CN-122003713-A - Calculation method for selecting personalized neoantigen vaccine

Abstract

The present invention provides systems and methods for identifying personalized tumor vaccines. Somatic variants of the subject are determined, and RNA sequence reads are obtained from tumors of the subject. Based on the RNA sequence reads, fusion proteins are determined, each fusion protein being a fusion of a portion of a first protein with a portion of a second protein. A candidate neoantigen is selected, the candidate neoantigen comprising a first subset of candidate neoantigens encoding a somatic variant and a second subset of candidate neoantigens encoding residues of the portion of the first protein and the portion of the second protein. The score of each candidate neoantigen is determined using a scoring function that includes a first scoring term that weights candidate neoantigens with relatively high MHC class I affinity given the subject's HLA class I. Two or more candidate neoantigens are selected for the tumor vaccine as a final set of neoantigens based on the respective scores.

Inventors

• T. O'Donnell
• J. Kodish
• N. Bahadevaj

Assignees

• 西奈山伊坎医学院

Dates

Publication Date

20260508

Application Date

20240412

Priority Date

20230412

Claims (20)

1. 1. A method for identifying a tumor vaccine personalized to a human subject having cancer, the method comprising: (A) Determining a first plurality of somatic variants of the subject; (B) Obtaining a first plurality of sequence reads from RNA molecules in a sample of a tumor obtained from the subject; (C) Determining one or more fusion proteins encoded by the first plurality of sequence reads from the first plurality of sequence reads, wherein each respective fusion protein of the one or more fusion proteins is a fusion of a portion of a respective first human protein with a portion of a respective second human protein; (D) Selecting a plurality of candidate neoantigens comprising a first subset of candidate neoantigens and a second subset of candidate neoantigens, wherein Each neoantigen in the first subset of candidate neoantigens encodes a somatic variant in the first plurality of somatic variants, and Each neoantigen in the second subset of candidate neoantigens encodes one or more residues from the portion of the respective first human protein and one or more residues from the portion of the respective second human protein; (E) Determining a respective score for each respective candidate neoantigen of the plurality of candidate neoantigens using a scoring function comprising a first scoring term that weights respective candidate neoantigens having a higher class I Major Histocompatibility Complex (MHC) affinity relative to respective candidate neoantigens having a lower class I major MHC affinity given the class I Human Leukocyte Antigen (HLA) type of the human subject, and (F) Two or more candidate neoantigens of the plurality of candidate neoantigens are selected for the tumor vaccine as a final set of neoantigens based on the respective scores of each candidate neoantigen of the plurality of candidate neoantigens.
2. 2. The method of claim 1, the method further comprising: Determining a respective allele-specific expression of each individual cell variant in the first plurality of variants using the first plurality of sequence reads, and The scoring function further includes a second scoring term that weights respective candidate neoantigens that represent alleles that are more abundant in the first plurality of sequence reads relative to respective candidate neoantigens that represent alleles that are less abundant in the first plurality of sequence reads.
3. 3. The method of claim 1, wherein the determining the first plurality of somatic variants of the subject comprises: obtaining a second plurality of sequence reads from DNA molecules in the tumor sample obtained from the subject; Obtaining a third plurality of sequence reads from a DNA molecule in a normal sample obtained from the subject, and Using the second plurality of sequence reads and the third plurality of sequence reads, identifying the first plurality of somatic variants of the subject by including in the first plurality of somatic variants those somatic variants that are observed in the second plurality of sequence reads but not observed in the third plurality of sequence reads.
4. 4. A method according to claim 3, the method further comprising: Sequencing the first exome at a coverage of at least 100x to obtain the second plurality of sequence reads, and Second exome sequencing at least 100x coverage to obtain the second plurality of sequence reads.
5. 5. The method of claim 3 or 4, wherein the normal sample is a tissue sample near the original location of the tumor of the subject.
6. 6. The method of claim 3 or 4, wherein the normal sample is a blood sample.
7. 7. The method of any one of claims 1 to 6, wherein the cancer is glioblastoma or prostate cancer.
8. 8. The method of any one of claims 1 to 7, wherein the cancer is a carcinoma, melanoma, lymphoma/leukemia, sarcoma, or glioma.
9. 9. The method of any one of claims 1 to 8, wherein the cancer is lung, pancreatic, colon, gastric or esophageal cancer, breast, ovarian, prostate or liver cancer.
10. 10. The method of any one of claims 1 to 9, wherein each somatic variant in the first plurality of somatic variants is a single nucleotide variant or an indel mutation.
11. 11. The method of any one of claims 1 to 10, wherein the sample of the tumor is a freshly frozen sample.
12. 12. The method of any one of claims 1 to 11, wherein each respective candidate neoantigen of the plurality of candidate neoantigens is N residues in length, wherein N is 8, 9, 10,11, 12, 13 or 14.
13. 13. The method of any one of claims 1 to 11, wherein each respective candidate neoantigen of the plurality of candidate neoantigens is 9 residues in length.
14. 14. The method of any one of claims 1 to 13, further comprising excluding from the plurality of candidate neoantigens or from the final set of candidate neoantigens those candidate neoantigens that comprise a mutation at the N-terminal residue position relative to the wild-type sequence.
15. 15. The method of any one of claims 1-14, further comprising determining the HLA class I type of the human subject using the first plurality of sequencing reads.
16. 16. The method of any one of claims 1 to 14, further comprising determining the HLA class I type of the human subject using polymerase chain reaction using a biological sample from the cancer subject.
17. 17. The method of any one of claims 1 to 16, the method further comprising: Determining the hydrophobicity of each respective candidate neoantigen of the plurality of candidate neoantigens, and The scoring function further includes scoring terms for hydrophobicity of the respective candidate neoantigen.
18. 18. The method of claim 17, wherein the determining the respective hydrophobicity of each candidate neoantigen of the plurality of candidate neoantigens comprises assigning the respective candidate neoantigen a maximum hydrophobicity score for any 7 mers within the respective candidate neoantigen.
19. 19. The method of claim 17 or 18, wherein the scoring for the hydrophobicity of the respective candidate neoantigen weights a more hydrophobic candidate neoantigen relative to a less hydrophobic candidate neoantigen.
20. 20. The method of any one of claims 1-19, wherein the scoring function further comprises an item for MHC class II affinity for the respective candidate neoantigen given the HLA class II type of the human subject that weights an increase in weight for the respective candidate neoantigen having a higher MHC class II affinity than for the respective candidate neoantigen having a lower MHC class II affinity.

