{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"bioRxiv"}],"indexed":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T11:12:56Z","timestamp":1790161976004,"version":"4.1.0"},"posted":{"date-parts":[[2026,9,18]]},"group-title":"Cancer Biology","reference-count":33,"publisher":"openRxiv","license":[{"start":{"date-parts":[[2026,9,18]],"date-time":"2026-09-18T00:00:00Z","timestamp":1789689600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2026,9,18]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                <jats:p>Neoantigen immunogenicity prediction is fundamental to personalized cancer vaccines, tumor-infiltrating lymphocyte (TIL) therapy, and TCR-T cell engineering. Existing computational predictors rely primarily on in-vitro correlates of peptide presentation or models trained against assay-based reactivity, and they are typically validated within a single therapeutic setting. We reasoned that the most direct evidence of neoantigen immunogenicity is longitudinal in-vivo elimination: under immune checkpoint blockade (ICB), subclones bearing recognized neoantigens are selectively depleted over time. Here, we present the Neoantigen Elimination Model (NEMo), a two-compartment (CD8 and CD4) machine learning classifier trained on the in-vivo editing (IVE) of neoantigens across serially sequenced, ICB-treated tumors. By using mechanistically inspired NeoPrecis features designed to capture determinants of immunogenicity beyond MHC binding affinity, NEMo recovered assay-confirmed immunogenic neoantigens across four independent, unseen clinical settings\u2014pre-existing immunogenicity screening, personalized cancer vaccines, TIL therapy, and a radiotherapy\u00b1ICB ctDNA cohort\u2014and stratified progression-free survival more strongly than ELISPOT-confirmed reactivity. The editing signal further revealed an immune-evasion architecture in which oncogenic drivers and neoantigens restricted to lost or silenced HLA alleles are systematically spared from editing.<\/jats:p>\n                <jats:sec>\n                  <jats:title>Highlights<\/jats:title>\n                  <jats:list list-type=\"simple\">\n                    <jats:list-item>\n                      <jats:label>-<\/jats:label>\n                      <jats:p>NEMo learns in-vivo immunoediting to prioritize immunogenic targets<\/jats:p>\n                    <\/jats:list-item>\n                    <jats:list-item>\n                      <jats:label>-<\/jats:label>\n                      <jats:p>NEMo outperforms existing tools across four independent clinical settings<\/jats:p>\n                    <\/jats:list-item>\n                    <jats:list-item>\n                      <jats:label>-<\/jats:label>\n                      <jats:p>Complete subclonal coverage and presentation integrity predict survival<\/jats:p>\n                    <\/jats:list-item>\n                    <jats:list-item>\n                      <jats:label>-<\/jats:label>\n                      <jats:p>Clonal oncogenic driver status and HLA damage spare neoantigens from editing<\/jats:p>\n                    <\/jats:list-item>\n                  <\/jats:list>\n                <\/jats:sec>","DOI":"10.64898\/2026.09.17.752425","type":"posted-content","created":{"date-parts":[[2026,9,18]],"date-time":"2026-09-18T21:35:12Z","timestamp":1789767312000},"source":"Crossref","is-referenced-by-count":0,"title":["Predicting Neoantigen Immunogenicity from\n                  <i>In Vivo<\/i>\n                  Immune Editing"],"prefix":"10.64898","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-4868-1509","authenticated-orcid":false,"given":"Timothy J.","family":"Sears","sequence":"first","affiliation":[{"name":"Bioinformatics and Systems Biology Program, University of California San Diego, La Jolla, CA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9162-3486","authenticated-orcid":false,"given":"Ko-han","family":"Lee","sequence":"additional","affiliation":[{"name":"Bioinformatics and Systems Biology Program, University of California San Diego, La Jolla, CA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mar\u00eda","family":"Mu\u00f1oz P\u00e9rez","sequence":"additional","affiliation":[{"name":"Department of Bioengineering, Universidad Carlos III de Madrid, Madrid, Spain"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Robert","family":"Rasmussen","sequence":"additional","affiliation":[{"name":"Bioinformatics and Systems Biology Program, University of California San Diego, La Jolla, CA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7591-6035","authenticated-orcid":false,"given":"Meghana