum-free glioma neural stem cell (GNS) culture medium containing Neurocult NS-A (STEMCELL Technologies Cat. #05751), supplemented with N-2 supplement (Invitrogen Cat. #17502048), B-27 supplement (vitamin A-free) (Invitrogen Cat. #12587010), 1% P / S, 1% glutamine, and 1% sodium pyrophosphate. Before immediate use in culture, supplement GNS medium with 20 ng / mL EGF (Peprotech Cat. #AF-100-15), bFGF (Peprotech Cat. #AF-100-18B), and PDGF-AA (Peprotech Cat. #AF-100-13A). Similar to the GBM PDX cell line, incubate overnight at 4°C with DPBS (containing calcium and magnesium) and 10% laminin (Gibco) before use. TM Cat. #23017015) cell culture plate. Jurkat76 cell culture: Jurkat76 cells were used as TCR α- and β-negative human T cell derivatives that allow for the non-competitive introduction of exogenous TCRs. CD8+ Jurkat76 cells were cultured in RPMI supplemented with 10% fetal bovine serum and 1% P / S. T2 cell culture: This study used T2 cells to monitor the response of immune cells to exogenous antigens of interest in a non-competitive environment. T2 cells lack peptide transporters (TAPs) involved in antigen processing; therefore, these cells were induced with exogenously administered peptides to enable them to interact with HLA molecules (especially HLA-A). 0102) Association and presentation. T2 cells were cultured in IMDM medium supplemented with 20% FBS. COS7 and K562 cell cultures: The COS7 (ATCC Cat. #CRL-1651) and K562 (ATCC Cat. #CCL-243) cell lines were selected as primate and artificial antigen-presenting cell (aAPC) models, respectively. These cell lines do not express HLA molecules, which allows for the introduction of HLA alleles of interest. COS7 cells were cultured in DMEM medium supplemented with 10% FBS and 1% P / S. K562 cells were cultured in IMDM medium supplemented with 10% FBS and 1% P / S. THP-1 cell culture:Immunoreactivity against neoantigens presented by dendritic cells was studied using THP-1 cells (ATCC Cat.#TIB-202). THP-1 cells were cultured in RPMI-1640 supplemented with 10% FBS.

[0262] siRNA-mediated knockdown of splicing-related genes

[0263] Cells were seeded in 2 mL of antibiotic-free medium in 6-well plates at the following densities: GBM115 – 45,000 cells per well, SF10417 – 100,000 cells per well, and SF10602 – 100,000 cells per well. Twenty-four hours after seeding, cells were transfected to a final concentration of 30 nM by adding 400 μL of reaction solution containing serum-free medium, 2.0 μL of DharmaFECT 1 reagent (Horizon, #T-2001-02), and its corresponding siRNA pool (an equimolar mixture of four siRNAs). The medium was replaced with complete medium 24 hours post-transfection. 72 hours post-transfection, RNA was isolated and purified using the Zymo Quick-RNA microprep kit (Zymo Research, #R1058).

[0264] Quantitative reverse transcription PCR (RT-qPCR)

[0265] 1000 ng of DNase-treated RNA was converted to cDNA using the iScript cDNA Synthesis Kit (BioRad, #1708891). The cDNA was then diluted 1:3 with ultrapure, nuclease-free water, and 2 μL was used for each qPCR reaction. Quantitative reverse transcription polymerase chain reaction (RT-qPCR) was performed using the Applied Biosystems POWER SYBR Green Master Mix (Applied Biosystems, #4367659). All samples were run in biological triplets, and each biological triplet was technically tripled using Quantstudio 5 (Thermo Scientific). All gene expression data were normalized to the housekeeping gene GUSB. The cycling program was as follows: 50°C for 2 min, 95°C for 10 min, then 40 cycles of 95°C for 15 s and 60°C for 60 s. Dissociation curves were run to confirm specific product amplification. Primer sequences for each gene were designed using NCBI Primer for mRNA expression analysis.

[0266] Amplicon sequencing for validating novel ligation expression

[0267] RNA from the respective cell lines was isolated and purified using the Zymo Quick-RNA microprep kit (Zymo Research, #R1058). 1000 ng of DNase-treated RNA was converted to cDNA using the iScript cDNA Synthesis Kit (BioRad, #1708891). The cDNA was then diluted 1:3 with ultrapure, nuclease-free water, and 2 μL was used for each PCR reaction. For each amplicon and for each cell line, 16 reactions were performed using a Q5 High Fidelity 2x Master Mixture (NEB, #M0492L) with primers containing partial Illumina adapters. The reaction mixture was set according to the manufacturer's instructions. The products were then purified by separation on a 1.0% agarose gel at a constant voltage of 100 volts for 1 h, followed by purification using the Monarch DNA Gel Extraction Kit (NEB, #T1020L). The purified product was quantified using the qubit high-sensitivity dsDNA kit (Invitrogen, #Q32851) and prepared and submitted according to the Azenta Life Sciences (Genewiz) amplicon sequencing guide.

[0268] Sensitization of healthy PBMCs from in vitro donors

[0269] Purchase HLA-A from StemExpress 02:01:01 Positive PBMCs, in fresh or frozen form. Immediately administer approximately 1×10 9 Fresh PBMCs (StemExpress Cat. #LE001F) were allocated proportionally to 3 × 10⁶ cells. 8 Aliquots of cells were aliquoted and cryopreserved in liquid nitrogen, with one aliquot used in downstream IVS. Approximately 3 × 10⁻⁶ cells were aliquoted from each vial. 8 One vial of cryopreserved PBMCs (StemExpress Cat. #PBMNC300C) was used for each IVS procedure. The PBMCs were thawed with 1:1000 Benzonase:RPMI (Sigma Aldrich Cat. #E8263). Following the manufacturer's instructions, a CD14+ population was isolated from the PBMCs using CD14+ Miltenyi microbeads (Miltenyi Biotec Cat. #130-050-201). The isolated CD14+ cells were then incubated in untreated 24-well plates at 5 × 10⁻⁶ cells / well. 5Cells were seeded at a density of 1000 cells / well in CellGenix GMP DC medium (CellGenixCat # 20801-0500) supplemented with 1% human serum (Sigma Aldrich Cat # H6914), 1% P / S, 1000 U / mL recombinant human IL-4 (Peprotech Cat. #200-04), and GM-CSF (Peprotech Cat. # 300-03). On day 3, recombinant human IL-4 and GM-CSF (both 1000 U / mL) were added to the DC cultures. On day 5, in addition to supplementing with recombinant human IL-4 and GM-CSF (both 1000 U / mL), the DC cultures were matured with 250 ng / mL LPS (Sigma Aldrich Cat. # L6529). On day 6, naive CD8+ T cells were isolated from the thawed CD14- population using the EasySep Human Naive CD8+ T Cell Isolation Kit (STEMCELL Technologies Cat. # 19258) according to the manufacturer's instructions. The isolated naive CD8+ T cells were cultured at 5 × 10⁶ cells per well. 5 Cells were seeded at a density of 1000 cells / well in 48-well plates in X-Vivo 15 medium (Lonza Cat. #04-418Q) supplemented with 5% human serum, 1% P / S, and 10 ng / mL recombinant human IL-7 (Peprotech Cat. #200-07). On day 8, adherent mature DCs were harvested from the plates with cold PBS. The collected DCs (1 × 10⁶ cells / well) were diluted with 1 μM of neoantigen peptide, influenza peptide, or no peptide. 6 DCs (cells / mL) were exogenously loaded at 37°C for 1 hour. Then, either peptide-loaded or unloaded DCs were co-cultured with naive CD8+ T cells in 48-well plates, with an optimal DC:T cell ratio of 1:4. Co-culture was maintained for 10 days in X-Vivo 15 medium supplemented with 10 ng / mL recombinant human IL-7, 10 ng / mL recombinant human IL-15 (Peprotech Cat. #200-15), and 60 ng / mL recombinant human IL-21 (Peprotech Cat. #200-21), with restimulation with IL-7 and IL-15 every 2 days. Cells were reseeded into subsequent 24-well, 12-well, and 6-well plates based on confluence. This completed the first round of IVS with the neoantigen and influenza peptide. On days 19 and 29, sensitized CD8+ T cells were reintroduced for the second and third rounds of stimulation with freshly loaded DCs, and co-culture was maintained for another 10 days until the second and third rounds of IVS were completed. Immunogenic cytokine assays were performed at the end of the second and third rounds of IVS to determine the presence of an expanded population of peptide-responsive T cells.

[0270] Mutation-specific ELISA screening

[0271] Aliquots containing CD8+ T cells from each parent IVS well were harvested and evenly divided into 96-well progeny plates, with each well containing 1×10⁶ cells. 5 T2 cells were used to stimulate triplicate progeny wells for 16 hours at an effector cell to target cell (E:T) ratio with T2 cells loaded with the neoantigen peptide of interest, control peptide, no peptide, or no T2 cells. T2 cells were loaded with 1 pM to 1 μM of the neoantigen peptide of interest, control peptide, or no peptide at 37°C for 1 hour. Influenza-reactive T cells were co-cultured with influenza peptide-loaded T2 cells as a positive control. The co-culture supernatant was collected and diluted according to the manufacturer's instructions for use in IFNγ (BD Biosciences Cat. #555142) and TNFα (BD Biosciences Cat. #555212) ELISA. ELISA readouts were performed on an Epoch microplate spectrophotometer (BioTek Instruments) using BioTek Gen5 data analysis software (version 1.11). Wells showing significant increases in IFNγ and TNFα expression levels were selected for downstream single-cell immunoassays using single-cell RNA and V(D)J sequencing.