Description

Calculation method for selecting personalized neoantigen vaccine Cross Reference to Related Applications The present application claims priority from U.S. provisional patent application Ser. No. 63/495,637 filed on 4/12 of 2023, which is hereby incorporated by reference in its entirety. Reference to an electronically submitted sequence Listing The present application contains a sequence table in XML format submitted electronically via EFS-Web. The content of the XML copy created at month 4 of 2024 was named "1045935046wo_sl.xml" and had a size of 22000 bytes. The sequence listing is incorporated by reference herein in its entirety. Technical Field The present disclosure relates generally to systems and methods for identifying a tumor vaccine personalized to a human subject having cancer, wherein the tumor vaccine comprises a neoantigen. Background Cancer specific neoantigens resulting from genetic alterations accumulated in tumor cells encode novel amino acid segments that are not present in the normal genome. Thus, these tumor-specific peptides are not negatively selected by the immune system as "self" proteomes. Thus, the total loading of neoantigens deduced by computer analysis of whole exome sequencing data from patient tumors (in silico analysis) can be used as predictors of positive response to immunotherapy regimens, as well as for vaccine design, including but not limited to peptide-based vaccines, RNA vaccines and/or DNA vaccine designs (Luksza, M.et al Nature 551,517-520 (2017), balachandran, V.P.et al Nature 551, S12-S16 (2017), charoentong, P.et al Cell report (CellReports) 18,248-262 (2016), hugo, W.et al Cell (Cell) 165,35-44 (2016)). Neoantigen vaccines were developed by comparing the genotype of tumor cells to that of normal tissue or blood matching the patient. The collected somatic missense mutations and frameshift mutations are then converted to the corresponding tumor-specific peptides, which are then screened for MHC-I epitopes by running experimental functional tests or in silico predictive algorithms. There are several algorithms that optimize the prediction of epitope-HLA interactions on a computer, so that MHC class I tumor neoepitopes can be predicted, and to a lesser extent MHC class II tumor neoepitopes. Alternatively, mass spectrometry-based methods for predicting tumor epitopes also exist today. These epitopes can then be used in short and long peptide-based vaccines, boost Dendritic Cell (DC) -based vaccination, cell therapies for priming and gene modification of adoptive autologous T cell transfer (Branca, m.a. (Nat. Biotechnol.)) 34,1019-1024 (2016)). In this regard, several research groups have reported encouraging results of neoantigen-based cancer vaccines to generate tumor antigen-specific immune responses in both mouse models and clinical trials. In addition, the number and quality of neoantigens have been demonstrated to be of predictive value for the clinical outcome of checkpoint blocking immunotherapy for certain tumor types. The recognition of neoantigens by vaccination or by adoptive T cell therapy may have unprecedented potential to drive the combination of cancer immunotherapy with other methods (Roudko v. Et al, front of immunology (front. Immunol.))) 11:27,1-11 (2020)). In view of the foregoing background, what is needed in the art are systems and methods for improving cancer treatment in humans. In addition, in view of the above background, there is a need to develop tumor vaccines that utilize tumor-associated antigens or neoantigens to improve immune responses. Disclosure of Invention The present disclosure addresses a need in the art for systems and methods for determining personalized treatments for human cancers. One of the obstacles to overcome in cancer therapy is the immunosuppressive tumor microenvironment (see, e.g., quail DF, joyce JA., cancer cells (CANCER CELL); 2017;31 (3): 326-341, which is hereby incorporated by reference in its entirety). Tumor mutations result in the formation of tumor-specific antigens (known as neoantigens) that are not present in normal tissues and can be recognized by the immune system, thereby providing specific targets for anti-tumor therapy. The neoantigen vaccine is a new tool for treating patients suffering from cancer. The neoantigen vaccines can play an important role in immunomodulation, as they are expected to expand the endogenous repertoire of tumor-specific T cells to eliminate any residual tumor cells. In general, such therapeutic approaches based on manipulation of the adaptive immune system may have relatively low risk of toxicity and thus have good side effect profiles due to the inherent specificity of the immune response induced, and in addition, these immune-based therapeutic agents may potentially promote prolonged relief of the disease due to the introduction and persistence of the immune cell population activated in vivo by the patient after receiving a course of treatment. Several groups have used