Methods of neoantigen identification
Inventors
Schoenberger, Stephen P. • Peters, Bjoern • Sette, Alessandro • Greenbaum, Jason • Cohen, Ezra • Miller, Aaron
Assignees
La Jolla Institute for Allergy and Immunology • University of California San Diego UCSD
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12427195-B1
Publication Date
2025-09-30
Expiration Date
Abstract
Systems and methods for identifying neoantigen peptide candidates are described. Such systems and methods can be used to generate engineered immune cells that target neoantigen peptides. In certain embodiments, the systems and methods do not require in silico epitope binding prediction algorithms. In certain embodiments, neoantigens are identified for a cancer or tumor with a low mutational burden.
Core Innovation
The invention relates to a method of identifying a neoantigen peptide candidate by generating tumor sequence reads from a tumor sample and normal sequence reads from a normal control sample, and comparing the tumor sequence reads with the normal sequence reads to identify one or more exome variants. The method selects from the one or more exome variants a set of exome variants that satisfy a variant calling policy, wherein the variant allele frequency of a variant among the set of exome variants in a normal exome is equal to about 50% or less.
For at least one exome variant of the set of exome variants that satisfy the variant calling policy, the method generates one or more synthetic peptides, each synthetic peptide having one or more mutated amino acids at one or more pre-selected locations of the synthetic peptide. The method excludes the use of an in silico MHC binding prediction algorithm, and evaluates immunogenicity for one or more of the peptides.
The method identifies at least one peptide demonstrating a predetermined immunogenic activity as a neoantigen peptide candidate. The predetermined immunogenic activity comprises production of IFN-γ or IL-5 by a T cell.
Claims Coverage
The independent claim defines a neoantigen peptide candidate identification method with 4 main inventive aspects: tumor and normal sequence comparison to exome variants, a normal allele-frequency-based variant calling policy, synthetic peptide generation at pre-selected locations, and immunogenicity evaluation defined by IFN-γ or IL-5 production by a T cell.
Tumor and normal sequence comparison to exome variants
Generating tumor sequence reads from a tumor sample and normal sequence reads from a normal control sample, comparing the tumor sequence reads with the normal sequence reads to identify one or more exome variants.
Variant calling policy with normal allele frequency cutoff
Selecting from the one or more exome variants a set of exome variants that satisfy a variant calling policy, wherein the variant allele frequency of a variant among the set of exome variants in a normal exome is equal to about 50% or less.
Synthetic peptides from selected exome variants without in silico MHC prediction
Generating for at least one exome variant of the set of exome variants that satisfy the variant calling policy one or more synthetic peptides, each synthetic peptide having one or more mutated amino acids at one or more pre-selected locations of the synthetic peptide, wherein the method excludes the use of an in silico MHC binding prediction algorithm.
Immunogenicity evaluation using IFN-γ or IL-5 production
Evaluating immunogenicity for one or more of the peptides and identifying at least one peptide demonstrating a predetermined immunogenic activity as a neoantigen peptide candidate, wherein the predetermined immunogenic activity comprises production of IFN-γ or IL-5 by a T cell.
Across the claim family, the method of identifying neoantigen peptide candidates is anchored by tumor and normal sequencing comparison to exome variants, selecting variants using a normal allele-frequency policy of about 50% or less, generating mutated synthetic peptides at pre-selected locations, and selecting peptides by immunogenic activity defined by T-cell production of IFN-γ or IL-5. Dependent claim refinements further narrow the input read types and immunogenicity characterization, and add downstream use for treating an individual with compositions comprising at least one identified peptide or an engineered immune cell that biologically recognizes the peptide.
Stated Advantages
Functional testing indicates that a substantial fraction of candidates stimulate T cells.
Conventional in silico prediction can miss immunogenic neoepitopes, including false negatives.
Excludes the use of an in silico MHC binding prediction algorithm.
Documented Applications
Use for peptide vaccine and engineered immune cells, including engineered TCR and chimeric antigen receptor (CAR)-based formats.
Use for low mutational burden cancers.
Treating an individual associated with a tumor sample using a composition comprising at least one identified peptide or an engineered immune cell that biologically recognizes the at least one peptide.
Interested in licensing this patent?