[0272] Single-cell immunoassay profile

[0273] Once the expanded neoantigen-reactive CD8+ T cell population from IVS was identified, single-cell RNA and V(D)J sequencing were performed using the 10x Genomics platform. Prior to sequencing, CD8+ T cells from the expanded neoantigen-reactive (ELISA-selected positive) wells were harvested and co-cultured with 1 μM of the neoantigen peptide of interest, control peptide, or peptide-free T2 cells at a 1:1 E:T ratio. One co-culture was repeated for 3 hours for single-cell RNA sequencing analysis, and another for 16 hours for IFNγ and TNFα ELISA confirmation. The final cell concentration was adjusted to approximately 1 × 10⁻⁶ cells / well. 4Cells / μL, with initial cell viability of at least 90% to maximize the likelihood of achieving the desired cell recovery target. Sequencing of independent CD8+ T cells and unloaded T2 single cultures with co-culture conditions was performed to differentiate cell types in downstream single-cell sequencing analysis. Preparation for single-cell sequencing analysis was performed using the ChromiumNext GEM Single Cell 5' Kit v2 (Dual Indexing) (10xGenomics, Cat. #CG000331). Gel beads in an emulsion (GEM) were generated by combining single-cell 5' gel beads, separator oil, and a master mixture containing cells onto a Chromium Next GEM chip K. Cell lysis and barcoded reverse transcription of RNA were performed within the corresponding GEM for all single cells. Barcoded cDNA products were recovered via GEM-RT post-cleaning and PCR amplification. cDNA quality control and quantification were performed on a Fragment Analyzer system (Agilent Technologies). 50 ng of cDNA was used to construct 5' gene expression libraries, and each sample was indexed using the Chromium i7 Sample Indexing Kit. The procedure was performed on an Illumina NovaSeq 6000 sequencer at the Institute for Human Genetics (IHG) at UCSF, with at least 20,000 read pairs for the 5' gene expression library per cell. Enrichment products were measured using the Fragment Analyzer system. 50 ng of the enriched TCR product was used for library construction. Single-cell V(D)J enriched libraries were then sequenced on the Illumina NovaSeq 6000, with at least 5,000 read pairs for the V(D)J library per cell. Cell Ranger 7.0.0 (10x Genomics Cloud Analysis) was used to preprocess raw single-cell RNA sequencing and identify V(D)J clonotypes. Annotation files 'vdj_GRCh38_alts_ensembl-3.1.0-3.1.0' and 'GRCh38-3.0.0' were used to demultiplex cell barcodes, perform read alignment, and generate a feature-barcode matrix. Only cells with usable clonoid information were retained for downstream analysis. Single-cell gene expression and corresponding V(D)J sequences of candidate T cell clonoids were analyzed on the Loupe V(D)J browser. Single cells with detectable CD8A expression were specifically isolated and characterized as CD8+ T cell populations and subsequently grouped according to their TCR clonoids. To identify T cell clonoids associated with neoantigen-specific responses, expanded TCR clonoids exhibiting significantly increased levels of IFNG, TNF, and GZMB expression under T2 conditions of T cells:neoantigen loading compared to T2 conditions of T cells:control loading and T2 conditions of T cells:unloaded were selected.

[0274] HLA typing

[0275] OptiType 1.3.1 is used to perform HLA allelic typing of available glioma cell lines from available WES data using default parameters.

[0276] plasmids and peptides

[0277] HLA-A Both the 02:01 and novel ligation-derived gene sequences were synthesized and cloned into the pTwist Lenti SFFVPuro WPRE vector (Twist Biosciences). Full-length and truncated multimeric versions encoding wild-type and mutant GNAS and RPL22 sequences were generated. TCRα / β was synthesized and cloned into the pTwist Lenti SFFV vector (TwistBiosciences). HPLC-grade novel ligation-derived neoantigen peptide multimers (>95%) were manufactured by TC Laboratories.

[0278] Lentiviral transduction

[0279] HEK293T cells were used at a rate of 1 × 10⁻⁶ cells per well. 6 Cells were plated at a density in 6-well plates containing 2 mL of DMEM medium supplemented with 10% FBS without antibiotics. Approximately 18 to 24 hours later, or at 90% confluence, HEK293T cells were transfected with the expression constructs (see above), as well as the lentiviral packaging plasmids pMD2.G (Addgene, #12259) and psPAX2 (Addgene, #12260). TCRα / β transduction:1.0 μg TCRα / β transfer plasmid, 0.75 μg psPAX2, and 0.25 μg pMD2.G were combined with 200 μL Opti-MEM (Thermo Fischer Scientific Cat. #31985062). 6 μL Xtremegene HP was added to this mixture, and complex formation was allowed to occur at room temperature for 15 minutes. The reaction mixture was then added to the corresponding HEK293T cells. The transfection medium was replaced with fresh DMEM after 24 hours. Viral supernatant was collected after 48 hours, and functional viral titers were performed by seeding Jurkat76 / CD8 cells or PBMC-derived CD8+ T cells in 6-well plates at 60–70% confluence. Viral transduction was performed by supplementing the viral stock solution with a final concentration of 4 μg / mL polybrene using a 3-fold serially diluted viral stock solution. The medium was replaced 24 hours after viral transduction. After 3–4 days, transduction efficiency was assessed by measuring the surface expression of TCRα / β and CD3 using fluorescence-activated cell sorting (FACS). Cells exhibiting high double-positive expression of TCRα / β and CD3 were flow-cytoscoped and maintained in downstream co-culture and immunogenicity assays. HLA and neoantigen transfer guide: HLA-A expression The 02:01 construct was linearized and digested with BamHI and XhoI (New England Biolabs) restriction enzymes, and purified using the Zymoclean Gel DNA Recovery Kit (Zymo Research Cat. #D4007). HLA-A was then... The 0201 sequence was linked downstream of the EF1A-core promoter and upstream of IRES in the lentiviral construct, followed by the blasticidin resistance gene. 1.0 μg of HLA-A 02:01 or a combination of neoantigen transfer plasmid, 0.75 μg psPAX2, and 0.25 μg pMD2.G with 200 μL Opti-MEM (Thermo Fischer Scientific Cat. #31985062). Add 6 μL Xtremegene HP to this mixture and allow complex formation to occur at room temperature for 15 minutes, at which point add the reaction mixture to the corresponding HEK293T cells. As described above, the neoantigen construct encodes a full-length or truncated version of the newly linked peptide. Replace the transfection medium with fresh DMEM after 24 hours. HLA-A testing is performed first before neoantigen lentiviral transduction and selection. 02:01 Lentiviral transduction and screening to simplify drug selection. Viral supernatant was collected 48 hours later, and functional viral titers were performed on 6-well plates inoculated with COS7 or K562 cells at 60-70% confluence. Viral transduction was performed using a 3-fold serially diluted viral stock solution supplemented with 4 μg / mL polybrene. The medium was replaced 24 hours after viral transduction with complete medium supplemented with blasticidin. Transduction efficiency was assessed 3-4 days later using drug screening. HLA-A was tested before assessing cell viability at each titer. The APCs transduced at 02:01 were cultured for approximately 7 days in medium treated with 10 μg / mL Blastidin. Neoantigen-lentivirus transduction was then performed, followed by HLA-A... APCs transduced with the 02:01 and neoantigen expression constructs were cultured for approximately 7 days in medium treated with 3 μg / mL puromycin. Cell viability was then assessed under all titer conditions. After 3–4 days, transduction efficiency was assessed by measuring fluorescence-activated cell sorting (FACS) of HLA-A2 surface expression.

[0280] Dose-dependent assessment of TCR responsiveness to neoantigens

[0281] The specificity of neoantigen-reactive CD8+ T cells and TCR-transduced T cells was assessed using human IFNγ (BD Biosciences Cat. #555142), IL-2 (BD Biosciences Cat. #555190), and TNFα ELISA (BD Biosciences Cat. #555212). TCR recognition of exogenously introduced neoantigen peptides presented by HLA molecules was assessed by co-culturing T cells with peptide-loaded T2 cells. The neoantigen peptide of interest, decoy peptide, or peptide-free T2 cells were cultured at 37°C for 1 hour at concentrations between 1 pM and 1 μM. Influenza-reactive T cells were co-cultured with influenza peptide-loaded T2 cells as a positive control. Cells were cultured at 1 x 10-1 cells per cell type. 5T cells and T2 cells were co-cultured in 200 μL of medium in 96-well round-bottom plates for 16 hours. The supernatant was collected and diluted for cytokine release assays according to the manufacturer's instructions. ELISA assays were performed using BioTek Gen5 data analysis software on an Epoch microplate spectrophotometer (input wavelength 450 nm, output wavelength 570 nm). To characterize dose-dependent activation of the TCR in transduced triple reporter Jurkat76 / CD8 cells, flow cytometry was performed to assess the expression levels of NFAT-GFP, NFκB-CFP, and AP-1-mCherry after 16 hours of co-culture. Similarly, the responsiveness of TCR-transduced PBMC-derived CD8+ T cells was evaluated by flow cytometry after staining with anti-CD107a (BioLegend, Cat#328620) and anti-CD137 antibody (4-1BB; Biolegend Cat#309804).

[0282] Synthesis of in vitro transcribed (IVT) mRNA

[0283] All constructs were subcloned into pcDNA3.1 (Invitrogen, 2520855) and linearized using the XhoI restriction enzyme, in which plasmid DNA templates were transcribed downstream of the phage T7 promoter sequence. 1 μg and 0.5 μg of template were used for long (>0.5 kb) and short (<0.5 kb) transcripts, respectively. The reaction was assembled at room temperature using the mMESSAGE mMACHINE T7 transcription kit (Invitrogen, 2582905) according to the manufacturer's instructions, and incubated at 37°C for 1 h for long transcripts and 16 h for short transcripts. Following DNase treatment, poly(A) tailing was performed for 1 h according to the HiScribe T7 ARCA manual (NEB, E2060S). Subsequently, the synthesized mRNA was purified by LiCl precipitation using 70% DEPC-based ethanol. The synthesized mRNA was heat-shocked (70°C, 5 min) with formaldehyde loading dye to verify quality by gel electrophoresis.

[0284] HLA-A 02:01. Transfection of truncated neoantigens and full-length neoligated mRNAs encoding mRNAs

[0285] According to the manufacturer's instructions, IVT-synthesized mRNA was transfected into COS7 and K562 cells via electroporation using a 100 μL Neon transfection system kit (Invitrogen, MPK10096). 100 μL of Neon... TM Washing and resuspending with resuspension buffer 1x10 6COS7 and K562 cells. 5 μg of HLA-A2 and 5 μg of candidate mRNA (truncated neoantigen sequence or full-length neolinked sequence) were added to the cell solution. Electroporation was performed on a Neon NxT electroporation system (Invitrogen, NEON1). Electroporation of COS7 cells was performed using the following optimized conditions: pulse voltage 1200 V, width 30 ms, and 2 pulses. Electroporation of K562 cells was performed using the following optimized conditions: pulse voltage 1450 V, width 10 ms, and 3 pulses. Transfected cells were immediately transferred to warm, antibiotic-free RPMI. Aliquots of transfected cells were retained for HLA-A2 expression validation by staining with an HLA-A2 monoclonal antibody (BB7.2, Thermo Scientific, 17-9876-42) and subsequent flow cytometry analysis.

[0286] Evaluation of TCR specificity against neoantigens processed endogenously and presented by HLA.