S.","family":"Pagadala","sequence":"additional","affiliation":[{"name":"Biomedical Sciences Program, University of California San Diego, La Jolla, CA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kairi","family":"Tanaka","sequence":"additional","affiliation":[{"name":"School of Biological Sciences, University of California San Diego, La Jolla, CA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7892-886X","authenticated-orcid":false,"given":"Ajay","family":"Subramanian","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology, Stanford University, Stanford, CA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7913-5351","authenticated-orcid":false,"given":"Everett J.","family":"Moding","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology, Stanford University, Stanford, CA, USA"},{"name":"Stanford Cancer Institute, Stanford University, Stanford, CA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6346-8776","authenticated-orcid":false,"given":"Maurizio","family":"Zanetti","sequence":"additional","affiliation":[{"name":"Moores Cancer Center, University of California San Diego, La Jolla, CA, USA"},{"name":"The Laboratory of Immunology, Moores Cancer Center and Department of Medicine, University of California San Diego, La Jolla, CA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1729-2463","authenticated-orcid":false,"given":"Hannah","family":"Carter","sequence":"additional","affiliation":[{"name":"Bioinformatics and Systems Biology Program, University of California San Diego, La Jolla, CA, USA"},{"name":"Moores Cancer Center, University of California San Diego, La Jolla, CA, USA"},{"name":"The Laboratory of Immunology, Moores Cancer Center and Department of Medicine, University of California San Diego, La Jolla, CA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"54368","reference":[{"key":"2026092303350931000_2026.09.17.752425v1.1","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-023-06063-y"},{"issue":"6","key":"2026092303350931000_2026.09.17.752425v1.2","doi-asserted-by":"crossref","first-page":"1994","DOI":"10.1038\/s41591-025-03627-5","article-title":"Neoantigen-specific tumor-infiltrating lymphocytes in gastrointestinal cancers: a phase 2 trial","volume":"31","year":"2025","journal-title":"Nat. Med"},{"key":"2026092303350931000_2026.09.17.752425v1.3","doi-asserted-by":"publisher","DOI":"10.1056\/NEJMoa2119662"},{"key":"2026092303350931000_2026.09.17.752425v1.4","doi-asserted-by":"crossref","unstructured":"N. McGranahan and C. Swanton , \u201cNeoantigen quality, not quantity,\u201d Sci. Transl. Med., vol. 11, no. 506, p. eaax7918, Aug. 2019.","DOI":"10.1126\/scitranslmed.aax7918"},{"key":"2026092303350931000_2026.09.17.752425v1.5","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-022-04735-9"},{"key":"2026092303350931000_2026.09.17.752425v1.6","doi-asserted-by":"crossref","unstructured":"Y. Wu et al., \u201cThe predictive value of tumor mutation burden on efficacy of immune checkpoint inhibitors in cancers: A systematic review and meta-analysis,\u201d Front. Oncol., vol. 9, p. 1161, Nov. 2019.","DOI":"10.3389\/fonc.2019.01161"},{"key":"2026092303350931000_2026.09.17.752425v1.7","doi-asserted-by":"publisher","DOI":"10.1038\/S41587-024-02420-Y"},{"key":"2026092303350931000_2026.09.17.752425v1.8","doi-asserted-by":"publisher","DOI":"10.1016\/j.ccell.2022.10.013"},{"key":"2026092303350931000_2026.09.17.752425v1.9","doi-asserted-by":"crossref","unstructured":"N. Xie , G. Shen , W. Gao , Z. Huang , C. Huang , and L. Fu , \u201cNeoantigens: promising targets for cancer therapy,\u201d Signal Transduct. Target. Ther., vol. 8, no. 1, p. 9, Jan. 2023.","DOI":"10.1038\/s41392-022-01270-x"},{"key":"2026092303350931000_2026.09.17.752425v1.10","doi-asserted-by":"publisher","DOI":"10.1158\/2159-8290.CD-18-1494"},{"key":"2026092303350931000_2026.09.17.752425v1.11","doi-asserted-by":"publisher","DOI":"10.1016\/j.cell.2020.09.015"},{"key":"2026092303350931000_2026.09.17.752425v1.12","doi-asserted-by":"publisher","DOI":"10.1038\/nature22991"},{"key":"2026092303350931000_2026.09.17.752425v1.13","doi-asserted-by":"publisher","DOI":"10.1016\/j.cell.2025.06.019"},{"key":"2026092303350931000_2026.09.17.752425v1.14","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-024-08507-5"},{"key":"2026092303350931000_2026.09.17.752425v1.15","doi-asserted-by":"publisher","DOI":"10.1172\/JCI134915"},{"key":"2026092303350931000_2026.09.17.752425v1.16","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-021-01462-y"},{"key":"2026092303350931000_2026.09.17.752425v1.17","doi-asserted-by":"crossref","unstructured":"J. B. Nilsson , J. Greenbaum , B. Peters , and M. Nielsen , \u201cNetMHCpan-4.2: improved prediction of CD8+ epitopes by use of transfer learning and structural features,\u201d Front. Immunol., vol. 16, no. 1616113, p. 1616113, Aug. 2025.","DOI":"10.3389\/fimmu.2025.1616113"},{"key":"2026092303350931000_2026.09.17.752425v1.18","doi-asserted-by":"crossref","unstructured":"J. B. Nilsson et al., \u201cAccurate prediction of HLA class II antigen presentation across all loci using tailored data acquisition and refined machine learning,\u201d Sci. Adv., vol. 9, no. 47, p. eadj6367, Nov. 2023.","DOI":"10.1126\/sciadv.adj6367"},{"key":"2026092303350931000_2026.09.17.752425v1.19","doi-asserted-by":"publisher","DOI":"10.1056\/NEJMoa1012863"},{"key":"2026092303350931000_2026.09.17.752425v1.20","doi-asserted-by":"crossref","unstructured":"J. Schmidt et al., \u201cPrediction of neo-epitope immunogenicity reveals TCR recognition determinants and provides insight into immunoediting,\u201d Cell Rep. Med., vol. 2, no. 2, p. 100194, Feb. 2021.","DOI":"10.1016\/j.xcrm.2021.100194"},{"issue":"8","key":"2026092303350931000_2026.09.17.752425v1.21","doi-asserted-by":"crossref","first-page":"861","DOI":"10.1038\/s42256-023-00694-6","article-title":"Deep neural networks predict class I major histocompatibility complex epitope presentation and transfer learn neoepitope immunogenicity,\u201d Nat","volume":"5","year":"2023","journal-title":"Mach. Intell"},{"key":"2026092303350931000_2026.09.17.752425v1.22","doi-asserted-by":"crossref","unstructured":"G. Wang et al., \u201cTLimmuno2: predicting MHC class II antigen immunogenicity through transfer learning,\u201d Brief. Bioinform., vol. 24, no. 3, p. bbad116, May 2023.","DOI":"10.1093\/bib\/bbad116"},{"key":"2026092303350931000_2026.09.17.752425v1.23","doi-asserted-by":"crossref","unstructured":"S. K. Dhanda et al., \u201cPredicting HLA CD4 immunogenicity in human populations,\u201d Front. Immunol., vol. 9, p. 1369, Jun. 2018.","DOI":"10.3389\/fimmu.2018.01369"},{"key":"2026092303350931000_2026.09.17.752425v1.24","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-024-08508-4"},{"key":"2026092303350931000_2026.09.17.752425v1.25","doi-asserted-by":"crossref","unstructured":"K.-H. Lee , T. J. Sears , M. Zanetti , and H. Carter , \u201cNeoPrecis: enhancing immunotherapy response prediction through integration of qualified immunogenicity and clonality-aware neoantigen landscapes,\u201d Nat. Commun., vol. 17, no. 1, p. 1966, Jan. 2026.","DOI":"10.1038\/s41467-026-68651-6"},{"issue":"2","key":"2026092303350931000_2026.09.17.752425v1.26","doi-asserted-by":"crossref","first-page":"440","DOI":"10.1038\/s41591-022-02163-w","article-title":"Persistent mutation burden drives sustained anti-tumor immune responses","volume":"29","year":"2023","journal-title":"Nat. Med"},{"key":"2026092303350931000_2026.09.17.752425v1.27","doi-asserted-by":"crossref","unstructured":"E. Hurtado , A. Bouchard-C\u00f4t\u00e9 , and A. Roth , \u201cPhyClone: accurate Bayesian reconstruction of cancer phylogenies from bulk sequencing,\u201d Bioinformatics, vol. 41, no. 7, p. btaf344, Jul. 2025.","DOI":"10.1093\/bioinformatics\/btaf344"},{"key":"2026092303350931000_2026.09.17.752425v1.28","doi-asserted-by":"publisher","DOI":"10.1038\/s41588-020-0687-1"},{"issue":"11","key":"2026092303350931000_2026.09.17.752425v1.29","doi-asserted-by":"crossref","first-page":"103773","DOI":"10.1016\/j.drudis.2023.103773","article-title":"Personalized neoantigen cancer vaccines: An analysis of the clinical and commercial potential of ongoing development programs","volume":"28","year":"2023","journal-title":"Drug Discov. Today"},{"key":"2026092303350931000_2026.09.17.752425v1.30","doi-asserted-by":"publisher","DOI":"10.1016\/j.cell.2017.10.001"},{"key":"2026092303350931000_2026.09.17.752425v1.31","doi-asserted-by":"publisher","DOI":"10.1038\/s41588-023-01313-1"},{"key":"2026092303350931000_2026.09.17.752425v1.32","doi-asserted-by":"publisher","DOI":"10.1038\/s41588-024-01883-8"},{"key":"2026092303350931000_2026.09.17.752425v1.33","doi-asserted-by":"publisher","DOI":"10.1200\/jco-26-00769"}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.64898\/2026.09.17.752425","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T10:35:25Z","timestamp":1790159725000},"score":1,"resource":{"primary":{"URL":"http:\/\/biorxiv.org\/lookup\/doi\/10.64898\/2026.09.17.752425"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,18]]},"references-count":33,"URL":"https:\/\/doi.org\/10.64898\/2026.09.17.752425","relation":{},"subject":[],"published":{"date-parts":[[2026,9,18]]},"subtype":"preprint"}}