[0287] By HLA-A 02:01 / Neoantigen-transfected COS7 or K562 cells were co-cultured with TCR-transduced T cells for endogenous processing and characterization of neoantigens presented by surface HLA. Similarly, 1x10 cells were used for each cell type. 5 T cells and COS7 / K562 cells were co-cultured in 200 μL of medium in 96-well plates for 16 hours. Supernatants were collected and diluted according to the manufacturer's instructions for cytokine release assays, and cytokine release levels were assessed using an Epoch microplate spectrophotometer and BioTek Gen5 data analysis software. In all cytokine release assays, the maximum cytokine release per well was determined by adding 0.2 μL of cell activation mixture (without brevidin A) (BioLegendCat. #423302) to each 100 μL cell solution. Cytokine release was determined by an E:T ratio of 1:1 (1 x 10⁶ cells per well in a 96-well plate). 5 TCR-transduced triple reporter Jurkat76 cells were co-cultured with glioma cells to evaluate the endogenous processing and presentation of neoantigens in glioma cell lines. Flow cytometry analysis was performed to assess the expression levels of NFAT-GFP, NFκB-CFP, and AP-1-mCherry after 16 hours of co-culture.

[0288] HLA-IP and LC-MS / MS

[0289] 10 μg of HLA-A encoded DNA was transfected using the Neon transfection system (100 μL pipette tip, settings: 1,050 V / 10 ms / 2 pulses). 02:01 Each mRNA containing the allele and the full-length coding sequence of the mutated GNAS or RPL22 was co-electroplated into COS-7 cells. Electroporation was performed at 20 × 10⁻⁶ cells per condition. 6 Cells were plated in six-well non-TC plates and incubated overnight. For GMB115 cell line samples, approximately 100 × 10⁶ cells were used. 6Cells were harvested by incubation at 37°C with 1 mM EDTA (Millipore Sigma) for 10 min. For immunoprecipitation, cells were lysed at 4°C in 8 mL of 1% CHAPS (Millipore Sigma) for 1 h, followed by centrifugation of the lysate at 20,000 g for 1 h at 4°C, and collection of the supernatant. For affinity-based immunopurification of HLA-I ligands, 40 mg of cyanogen bromide-activated Sepharose 4B (Millipore Sigma) was activated with 1 mM hydrochloric acid (Millipore Sigma) for 30 min. Subsequently, 1 mg of W6 / 32 antibody (Bio X Cell) was conjugated to Sepharose in the presence of binding buffer (150 mM sodium chloride, 50 mM sodium bicarbonate, pH 8.3; sodium chloride) at room temperature for 2 h. Sepharose was blocked with glycine for 1 h and washed three times with PBS. The supernatant of cell lysates was run on an affinity column overnight at 4°C using a peristaltic pump at a flow rate of 6 mL / min. The HLA complex and binding peptide were eluted five times from the column using 1% TFA. The peptide and HLA-I complex were separated using a C18 column (Sep-Pak C18 1 cc Vac Cartridge, 50 mg adsorbent per column, 37–55 μm particle size, Waters). The C18 column was pretreated with 80% ACN (Millipore Sigma) in 0.1% TFA and equilibrated with two washes in 0.1% TFA. The sample was loaded, washed twice with 0.1% TFA, and eluted in 30%, 40%, and 50% acetonitrile in 0.1% TFA, 300 μL each time. All three fractions were combined, dried by vacuum centrifugation, and stored at -80°C until further processing. The HLA-I ligand was separated by solid-phase extraction using a homemade C18 microcolumn. Samples were analyzed using high-resolution / high-precision LC-MS / MS (Lumos Fusion, Thermo Fisher Scientific). COS-7 samples were run in DDA mode, while GMB115 samples were run in DIA mode. The operating resolutions for MS and MS / MS were 60,000 and 30,000, respectively. Only charge states 1, 2, and 3 were allowed. A 1.6 Thomson spectral density was selected as the isolation window, and the collision energy was set to 30%. For MS / MS, the maximum injection time was 100 ms, and the automatic gain control was 50,000. MS data were processed using FragPipe. The protein FDR was set to 1%. For all samples, methionine oxidation, phosphorylation of serine, threonine, and tyrosine, and N-terminal acetylation were set as variable modifications.Targets include UniProt Cercopithecus aethiops or UniProt human review protein (supplemented with human HLA-A). The samples were searched using databases containing allele sequences (02:01), mutRPL22 and mutGNAS, and common pollutants.

[0290] Characteristics of CD8+ T cell-mediated antitumor response

[0291] To determine whether TCR-transduced T cells could produce an antitumor response, TCR-transduced Jurkat76 / CD8 or PBMC-derived CD8+ T cells were co-cultured with patient-derived GBM or LGG cell lines. CD8+ T cells were isolated from healthy donor-derived PBMCs using the EasySep™ Human CD8+ T Cell Isolation Kit (STEMCELL Technologies, Cat. #17953). Cells were then isolated in 25 μL / 1 × 10⁻⁶ cells. 6 The concentration of each cell was used to activate CD8+ T cells with Dynabeads™ Human T-Activator CD3 / CD28 (Thermo Scientific, Cat. #11161D) for T cell expansion and activation. Supplementation with IL-7 (30 μL / 1 × 10⁻⁶ cells) was also employed. 6 CD8+ T cells were cultured in a medium containing 100 cells for 7 days, replenished every 2 days. The CD8+ T cells were then transduced using the neoantigen-specific TCR and a heterozygous mouse TCR constant region via the transduction procedure described above. This additional step eliminated the possibility of TCR α- and β-chain mismatches and allowed us to assess TCR transduction efficiency by staining with an anti-mouse TCR constant region antibody (clone H57-597; BioLegend Cat. #109208). Highly transduced CD8+ T cells were isolated by flow cytometry sorting using cells strongly stained with anti-CD3 and anti-mouse TCR constant region antibodies. The sorted transduced CD8+ T cells were expanded for 7 days prior to use in co-culture assays. Kill assays were performed using an xCELLigence RTCA S16 real-time cell analyzer. Tumor cells were cultured for 48 hours in a medium pretreated with 100 ng / mL IFNγ (Peprotech, Cat. #300-02) and washed twice with PBS before inoculation. 1×10 plates were laid in each well of a 96-well E-plate (Agilent). 4 Tumor cells were incubated, and impedance was read continuously for 16 hours during incubation. TCR-transduced CD8+ T cells were introduced into each well at an E:T ratio of 1:1 or 2:1, and tumor-specific killing was measured by changes in cell index over 24–48 hours.

[0292] Identification of HLA-restricted CD8+ T cell-mediated reactivity to neoantigens

[0293] HLA-restricted reactivity was evaluated by introducing anti-HLA antibodies to disrupt TCR and HLA:peptide interactions. In the dose-dependent immunogenicity assay, the concentration in each well of a 96-well plate was 1 × 10⁻⁶. 5 T2 cells containing tumor cells were washed twice with PBS and incubated for 30 min with either a blocking anti-HLA antibody (50 μg / well; clone W6 / 32, BioX Cell, Cat. #BE0079) or an allotype control (50 μg / well; BioX Cell, Cat. #BE0085) in a total volume of 100 μL. Without further washing, T cells were added to reach a final volume of 200 μL. For the tumor killing assay, tumor cells were added to each well of a 96-well E-plate for initial seeding in a total volume of 50 μL. Thirty minutes before adding T cells, either the anti-HLA antibody or the allotype control (50 μg / well) was added to each well to reach a total volume of 100 μL. T cells were added to each well to reach a final volume of 200 μL, and impedance was measured over the next 24–48 hours.

[0294] Immune surveillance in cancer patients expressing mutGNAS-NJ

[0295] Using a two-color HLA-A assay of the mutGNAS peptide identified by MS. 02 dextramer's FACS test showed HLA-A expression of GNAS NJ. 02. mutGNAS-specific CD8+ T cells are present in the circulation of cancer patients. Two weeks prior to FACS staining, HLA-A cells expressing NJ were used. 02. Matched moDCs were generated by stimulating patient CD8+ cells in vitro for 2 weeks. To generate moDCs, HLA-A... 02 Healthy donor PBMCs at 1×10 6 cells / cm 2Plates were placed in tissue culture flasks and incubated at 37°C for 2 hours in cytokine-free complete medium to separate adherent (monocyte-containing) and non-adherent (T cell-containing) fractions. Adherent fractions were washed with PBS, and fresh medium containing human A / B serum, supplemented with recombinant human IL-4 and GM-CSF (400 IU ml⁻¹), was provided every 3 days. On day 6, moDCs were matured for 24 hours prior to transfection using LPS (Invitrogen) and IFN-γ (Miltenyi Biotec). moDCs were electroporated using a Neon transfection system (10-μl tip, settings: 1,325 V / 10 ms / 3 pulses) with 100 μg ml⁻¹ of mRNA encoding full-length mutGNAS. Patient CD8+ cells were enriched from PBMCs using negative selection (STEMCELL Technologies) and, in tissue-free 24-well plates (FALCON), in the presence of 300 IU ml-1 IL-2 and 50 ng ml-1 IL-7, IL-15, and IL-21, were compared with HLA-A cells expressing mutGNAS-NJ in a 2:1 ratio. 02. Matched moDCs were co-cultured. Cytokines were supplemented every 3 days. As a control, HLA-A cells isolated and co-cultured similarly from healthy donors were used. 02:01 matched CD8+ cells. For the dextramer marker, HLA-A conjugates bound to mutGNAS and conjugated to PE or APC were purchased from Immudex. 02. Multimer. As a specificity control, HLA-A binding to the 9-mer peptide of P53 R175H (HMTEVVRHC) was used. 02. Multimers. Cells were labeled with dual-luciferase-conjugated dextramer for 15 minutes at room temperature, and then additionally labeled for 15 minutes at 4°C with surface antibodies against CD3-BV785, CD4-BV421, and CD8-BV650 (Biolegend). Cells were washed twice, stained with the viability dye 7-AAD (Biolegend), and harvested using a BD Fortessa X20 flow cytometer.

[0296] FACS analysis and antibodies

[0297] TCR-transduced cell lines were stained with anti-human TCRα / β (clone IP26, BioLegend Cat. #306717) and anti-human CD3 antibody (clone HIT3a, BioLegend Cat. #300307) to assess the surface expression level of transduced TCRs. CD8+ T cells were stained with anti-CD107a (BioLegend, Cat #328620) and anti-CD137 antibody (4-1BB; Biolegend Cat #309804) to assess CD8+ T cell degranulation and TCR activation, respectively. Zombie Green antibody was used. TM Cell viability was assessed using a fixation viability kit (BioLegend, Cat. #423111). APCs and patient-derived glioma cell lines were stained with an HLA-A2 monoclonal antibody (clone BB7.2, Thermo Fisher Scientific Cat. #17-9876-42). Approximately 1 × 10⁻⁶ cells were added per 100 μL of FACS buffer (PBS supplemented with 1% BSA (Sigma Aldrich Cat. #L6529)) as directed by the manufacturer. 6 One cell was incubated with one test volume of antibody for 20 minutes. The stained cells were washed once with FACS buffer and then resuspended to 4 × 10⁶ cells / 100 μL FACS buffer. 5 The concentration of cells was then determined. The cells were then analyzed using an Attune NxT flow cytometer (Thermo Fischer Scientific). Inaki, please add the patient's immune monitoring methods here.

[0298] Gene set enrichment analysis

[0299] Differential gene expression was performed and quantified using DESeq2 for TCGA, GTEx, and UCSF GBM / LGG RNA sequencing. Only genes with an absolute fold change >1.5 called by DESeq2 and a Benjamini-Hochberg adjusted p-value <0.05 were considered differentially expressed. Pre-ranked gene set enrichment analysis (GSEA) was performed by ranking genes by the product of their fold change sign and -log10 (adjusted p-value). Disease subtype-specific differential gene analysis:GSEA comparisons were performed between IDH1 mutant subtypes (wild-type IDH1 and mutant IDH1) and glioma disease subtypes (wild-type IDH1 glioblastoma, mutant IDH1 astrocytoma, and mutant IDH1 oligodendroglioma). Gene sets related to splicing were selected based on keyword searches, and gene sets with adjusted p-values ​​<0.05 when comparing two groups were considered differentially enriched. Unbiased hierarchical clustering of differentially enriched gene sets allowed for characterization of subgroup-specific upregulated genes. New connection load specificity difference base Analysis: TCGA LGG and GBM samples were ranked based on the putative total number of new connections expressed per sample. Within each disease subtype, high (NJ) levels were ranked. HI ) and low new connection load (NJ LO The samples were characterized as the upper and lower 0.10 percentiles of the ranked samples, respectively. In each disease subgroup, NJ... HI and NJ LO GSEA was performed between samples. Gene sets with unidirectional fold changes and adjusted p-values ​​<0.05 were considered enriched gene sets associated with neoconnection load. Gene sets associated with splicing were selected based on keyword searches. Leading marginal genes shared across all disease subgroups within the same gene set were defined as enriched genes associated with neoconnection load.

[0300] Correlation analysis of new connections and splicing-related genes

[0301] The selection of upregulated genes in mutant IDH1 cases was determined by a significant (p<0.05) log2-fold increase in expression of splicing-related genes compared to their wild-type counterparts. The selection of splicing genes affected by oligodendroglioma-specific chromosome 1p / 19q loss was determined by a significant (p<0.05) log2-fold decrease in expression of chromosome 1p / 19q-related genes in IDH-O cases compared to both IDH-A and IDH-wt cases. Splicing-related genes were selected for in vitro validation based on previously reported confirmation of aberrant splicing due to their dysregulation. Pearson correlation analysis was performed for each pair of novel links and splicing-related genes of interest to determine the correlation factor between each identified common neolink and each splicing-related gene of interest. The neolinks with the highest positive correlation scores against the selected mutant IDH1 upregulated genes (CELF2, ELAVL4) were averaged across all three glioma subtypes and tested using downstream qPCR assays. Similarly, the new linker with the most negative correlation score targeting selected chromosome 1p or 19q splicing-related genes downregulated in IDH1mut-O cases (SNRPD2, SF3A3) on average across all three glioma subtypes was also tested in a downstream qPCR assay.

[0302] AlphaFold2 structural prediction

[0303] AlphaFold v2.3.2 and its reference database were installed. AlphaFold was run in multi-merchant mode with default options, and the highest-ranked PDB file was visualized using PyMOL. Images were exported using the settings "ray 5000,5000" and "png image,dpi=2400".

[0304] Quantitative and statistical analysis

[0305] Perform all statistical analyses in basic R.

[0306] Method-specific references

[0307] 1. Vaubel, RA, Tian, ​​S., Remonde, D., Schroeder, MA, Mladek, AC, Kitange, GJ, Caron, A., Kollmeyer, TM, Grove, R., Peng, S., et al. (2020). Genomic and Phenotypic Characterization of a Broad Panel of Patient-Derived Xenografts Reflects the Diversity of Glioblastoma. Clin. Cancer Res.26, 1094–1104.

[0308] 2. Edwards, NJ, Oberti, M., Thangudu, RR, Cai, S., McGarvey, PB, Jacob, S., Madhavan, S., and Ketchum, KA (2015). The CPTAC Data Portal: A Resource for Cancer Proteomics Research. J. Proteome Res. 14, 2707–2713.

[0309] 3. Kahles, A., Lehmann, K.-V., Toussaint, N.C., Hüser, M., Stark, S.G., Sachsenberg, T., Stegle, O., Kohlbacher, O., Sander, C., Cancer Genome Atlas Research Network, et al. (2018). Comprehensive Analysis of Alternative Splicing Across Tumors from 8,705 Patients. Cancer Cell 34, 211-224.e6.

[0310] 4. Jones, L.E., Hilz, S., Grimmer, M.R., Mazor, T., Najac, C., Mukherjee, J., McKinney, A., Chow, T., Pieper, R.O., Ronen, S.M., et al. (2020). Patient-derived cells from recurrent tumors that model the evolution of IDH-mutant glioma. Neurooncol Adv 2, vdaa088.

[0311] 5. Stathias, V., Jermakowicz, A.M., Maloof, M.E., Forlin, M., Walters, W., Suter, R.K., Durante, M.A., Williams, S.L., Harbour, J.W., Volmar, C.-H., et al. (2018). Drug and disease signature integration identifies synergistic combinations in glioblastoma. Nat. Commun. 9, 5315.

[0312] 6. Jermakowicz, A.M., Rybin, M.J., Suter, R.K., Sarkaria, J.N., Zeier, Z., Feng, Y., and Ayad, N.G. (2021). The novel BET inhibitor UM-002 reduces glioblastoma cell proliferation and invasion. Sci. Rep. 11, 23370.

[0313] 7. Zhang, Y., Liu, Y., Moxley, K.M., Golden-Mason, L., Hughes, M.G., Liu, T., Heemskerk, M.H.M., Rosen, H.R., and Nishimura, M.I. (2010). Transduction of human T cells with a novel T-cell receptor confers anti-HCV reactivity. PLoS Pathog. 6, e1001018.

[0314] 8. Bossi, G., Gerry, A.B., Paston, S.J., Sutton, D.H., Hassan, N.J., and Jakobsen, B.K. (2013). Examining the presentation of tumor-associated antigens on peptide-pulsed T2 cells. Oncoimmunology 2, e26840.

[0315] 9. Hosken, N.A., and Bevan, M.J. (1990). Defective presentation of endogenous antigen by a cell line expressing class I molecules. Science 248, 367–370.

[0316] 10. Chandran, S.S., Ma, J., Klatt, M.G., Dündar, F., Bandlamudi, C., Razavi, P., Wen, H.Y., Weigelt, B., Zumbo, P., Fu, S.N., et al. (2022). Immunogenicity and therapeutic targeting of a public neoantigen derived from mutated PIK3CA. Nat. Med. 28, 946–957.

[0317] 11. Britten, C.M., Meyer, R.G., Kreer, T., Drexler, I., Wölfel, T., and Herr, W. (2002). The use of HLA-A 0201-transfected K562 as standard antigen-presenting cells for CD8(+) T lymphocytes in IFN-gamma ELISPOT assays. J. Immunol. Methods 259, 95–110.

[0318] 12. Suhoski, M.M., Golovina, T.N., Aqui, N.A., Tai, V.C., Varela-Rohena, A., Milone, M.C., Carroll, R.G., Riley, J.L., and June, C.H. (2007). Engineering artificial antigen-presenting cells to express a diverse array of co-stimulatory molecules. Mol. Ther. 15, 981–988.

[0319] 13. Love, M.I., Huber, W., and Anders, S. (2014). Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 550.

[0320] 14. Matsushima, S., Ajiro, M., Iida, K., Chamoto, K., Honjo, T., and Hagiwara, M. (2022). Chemical induction of splice-neoantigens attenuatestumor growth in a preclinical model of colorectal cancer. Sci. Transl. Med.14, eabn6

[0321] 15. Subramanian , A , Tamayo , P , Mootha , VK , Mukherjee , S , Ebert , BL , Gillette , MA , Paulovich , A , Pomeroy , SL , Golub , TR , Lander , ES , Other . (2005). Gene set enrichment analysis: a knowledge-based approachfor interpreting genome-wide expression profiles. Proc. Natl. Acad. Sci. U.S.A. 102 , 15545–15550.

[0322] 16. Pique, L., Martinez de Paz, A., Piñeyro, D., Martinez-Cardus, A.,Castro de Moura, M., Llinàs-Arias, P., Setien, F., Gomez-Miragaya, J.,Gonzalez-Suarez, E., Sigurdsson, S., Paper. (2019). Epigenetic inactivation of the splicing RNA-binding protein CELF2 in human breast cancer. Oncogene 38.7106–7112.

[0323] 17. Yanaizu, M., Washizu, C., Nukina, N., Satoh, J.-I., and Kino, Y. (2020). CELF2 regulates the species-specific alternative splicing of TREM2. Sci. Rep. 10, 17995.

[0324] 18. Bowles, K.R., Silva, M.C., Whitney, K., Bertucci, T., Berlind, J.E., Lai, J.D., Garza, J.C., Boles, N.C., Mahali, S., Strang, K.H., et al. (2021). ELAVL4, splicing, and glutamatergic dysfunction precede neuron loss in MAPT mutation cerebral organoids. Cell 184, 4547-4563.e17.

[0325] 19. Cie la, M., Ngoc, P.C.T., Cordero, E., Martinez, Á.S., Morsing, M., Muthukumar, S., Beneventi, G., Madej, M., Munita, R., Jönsson, T., et al. (2021). Oncogenic translation directs spliceosome dynamics revealing an integral role for SF3A3 in breast cancer. Mol. Cell 81, 1453-1468.e12.

[0326] 20. Wyatt, C.D.R., Pernaute, B., Gohr, A., Miret-Cuesta, M., Goyeneche, L., Rovira, Q., Salzer, M.C., Boke, E., Bogdanovic, O., Bonnal, S., et al. (2022). A developmentally programmed splicing failure contributes to DNA damage response attenuation during mammalian zygotic genome activation. Sci Adv 8, eabn4935.

[0327] References

[0328] 1. Weber, E.W., Maus, M.V., and Mackall, C.L. (2020). The Emerging Landscape of Immune Cell Therapies. Cell 181, 46–62.

[0329] 2. Touat, M., Li, Y.Y., Boynton, A.N., Spurr, L.F., Iorgulescu, J.B., Bohrson, C.L., Cortes-Ciriano, I., Birzu, C., Geduldig, J.E., Pelton, K., et al. (2020). Mechanisms and therapeutic implications of hypermutation in gliomas. Nature 580, 517–523.

[0330] 3. Russo, M., Crisafulli, G., Sogari, A., Reilly, NM, Arena, S.,Lamba, S., Bartolini, A., Amodio, V., Magrì, A., Novara, L., Others. (2019).Adaptive mutability of colorectal cancers in response to targeted therapies.Science 366, 1473–1480.

[0331] 4. Huang, Y., Jiang, X., Liang, X., and Jiang, G. (2018). Molecularand cellular mechanisms of castration resistant prostate cancer. Oncol. Lett.15, 6063–6076.

[0332] 5. Kim, C., Gao, R., Sei, E., Brandt, R., Hartman, J., Hatschek, T.,Crosetto, N., Foukakis, T., and Navin, NE (2018). Evolution of Chemoresistance in Triple-Negative Breast Cancer Delineated by Single-Cell Sequencing. Cell173, 879–893.e13.

[0333] 6. Topalian , SL , Hodi , FS , Brahmer , JR , Gettinger , SN , Smith , DC , McDermott , DF , Powderly , JD , Carvajal , RD , Sosman , JA , Atkins , MB , et al. (2012). Safety, Activity, and Immune Correlates of Anti–PD-1Antibody in Cancer. N. Engl. J. Med. Rev. 366 , 2443–2454.

[0334] 7. Motzer , RJ , Tannir , NM , McDermott , DF , Sand Border , O. ,Melichar , B. , Choueiri , TK , Plimack , ER , Barthélémy , P. , Porta , C. ,George , S. . (2018). Nivolumab plus Ipilimumab versus Sunitinib inAdvanced Renal-Cell Carcinoma. N. Engl. J. Med. Rev. 378 , 1277–1290.

[0335] 8. Le, DT, Uram, JN, Wang, H, Bartlett, BR, Kemberling, H,Eyring, AD, Skora, AD, Luber, BS, Azad, NS, Laheru, D, Private. (2015).PD-1 Blockade in Tumors with Mismatch-Repair Deficiency. N. Engl. J. Med.372, 2509–2520.

[0336] 9. Scholler, J., Brady, TL, Binder-Scholl, G., Hwang, W.-T., Plesa,G., Hege, KM, Vogel, AN, Kalos, M., Riley, JL, Deeks, SG. Decade-long safety and function of retroviral-modified chimeric antigen receptor T cells. Sci. Transl. Med. 4, 132ra5

[0337] 10. Kantoff, P.W., Higano, C.S., Shore, N.D., Berger, E.R., Small, E.J., Penson, D.F., Redfern, C.H., Ferrari, A.C., Dreicer, R., Sims, R.B., et al. (2010). Sipuleucel-T Immunotherapy for Castration-Resistant Prostate Cancer. N. Engl. J. Med. 363, 411–422.

[0338] 11. Ott, P.A., Hu, Z., Keskin, D.B., Shukla, S.A., Sun, J., Bozym, D.J., Zhang, W., Luoma, A., Giobbie-Hurder, A., Peter, L., et al. (2017). An immunogenic personal neoantigen vaccine for patients with melanoma. Nature 547, 217–221.

[0339] 12. Tirosh, I., Venteicher, A.S., Hebert, C., Escalante, L.E., Patel, A.P., Yizhak, K., Fisher, J.M., Rodman, C., Mount, C., Filbin, M.G., et al. (2016). Single-cell RNA-seq supports a developmental hierarchy in human oligodendroglioma. Nature 539, 309–313.

[0340] 13. Neftel, C., Laffy, J., Filbin, M.G., Hara, T., Shore, M.E., Rahme, G.J., Richman, A.R., Silverbush, D., Shaw, M.L., Hebert, C.M., et al. (2019). An Integrative Model of Cellular States, Plasticity, and Genetics for Glioblastoma. Cell 178, 835-849.e21.

[0341] 14. Nicholson, J.G., and Fine, H.A. (2021). Diffuse Glioma Heterogeneity and Its Therapeutic Implications. Cancer Discov. 11, 575–590.

[0342] 15. Teixeira, V.H., Pipinikas, C.P., Pennycuick, A., Lee-Six, H., Chandrasekharan, D., Beane, J., Morris, T.J., Karpathakis, A., Feber, A., Breeze, C.E., et al. (2019). Deciphering the genomic, epigenomic, and transcriptomic landscapes of pre-invasive lung cancer lesions. Nat. Med. 25, 517–525.

[0343] 16. van Galen , P , Hovestadt , V , Wadsworth , MH , Ii , Hughes , TK , Griffin , GK , Battaglia , S , Verga , JA , Stephansky , J , Pastika , TJ , Lombardi Story , J , Principal. (2019). Single-Cell RNA-Seq Reveals AML HierarchiesRelevant to Disease Progression and Immunity. Cell 176, 1265–1281.e24.

[0344] 17. Grosselin, K., A. Durand, J. Marsolier, A. Poitou, E. Marangoni, F. Nemati, A. Dahmani, S. Lameiras, F. Royal, O. Frenoy. High-throughput single-cell ChIP-seq identifies heterogeneity ofchromatin states in breast cancer. Nat. Genet. Rev. 51 , 1060–1066.

[0345] 18. Gerlinger , M , Rowan , AJ , Horswell , S , Larkin , J ,Endesfelder , D , Gronroos , E , Martinez , P , Matthews , N , Stewart , A , Tarpey , P , et al. (2012). Intratumor Heterogeneity and Branched EvolutionRevealed by Multiregion Sequencing. N. Engl. J. Med. Rev. 366 , 883–892.

[0346] 19. Yachida, S., Jones, S., Bozic, I., Antal, T., Leary, R., Fu, B.,Kamiyama, M., Hruban, RH, Eshleman, JR, Nowak, MA, et al. (2010).Distant metastasis occurs late during the genetic evolution of pancreaticcancer. Nature 467, 1114–1117.

[0347] 20. Sotillo, E., Barrett, DM, Black, KL, Bagashev, A., Oldridge, D., Wu, G., Sussman, R., Lanauze, C., Ruella, M., Gazzara, MR. Convergence of Acquired Mutations and Alternative Splicing of CD19 Enables Resistance to CART-19 Immunotherapy. Cancer Discov. 5, 1282–1295.

[0348] 21. Le , DT , Durham , JN , Smith , KN , Wang , H , Bartlett , BR ,Aulakh , LK , Lu , S , Kemberling , H , Wilt , C , Luber , BS , et al. (2017).Mismatch repair deficiency predicts response of solid tumors to PD-1blockade. Science 357, 409–413.

[0349] 22. Hellmann , MD , Nathanson , T , Rizvi , H , Creelan , BC , Sanchez-Vega , F , Ahuja , A , Ni , A , Novik , JB , Mangarin , LMB , Abu-Akeel , M . (2018). Genomic Features of Response to Combination Immunotherapy inPatients with Advanced Non-Small-Cell Lung Cancer. Cancer Cell 33, 843–852.e4.

[0350] 23. Samstein , RM , Lee , C.-H. , Shoushtari , AN , Hellmann , MD ,Shen , R. , Janjigian , YY , Barron , DA , Zehir , A. , Jordan , EJ , Omuro , A. ,person. (2019). Tumor mutational load predicts survival after immunotherapyacross multiple cancer types. Nat. Genet. 51 , 202–206 .

[0351] 24. Strønen, E., Toebes, M., Kelderman, S., van Buuren, MM, Yang,W., van Rooij, N., Donia, M., Böschen, M.-L., Lund-Johansen, F., Olweus, J.,et al. (2016). Targeting of cancer neoantigens with donor-derived T cell receptor repertoires. Science 352 , 1337–1341.

[0352] 25. Lawrence, M.S., Stojanov, P., Polak, P., Kryukov, G.V., Cibulskis, K., Sivachenko, A., Carter, S.L., Stewart, C., Mermel, C.H., Roberts, S.A., et al. (2013). Mutational heterogeneity in cancer and the search for new cancer-associated genes. Nature 499, 214–218.

[0353] 26. Klebanoff, C.A., Rosenberg, S.A., and Restifo, N.P. (2016). Prospects for gene-engineered T cell immunotherapy for solid cancers. Nat. Med. 22, 26–36.

[0354] 27. Kahles, A., Lehmann, K.-V., Toussaint, N.C., Hüser, M., Stark, S.G., Sachsenberg, T., Stegle, O., Kohlbacher, O., Sander, C., Cancer Genome Atlas Research Network, et al. (2018). Comprehensive Analysis of Alternative Splicing Across Tumors from 8,705 Patients. Cancer Cell 34, 211-224.e6.

[0355] 28. Tretter, C., de Andrade Krätzig, N., Pecoraro, M., Lange, S.,Seifert, P., von Frankenberg, C., Untch, J., Zuleger, G., Wilhelm, M., Zolg,DP. (2023). Proteogenomic analysis reveals RNA as a source for tumor-agnostic neoantigen identification. Nat. Commun. 14 , 4632 .

[0356] 29. Dvinge, H., and Bradley, RK (2015). Widespread intron retentiondiversifies most cancer transcriptomes. Genome Med. 7,

[0357] 30. Smart , AC , Margolis , CA , Pimentel , H. , He , MX , Miao , D. , Hunter , D. , Fugmann , T. , Wong , K.-K. Intronretention is a source of neoepitopes in cancer. Nat. Biotechnol. Rev. 36 , 1056–1058.

[0358] 31. Bigot , J. , Lalanne , AI , Lucibello , F. , Gueguen , P. , Houy , A. , Dayot , S. , Ganier , O. , Gilet , J. , Tosello , J. , Nemati , F. , Person. (2021).Splicing Patterns in SF3B1-Mutated Uveal Melanoma Generate Shared ImmunogenicTumor-Specific Neoepitopes. Cancer Discov. 11, 1938–1951.

[0359] 32. Merlotti, A., Sadacca, B., Arribas, YA, Ngoma, M., Burbage, M.,Goudot, C., Houy, A., Rocañin-Arjó, A., Lalanne, A., Seguin-Givelet, A., et al. (2023). Noncanonical splicing junctions between exons and transposableelements represent a source of immunogenic recurrent neo-antigens in patientswith lung cancer. Immun Sci 8, eabm6359.

[0360] 33. Burbage, M., Rocañin-Arjó, A., Baudon, B., Arribas, YA,Merlotti, A., Rookhuizen, DC, Heurtebise-Chrétien, S., Ye, M., Houy, A.,Burgdorf, N., Principal. (2023). Epigenetically controlled tumor antigens derivedfrom splice junctions between exons and transposable elements. Sci Immunol 8,eabm6360.

[0361] (2023) 34. Pan , Y. , Phillips , JW , Zhang , BD , Noguchi , M. , Kutschera , E. , McLaughlin , J. , Nesterenko , PA , Mao , Z. , Bangayan , NJ , Wang , R. . IRIS: Discovery of cancer immunotherapy targets arising from pre-mRNAalternative splicing. Proc. Natl. Acad. Sci. U.S.A. 120, e2221116120.

[0362] 35. Kim, G.B., Fritsche, J., Bunk, S., Mahr, A., Unverdorben, F., Tosh, K., Kong, H., Maldini, C.R., Lau, C., Srivatsa, S., et al. (2022). Quantitative immunopeptidomics reveals a tumor stroma-specific target for T cell therapy. Sci. Transl. Med. 14, eabo6135.

[0363] 36. Aran, D., Sirota, M., and Butte, A.J. (2015). Systematic pan-cancer analysis of tumour purity. Nat. Commun. 6, 8971.

[0364] 37. Ceccarelli, M., Barthel, F.P., Malta, T.M., Sabedot, T.S., Salama, S.R., Murray, B.A., Morozova, O., Newton, Y., Radenbaugh, A., Pagnotta, S.M., et al. (2016). Molecular Profiling Reveals Biologically Discrete Subsets and Pathways of Progression in Diffuse Glioma. Cell 164, 550–563.

[0365] 38. Nejo, T., Wang, L., Leung, K.K., Wang, A., Lakshmanachetty, S., Gallus, M., Kwok, D.W., Hong, C., Chen, L.H., Carrera, D.A., et al. (2024). Challenges in the discovery of tumor-specific alternative splicing-derived cell-surface antigens in glioma. Sci. Rep. 14, 6362.

[0366] 39. Quesnel-Vallières, M., Jewell, S., Lynch, K.W., Thomas-Tikhonenko, A., and Barash, Y. (2024). MAJIQlopedia: an encyclopedia of RNA splicing variations in human tissues and cancer. Nucleic Acids Res. 52, D213–D221.

[0367] 40. Yamamoto, T.N., Kishton, R.J., and Restifo, N.P. (2019). Developing neoantigen-targeted T cell-based treatments for solid tumors. Nat. Med. 25, 1488–1499.

[0368] 41. Ku, A.T., Wilkinson, S., and Sowalsky, A.G. (2021). Comparison of approaches to transcriptomic analysis in multi-sampled tumors. Brief. Bioinform. 22, bbab337.

[0369] 42. Joung, J.-G., Bae, J.S., Kim, S.C., Jung, H., Park, W.-Y., and Song, S.-Y. (2016). Genomic Characterization and Comparison of Multi-Regional and Pooled Tumor Biopsy Specimens. PLoS One 11, e0152574.

[0370] 43. Yang, C., Zhang, S., Cheng, Z., Liu, Z., Zhang, L., Jiang, K., Geng, H., Qian, R., Wang, J., Huang, X., et al. (2022). Multi-region sequencing with spatial information enables accurate heterogeneity estimation and risk stratification in liver cancer. Genome Med. 14, 142.

[0371] 44. Jeon, A.-J., Teo, Y.-Y., Sekar, K., Chong, S.L., Wu, L., Chew, S.-C., Chen, J., Kendarsari, R.I., Lai, H., Ling, W.H., et al. (2023). Multi-region sampling with paired sample sequencing analyses reveals sub-groups of patients with novel patient-specific dysregulation in Hepatocellular Carcinoma. BMC Cancer 23, 118.

[0372] 45. Zhai, W., Lai, H., Kaya, N.A., Chen, J., Yang, H., Lu, B., Lim, J.Q., Ma, S., Chew, S.C., Chua, K.P., et al. (2021). Dynamic phenotypic heterogeneity and the evolution of multiple RNA subtypes in hepatocellular carcinoma: the PLANET study. Natl Sci Rev 9, nwab192.

[0373] 46. ​​Arnadottir , SS , Jeppesen , M , Lamy , P , Bramsen , JB , Nordentoft , I , Knudsen , M , Vang , S , Madsen , MR , Thastrup , O , Thastrup , J . (2018). Characterization of genetic intratumor heterogeneity in colorectal cancer and matching patient-derived spheroid cultures. Mol. Oncol.12, 132–147.

[0374] 47. Biswas, D., Birkbak, NJ, Rosenthal, R., Hiley, CT, Lim, EL,Papp, K., Boeing, S., Krzystanek, M., Djureinovic, D., La Fleur, L., Opinion. A clonal expression biomarker associates with lung cancer mortality.Nat. Med. Rev. 25, 1540–1548.

[0375] 48. Meiller, C., Montagne, F., Hirsch, TZ, Caruso, S., de Wolf, J.,Bayard, Q., Assié, J.-B., Meunier, L., Blum, Y., Quetel, L., et al. (2021).Multi-site tumor sampling highlights molecular intra-tumor heterogeneity in malignant pleural mesothelioma. Genome Med. 13 ,

[0376] 49. Bhaduri, A., Di Lullo, E., Jung, D., Müller, S., Crouch, EE,Espinosa, CS, Ozawa, T., Alvarado, B., Spatazza, J., Cadwell, CR. Outer Radial Glia-like Cancer Stem Cells Contribute to Heterogeneityof Glioblastoma. Cell Stem Cell 26, 48–63.e6.

[0377] 50. Venteicher, AS, Tirosh, I, Hebert, C, Yizhak, K, Neftel, C,Filbin, MG, Hovestadt, V, Escalante, LE, Shaw, ML, Rodman, C. (2017). Decoupling genetics, lineages, and microenvironment in IDH-mutant gliomas by single-cell RNA-seq. Science 355. 10.1126 / science.aai8478.

[0378] 51. Brennan , CW , Verhaak , RGW , McKenna , A , Campos , B , Noushmehr , H , Salama , SR , Zheng , S , Chakravarty , D , Sanborn , JZ , Berman , SH , et al. (2013). The somatic genomic landscape of glioblastoma.Cell 155, 462–477.

[0379] 52. McKinney, A.M., Mathur, R., Stevers, N.O., Molinaro, A.M., Chang, S.M., Phillips, J.J., and Costello, J.F. (2022). GABP couples oncogene signaling to telomere regulation in TERT promoter mutant cancer. Cell Rep. 40, 111344.

[0380] 53. Appin, C.L., Hong, C., Suwala, A.K., Hilz, S., Mathur, R., Solomon, D.A., Smirnov, I.V., Stevers, N.O., Shai, A., Wang, A., et al. (2024). Whole tumor analysis reveals early origin of the TERT promoter mutation and intercellular heterogeneity in TERT expression. Neuro. Oncol. 26, 640–652.

[0381] 54. Mathur, R., Wang, Q., Schupp, P.G., Nikolic, A., Hilz, S., Hong, C., Grishanina, N.R., Kwok, D., Stevers, N.O., Jin, Q., et al. (2024). Glioblastoma evolution and heterogeneity from a 3D whole-tumor perspective. Cell 187, 446-463.e16.

[0382] 55. Spain, L., Coulton, A., Lobon, I., Rowan, A., Schnidrig, D., Shepherd, S.T.C., Shum, B., Byrne, F., Goicoechea, M., Piperni, E., et al. (2023). Late-Stage Metastatic Melanoma Emerges through a Diversity of Evolutionary Pathways. Cancer Discov. 13, 1364–1385.

[0383] 56. Van Nostrand, E.L., Freese, P., Pratt, G.A., Wang, X., Wei, X., Xiao, R., Blue, S.M., Chen, J.-Y., Cody, N.A.L., Dominguez, D., et al. (2020). A large-scale binding and functional map of human RNA-binding proteins. Nature 583, 711–719.

[0384] 57. Sveen, A., Kilpinen, S., Ruusulehto, A., Lothe, R.A., and Skotheim, R.I. (2016). Aberrant RNA splicing in cancer; expression changes and driver mutations of splicing factor genes. Oncogene 35, 2413–2427.

[0385] 58. Seiler, M., Peng, S., Agrawal, A.A., Palacino, J., Teng, T., Zhu, P., Smith, P.G., Cancer Genome Atlas Research Network, Buonamici, S., and Yu, L. (2018). Somatic Mutational Landscape of Splicing Factor Genes and Their Functional Consequences across 33 Cancer Types. Cell Rep. 23, 282-296.e4.

[0386] 59. PCAWG Transcriptome Core Group, Calabrese, C., Davidson, N.R., Demircio lu, D., Fonseca, N.A., He, Y., Kahles, A., Lehmann, K.-V., Liu, F., Shiraishi, Y., et al. (2020). Genomic basis for RNA alterations in cancer. Nature 578, 129–136.

[0387] 60. Bettegowda, C., Agrawal, N., Jiao, Y., Sausen, M., Wood, L.D., Hruban, R.H., Rodriguez, F.J., Cahill, D.P., McLendon, R., Riggins, G., et al. (2011). Mutations in CIC and FUBP1 contribute to human oligodendroglioma. Science 333, 1453–1455.

[0388] 61. Chan, A.K.-Y., Pang, J.C.-S., Chung, N.Y.-F., Li, K.K.-W., Poon, W.S., Chan, D.T.-M., Shi, Z., Chen, L., Zhou, L., and Ng, H.-K. (2014). Loss of CIC and FUBP1 expressions are potential markers of shorter time to recurrence in oligodendroglial tumors. Mod. Pathol. 27, 332–342.

[0389] 62. Ajiro, M., Awaya, T., Kim, Y.J., Iida, K., Denawa, M., Tanaka, N., Kurosawa, R., Matsushima, S., Shibata, S., Sakamoto, T., et al. (2021). Therapeutic manipulation of IKBKAP mis-splicing with a small molecule to cure familial dysautonomia. Nat. Commun. 12, 4507.

[0390] 63. Piqué, L., Martinez de Paz, A., Piñeyro, D., Martínez-Cardús, A., Castro de Moura, M., Llinàs-Arias, P., Setien, F., Gomez-Miragaya, J., Gonzalez-Suarez, E., Sigurdsson, S., et al. (2019). Epigenetic inactivation of the splicing RNA-binding protein CELF2 in human breast cancer. Oncogene 38, 7106–7112.

[0391] [ PubMed ] 64. Yanaizu, M., Washizu, C., Nukina, N., Satoh, J.-I., and Kino, Y. (2020). CELF2 regulates species-specific alternative splicing of TREM2.Sci. Rep. 10 , 17995 .

[0392] 65. Jones , LE , Hilz , S. , Grimmer , MR , Mazor , T. , Najac , C. , Mukherjee , J. , McKinney , A. , Chow , T. , Pieper , RO , Ronen , SM . Patient-derived cells from recurrent tumors that model the evolutionof IDH-mutant glioma. Neuroncol Adv 2, vdaa088.

[0393] 66. Wyatt, CDR, Pernaute, B., Gohr, A., Miret-Cuesta, M.,Goyeneche, L., Rovira, Q., Salzer, MC, Boke, E., Bogdanovic, O., Bonnal,S. (2022). A developmentally programmed splicing failure contributesto the attenuation of the DNA damage response during mammalian zygotic genomeactivation. Sci Adv 8 , eabn4935 .

[0394] 67. Furney, S.J., Pedersen, M., Gentien, D., Dumont, A.G., Rapinat, A., Desjardins, L., Turajlic, S., Piperno-Neumann, S., de la Grange, P., Roman-Roman, S., et al. (2013). SF3B1 mutations are associated with alternative splicing in uveal melanoma. Cancer Discov. 3, 1122–1129.

[0395] 68. Vaubel, R.A., Tian, S., Remonde, D., Schroeder, M.A., Mladek, A.C., Kitange, G.J., Caron, A., Kollmeyer, T.M., Grove, R., Peng, S., et al. (2020). Genomic and Phenotypic Characterization of a Broad Panel of Patient-Derived Xenografts Reflects the Diversity of Glioblastoma. Clin. Cancer Res. 26, 1094–1104.

[0396] 69. Edwards, N.J., Oberti, M., Thangudu, R.R., Cai, S., McGarvey, P.B., Jacob, S., Madhavan, S., and Ketchum, K.A. (2015). The CPTAC Data Portal: A Resource for Cancer Proteomics Research. J. Proteome Res. 14, 2707–2713.

[0397] 70. Bader, J.M., Deigendesch, N., Misch, M., Mann, M., Koch, A., and Meissner, F. (2023). Proteomics separates adult-type diffuse high-grade gliomas in metabolic subgroups independent of 1p / 19q codeletion and across IDH mutational status. Cell Rep Med 4, 100877.

[0398] 71. Wong, D., Lee, T.H., Lum, A., Tao, V.L., and Yip, S. (2022). Integrated proteomic analysis of low-grade gliomas reveals contributions of 1p-19q co-deletion to oligodendroglioma. Acta Neuropathol Commun 10, 70.

[0399] 72. Blass, E., and Ott, P.A. (2021). Advances in the development of personalized neoantigen-based therapeutic cancer vaccines. Nat. Rev. Clin. Oncol. 18, 215–229.

[0400] 73. O’Donnell, T.J., Rubinsteyn, A., and Laserson, U. (2020). MHCflurry2.0: Improved Pan-Allele Prediction of MHC Class I-Presented Peptides by Incorporating Antigen Processing. Cell Systems 11, 42-48.e7.

[0401] 74. Sarkizova , S. , Klaeger , S. , Le , PM , Li , LW , Oliveira , G. , Keshishian , H. , Hartigan , CR , Zhang , W. , Braun , DA , Ligon , KL . A large peptidome dataset improves HLA class I epitope predictionacross most of the human population. Nat. Biotechnol. 38 , 199–209 .

[0402] 75. Gonzalez-Galarza , FF , McCabe , A , Santos , EJMD , Jones , J , Takeshita , L , Ortega-Rivera , ND , Cid-Pavon , GMD , Ramsbottom , K , Ghattaoraya , G , Alfirevic , A , Others . (2020). Allele Frequency Net Database (AFND) 2020 Update: Gold-Standard Data Classification, Open Access Genotype Data and New Query Tools. Nucleic Acids Res. 48, D783–D788.

[0403] 76. Arrieta-Bolaños, E., Hernandez-Zaragoza, DI, and Barquera, R. (2023). An HLA map of the world: A comparison of HLA frequencies in 200worldwide populations reveals diverse patterns for class I and class II.Front. Genet. 14 , 866407 .

[0404] 77. Olivier, T., Haslam, A., Tuia, J., and Prasad, V. (2023). Eligibility for Human Leukocyte Antigen-Based Therapeutics by Race and Ethnicity. JAMA Netw Open 6, e2338612.

[0405] 78. Lo, W., Parkhurst, M., Robbins, P.F., Tran, E., Lu, Y.-C., Jia, L., Gartner, J.J., Pasetto, A., Deniger, D., Malekzadeh, P., et al. (2019). Immunologic Recognition of a Shared p53 Mutated Neoantigen in a Patient with Metastatic Colorectal Cancer. Cancer Immunol Res 7, 534–543.

[0406] 79. Hsiue, E.H.-C., Wright, K.M., Douglass, J., Hwang, M.S., Mog, B.J., Pearlman, A.H., Paul, S., DiNapoli, S.R., Konig, M.F., Wang, Q., et al. (2021). Targeting a neoantigen derived from a common TP53 mutation. Science 371. 10.1126 / science.abc8697.

[0407] 80. Carreno, B.M., Magrini, V., Becker-Hapak, M., Kaabinejadian, S., Hundal, J., Petti, A.A., Ly, A., Lie, W.-R., Hildebrand, W.H., Mardis, E.R., et al. (2015). Cancer immunotherapy. A dendritic cell vaccine increases the breadth and diversity of melanoma neoantigen-specific T cells. Science 348, 803–808.

[0408] 81. Chen, F., Zou, Z., Du, J., Su, S., Shao, J., Meng, F., Yang, J., Xu, Q., Ding, N., Yang, Y., et al. (2019). Neoantigen identification strategies enable personalized immunotherapy in refractory solid tumors. J. Clin. Invest. 129, 2056–2070.

[0409] 82. Ali, M., Foldvari, Z., Giannakopoulou, E., Böschen, M.-L., Strønen, E., Yang, W., Toebes, M., Schubert, B., Kohlbacher, O., Schumacher, T.N., et al. (2019). Induction of neoantigen-reactive T cells from healthy donors. Nat. Protoc. 14, 1926–1943.

[0410] 83. Chandran, S.S., Ma, J., Klatt, M.G., Dündar, F., Bandlamudi, C., Razavi, P., Wen, H.Y., Weigelt, B., Zumbo, P., Fu, S.N., et al. (2022). Immunogenicity and therapeutic targeting of a public neoantigen derived from mutated PIK3CA. Nat. Med. 28, 946–957.

[0411] 84. Bossi, G., Gerry, A.B., Paston, S.J., Sutton, D.H., Hassan, N.J., and Jakobsen, B.K. (2013). Examining the presentation of tumor-associated antigens on peptide-pulsed T2 cells. Oncoimmunology 2, e26840.

[0412] 85. Jutz, S., Leitner, J., Schmetterer, K., Doel-Perez, I., Majdic, O., Grabmeier-Pfistershammer, K., Paster, W., Huppa, J.B., and Steinberger, P. (2016). Assessment of costimulation and coinhibition in a triple parameter T cell reporter line: Simultaneous measurement of NF-κB, NFAT and AP-1. J. Immunol. Methods 430, 10–20.

[0413] 86. Rosskopf, S., Leitner, J., Paster, W., Morton, L.T., Hagedoorn, R.S., Steinberger, P., and Heemskerk, M.H.M. (2018). A Jurkat 76 based triple - parameter reporter system to evaluate TCR functions and adoptive T cell strategies. Oncotarget 9, 17608–17619.

[0414] 87. Weiss, G.A., Watanabe, C.K., Zhong, A., Goddard, A., and Sidhu, S.S. (2000). Rapid mapping of protein functional epitopes by combinatorial alanine scanning. Proc. Natl. Acad. Sci. U. S. A. 97, 8950–8954.

[0415] 88. Sanderson, J.P., Crowley, D.J., Wiedermann, G.E., Quinn, L.L., Crossland, K.L., Tunbridge, H.M., Cornforth, T.V., Barnes, C.S., Ahmed, T., Howe, K., et al. (2020). Preclinical evaluation of an affinity - enhanced MAGE - A4 - specific T - cell receptor for adoptive T - cell therapy. Oncoimmunology 9, 1682381.

[0416] 89. Mardis, E.R., Ding, L., Dooling, D.J., Larson, D.E., McLellan, M.D., Chen, K., Koboldt, D.C., Fulton, R.S., Delehaunty, K.D., McGrath, S.D., et al. (2009). Recurring mutations found by sequencing an acute myeloid leukemia genome. N. Engl. J. Med. 361, 1058–1066.

[0417] 90. Cancer Genome Atlas Research Network, Ley, T.J., Miller, C., Ding, L., Raphael, B.J., Mungall, A.J., Robertson, A.G., Hoadley, K., Triche, T.J., Jr, Laird, P.W., et al. (2013). Genomic and epigenomic landscapes of adult de novo acute myeloid leukemia. N. Engl. J. Med. 368, 2059–2074.

[0418] 91. Marcucci, G., Maharry, K., Wu, Y.-Z., Radmacher, M.D., Mrózek, K., Margeson, D., Holland, K.B., Whitman, S.P., Becker, H., Schwind, S., et al. (2010). IDH1 and IDH2 gene mutations identify novel molecular subsets within de novo cytogenetically normal acute myeloid leukemia: a Cancer and Leukemia Group B study. J. Clin. Oncol. 28, 2348–2355.

[0419] 92. Borger, D.R., Tanabe, K.K., Fan, K.C., Lopez, H.U., Fantin, V.R., Straley, K.S., Schenkein, D.P., Hezel, A.F., Ancukiewicz, M., Liebman, H.M., et al. (2012). Frequent mutation of isocitrate dehydrogenase (IDH)1 and IDH2 in cholangiocarcinoma identified through broad-based tumor genotyping. Oncologist 17, 72–79.

[0420] 93. Boscoe, A.N., Rolland, C., and Kelley, R.K. (2019). Frequency and prognostic significance of isocitrate dehydrogenase 1 mutations in cholangiocarcinoma: a systematic literature review. J. Gastrointest. Oncol. 10, 751–765.

[0421] 94. Amary, M.F., Bacsi, K., Maggiani, F., Damato, S., Halai, D., Berisha, F., Pollock, R., O’Donnell, P., Grigoriadis, A., Diss, T., et al. (2011). IDH1 and IDH2 mutations are frequent events in central chondrosarcoma and central and periosteal chondromas but not in other mesenchymal tumours. J. Pathol. 224, 334–343.

[0422] 95. Glöss, S., Jurmeister, P., Thieme, A., Schmid, S., Cai, W.Y., Serrette, R.N., Perner, S., Ribbat-Idel, J., Pagenstecher, A., Bläker, H., et al. (2021). IDH2 R172 Mutations Across Poorly Differentiated Sinonasal Tract Malignancies: Forty Molecularly Homogenous and Histologically Variable Cases With Favorable Outcome. Am. J. Surg. Pathol. 45, 1190–1204.

[0423] 96. Guilmette, J., and Sadow, P.M. (2019). High-Grade Sinonasal Carcinoma: Classification Through Molecular Profiling. Arch. Pathol. Lab. Med. 143, 1416–1419.

[0424] 97. Mito, J.K., Bishop, J.A., Sadow, P.M., Stelow, E.B., Faquin, W.C., Mills, S.E., Krane, J.F., French, C.A., Fletcher, C.D.M., Hornick, J.L., et al. (2018). Immunohistochemical Detection and Molecular Characterization of IDH-mutant Sinonasal Undifferentiated Carcinomas. Am. J. Surg. Pathol. 42, 1067–1075.

[0425] 98. Cairns, RA, Iqbal, J., Lemonnier, F., Kucuk, C., de Leval, L.,Jais, J.-P., Parrens, M., Martin, A., Xerri, L., Brousset, P., et al. (2012).IDH2 mutations are frequent in angioimmunoblastic T-cell lymphoma. Blood 119,1901–1903.

[0426] 99. Wang , C , McKeithan , TW , Gong , Q , Zhang , W , Bouska , A ,Rosenwald , A , Gascoyne , RD , Wu , X , Wang , J , Muhammad , Z . (2015).IDH2R172 mutations define a unique subgroup of patients withangioimmunoblastic T-cell lymphoma. Blood 126, 1741–1752.

[0427] [ PubMed ] 100. Schumacher, T., Bunse, L., Pusch, S., Sahm, F., Wiestler, B.,Quandt, J., Menn, O., Osswald, M., Oezen, I., Ott, M., Ott. (2014). IDH1 mutant targeting vaccine induces antitumor immunity. Nature 512, 324–327.

[0428] 101. Grassl, N., Poschke, I., Lindner, K., Bunse, L., Mildenberger, I., Boschert, T., Jähne, K., Green, E.W., Hülsmeyer, I., Jünger, S., et al. (2023). A H3K27M-targeted vaccine in adults with diffuse midline glioma. Nat. Med. 10.1038 / s41591-023-02555-6.

[0429] 102. Tran, E., Turcotte, S., Gros, A., Robbins, P.F., Lu, Y.-C., Dudley, M.E., Wunderlich, J.R., Somerville, R.P., Hogan, K., Hinrichs, C.S., et al. (2014). Cancer immunotherapy based on mutation-specific CD4+ T cells in a patient with epithelial cancer. Science 344, 641–645.

[0430] 103. Linnemann, C., van Buuren, M.M., Bies, L., Verdegaal, E.M.E., Schotte, R., Calis, J.J.A., Behjati, S., Velds, A., Hilkmann, H., Atmioui, D.E., et al. (2015). High-throughput epitope discovery reveals frequent recognition of neo-antigens by CD4+ T cells in human melanoma. Nat. Med. 21, 81–85.

[0431] 104. Stathias, V., Jermakowicz, A.M., Maloof, M.E., Forlin, M., Walters, W., Suter, R.K., Durante, M.A., Williams, S.L., Harbour, J.W., Volmar, C.-H., et al. (2018). Drug and disease signature integration identifies synergistic combinations in glioblastoma. Nat. Commun. 9, 5315.

[0432] 105. Jermakowicz, A.M., Rybin, M.J., Suter, R.K., Sarkaria, J.N., Zeier, Z., Feng, Y., and Ayad, N.G. (2021). The novel BET inhibitor UM-002 reduces glioblastoma cell proliferation and invasion. Sci. Rep. 11, 23370.

[0433] 106. Zhang, Y., Liu, Y., Moxley, K.M., Golden-Mason, L., Hughes, M.G., Liu, T., Heemskerk, M.H.M., Rosen, H.R., and Nishimura, M.I. (2010). Transduction of human T cells with a novel T-cell receptor confers anti-HCV reactivity. PLoS Pathog. 6, e1001018.

[0434] 107. Hosken, N.A., and Bevan, M.J. (1990). Defective presentation of endogenous antigen by a cell line expressing class I molecules. Science 248, 367–370.

[0435] 108. Britten, C.M., Meyer, R.G., Kreer, T., Drexler, I., Wölfel, T., and Herr, W. (2002). The use of HLA-A 0201-transfected K562 as standard antigen-presenting cells for CD8(+) T lymphocytes in IFN-gamma ELISPOT assays. J. Immunol. Methods 259, 95–110.

[0436] 109. Suhoski, M.M., Golovina, T.N., Aqui, N.A., Tai, V.C., Varela-Rohena, A., Milone, M.C., Carroll, R.G., Riley, J.L., and June, C.H. (2007). Engineering artificial antigen-presenting cells to express a diverse array of co-stimulatory molecules. Mol. Ther. 15, 981–988.

[0437] 110. Love, M.I., Huber, W., and Anders, S. (2014). Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 550.

[0438] 111. Matsushima, S., Ajiro, M., Iida, K., Chamoto, K., Honjo, T., and Hagiwara, M. (2022). Chemical induction of splice-neoantigens attenuates tumor growth in a preclinical model of colorectal cancer. Sci. Transl. Med. 14, eabn6056.

[0439] 112. Subramanian , A , Tamayo , P , Mootha , VK , Mukherjee , S , Ebert , BL , Gillette , MA , Paulovich , A , Pomeroy , SL , Golub , TR , Lander , ES . (2005). Gene set enrichment analysis: a knowledge-based approachfor interpreting genome-wide expression profiles. Proc. Natl. Acad. Sci. U.S.A. 102 , 15545–15550.

[0440] (2021) Bowles, KR, Silva, MC, Whitney, K, Bertucci, T, Berlind,JE, Lai, JD, Garza, JC, Boles, NC, Mahali, S, Strang, KH. ELAVL4, splicing, and glutamatergic dysfunction precede neuron loss in MAPT mutation brain organoids. Cell 184, 4547–4563.e17.

[0441] 114. Cie la, M., Ngoc, PCT, Lamb, E., Martinez, Á.S., Morsing,M., Muthukumar, S., Beneventi, G., Madej, M., Munita, R., Jönsson, T., Private.(2021). Oncogenic translation directs spliceosome dynamics revealing an integral role for SF3A3 in breast cancer. Mol. Cell 81, 1453–1468.e12.

[0442] Although the invention has been described with reference to specific embodiments thereof, those skilled in the art will understand that various changes and substitutions may be made without departing from the true spirit and scope of the invention. Furthermore, many modifications may be made to adapt to particular circumstances, materials, compositions, processes, process steps, or steps to be consistent with the object, spirit, and scope of the invention. All such modifications are intended to be within the scope of the claims appended herein.

Claims

1. An isolated peptide comprising the amino acid sequence of SEQ ID NO: 1 or 2.

2. The peptide of claim 1, wherein the peptide has a length of 9 to 40 amino acids.

3. The isolated peptide of claim 1, wherein the peptide is composed of SEQ ID NO: 1, 2 or 3.

4. A composition comprising: The peptide or the RNA or DNA encoding it according to any one of claims 1-3; and Pharmaceutically acceptable excipients.

5. A composition comprising a major histocompatibility complex (MHC)-peptide complex, said complex comprising MHC and the peptide of any one of claims 1-3.

6. A cell expressing the composition of claim 5 on its surface.

7. The cell of claim 6, wherein the cell is an antigen-presenting cell.

8. The cell of claim 7, wherein the cell is a dendritic cell.

9. A method for sensitizing T cells, comprising: To bring the cells of any one of claims 7 or 8 into contact with the T cell population; and T cells are incubated together with the cells under conditions that activate and / or expand T cells that recognize MHC-peptide complexes.

10. The method of claim 9, wherein the cells are brought into contact in vivo.

11. The method of claim 9, wherein the cells are contacted in vitro or ex vivo.

12. A method for inducing an immune response against cancer cell antigens, the method comprising: Administer to an individual an effective dose of any of claims 1-8, the peptide, composition, or cell or RNA or DNA encoding the peptide.

13. T cell receptors, which include: α-chains having CDR sequences of SEQ ID NO: 5, 6 and 7; and β-chain having CDR sequences of SEQ ID NO: 9, 10 and 11; or α-chains having CDR sequences of SEQ ID NO: 13, 14 and 15; and β chain having CDR sequences of SEQ ID NO: 17, 18 and 28.

14. The T-cell receptor of claim 13, comprising: An α-chain having at least 90% of the same amino acid sequence as SEQ ID NO: 4; and A β-chain having at least 90% of the same amino acid sequence as SEQ ID NO: 8; or An α-chain having at least 90% of the same amino acid sequence as SEQ ID NO: 12; and A β chain having at least 90% of the same amino acid sequence as SEQ ID NO:

16.

15. An immune cell engineered to express the T-cell receptor of claim 13 or 14.

16. The immune cell of claim 15, wherein the cell is a T cell.

17. A treatment method comprising administering to an individual in need the peptide, composition, or cell of any one of claims 1-8 or 15-16.

18. The method of claim 17, wherein the individual has glioblastoma or low-grade glioma.

19. A method for screening HLA-binding peptides, wherein the HLA-binding peptides comprise novel linkages, the method comprising: a) Obtain transcriptome sequence data from tumor tissue samples and normal tissue samples and identify one or more newly linked transcripts in tumor tissue samples; b) Input a sequence containing one or more newly linked transcripts into a computer-readable medium and generate a computer-simulated library of possible multipeptide sequences from the sequence containing one or more newly linked transcripts; c) For each peptide in the multipeptide library obtained through computer simulation, generate: i) The first MHC I presentation score obtained through the antigen treatment advantage algorithm. ii) The second MHC I presentation score obtained by the pan-HLA-A binding affinity algorithm; and d) Identify one or more HLA-binding peptides determined by the first MHC-I presentation score and the second MHC-I presentation score.

20. The method of claim 19, wherein identifying one or more transcripts containing the new link comprises the following steps: i) Quantify splice junctions detected in protein-coding transcripts expressed in tumor tissue samples and quantify splice junctions detected in protein-coding transcripts expressed in normal tissue samples. ii) Identify unannotated splice joins as part of the set of detected splice joins that are not present in the annotated set of classic splice joins; iii) Define the splice frequency of uncommented splice joins as: iv) Identify new connections as unannotated splice connections that have a splice frequency of at least 0.1 in tumor tissue samples and are detected in less than 1% of multiple normal tissue samples.

21. The method of any one of claims 19-20, wherein generating a computer-simulated library of possible multimeric peptide sequences comprises the following steps: i) Translate transcripts containing new links into their corresponding amino acid sequences using computer simulation; ii) For each amino acid sequence, divide the amino acid sequence into all possible multipeptide sequences of length 8 to 11 amino acids; iii) Discard every multipeptide sequence that is detectable in the reference proteome dataset of normal tissue samples.

22. The method of any one of claims 19-21, wherein identifying one or more peptides that bind to HLA comprises the following steps: i) Sort the multipeptides in the library according to the first MHC I presentation score and the second MHC I presentation score; as well as ii) Polymer peptides whose first and second MHC I presentation scores are both in the top 10% are identified as one or more HLA-binding peptides.

23. A method for screening, identifying, and incorporating new linker epitopes (TCRs), the method comprising: a) Introducing a population of antigen-presenting cells (APCs) containing a newly linked HLA-binding peptide to generate a population of APCs loaded with the HLA-binding peptide. b) Expose the T cell population to the APC population loaded with HLA-binding peptide; c) Paired single-cell RNA and V(D)J sequencing of T cells; and d) Determine the TCR sequence of reactive T cells by single-cell RNA sequencing.

Images