跳到正文
原文
Nature 健康护理主题·· 2026-10-14AI 评分58

个体化新抗原靶向疫苗联合nivolumab治疗卵巢癌的I期试验

Personalized neoantigen-targeting cancer vaccines in ovarian cancer: a phase 1 trial

AI 导读

Dana-Farber癌症研究院团队在Nature Medicine发表个体化新抗原靶向疫苗(NeoVax)联合nivolumab治疗卵巢癌的I期试验。扩展数据报告了疫苗诱导的外周ELISpot应答与TCRβ克隆扩增,以及肿瘤浸润T细胞对患者来源类器官的反应。1例进展患者的活检发现JAK1结构变异,部分患者疫苗前即存在HLA杂合性缺失。

正文

Data availability

Bulk WES, RNA-seq, WGS and TCR-seq, and single-cell RNA and single-cell TCR-seq data are deposited and available on dbGaP under accession code phs004380.v1. Source data are provided with this paper.

Code availability

All code used in the processing and analysis of bulk TCR-seq data is available on GitHub at https://github.com/jeremymsimon/. Other code using established tools will be made available within 30 days upon request to the corresponding authors.

References

  1. Zhang, L. et al. Intratumoral T cells, recurrence, and survival in epithelial ovarian cancer. N. Engl. J. Med. 348, 203–213 (2003).

    Article  CAS  PubMed  Google Scholar 

  2. Sato, E. et al. Intraepithelial CD8+ tumor-infiltrating lymphocytes and a high CD8+/regulatory T cell ratio are associated with favorable prognosis in ovarian cancer. Proc. Natl Acad. Sci. USA 102, 18538–18543 (2005).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  3. Varga, A. et al. Pembrolizumab in patients with programmed death ligand 1-positive advanced ovarian cancer: Analysis of KEYNOTE-028. Gynecol. Oncol. 152, 243–250 (2019).

    Article  CAS  PubMed  Google Scholar 

  4. Hamanishi, J. et al. Safety and antitumor activity of anti-PD-1 antibody, nivolumab, in patients with platinum-resistant ovarian cancer. J. Clin. Oncol. 33, 4015–4022 (2015).

    Article  CAS  PubMed  Google Scholar 

  5. Colombo, N. et al. Pembrolizumab plus weekly paclitaxel in platinum-resistant recurrent ovarian cancer (ENGOT-ov65/KEYNOTE-B96): a multicentre, randomised, double-blind, phase 3 study. Lancet 407, 1525–1537 (2026).

    Article  CAS  PubMed  Google Scholar 

  6. Cancer Genome Atlas Research Network. Integrated genomic analyses of ovarian carcinoma. Nature 474, 609–615 (2011).

    Article  Google Scholar 

  7. Liu, M. et al. Improved T-cell immunity following neoadjuvant chemotherapy in ovarian cancer. Clin. Cancer Res. 28, 3356–3366 (2022).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  8. Kandalaft, L. E., Dangaj Laniti, D. & Coukos, G. Immunobiology of high-grade serous ovarian cancer: lessons for clinical translation. Nat. Rev. Cancer 22, 640–656 (2022).

    Article  CAS  PubMed  Google Scholar 

  9. Vazquez-Garcia, I. et al. Ovarian cancer mutational processes drive site-specific immune evasion. Nature 612, 778–786 (2022).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  10. Hornburg, M. et al. Single-cell dissection of cellular components and interactions shaping the tumor immune phenotypes in ovarian cancer. Cancer Cell 39, 928–944 e926 (2021).

    Article  CAS  PubMed  Google Scholar 

  11. Keskin, D. B. et al. Neoantigen vaccine generates intratumoral T cell responses in phase Ib glioblastoma trial. Nature 565, 234–239 (2019).

    Article  CAS  PubMed  Google Scholar 

  12. Braun, D. A. et al. A neoantigen vaccine generates antitumour immunity in renal cell carcinoma. Nature 639, 474–482 (2025).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  13. Rojas, L. A. et al. Personalized RNA neoantigen vaccines stimulate T cells in pancreatic cancer. Nature 618, 144–150 (2023).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  14. Sarkizova, S. et al. A large peptidome dataset improves HLA class I epitope prediction across most of the human population. Nat. Biotechnol. 38, 199–209 (2020).

    Article  CAS  PubMed  Google Scholar 

  15. Jurtz, V. et al. NetMHCpan-4.0: improved peptide-MHC class I interaction predictions integrating eluted ligand and peptide binding affinity data. J. Immunol. 199, 3360–3368 (2017).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  16. Parsons, H. A. et al. Sensitive detection of minimal residual disease in patients treated for early-stage breast cancer. Clin. Cancer Res. 26, 2556–2564 (2020).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  17. Moodie, Z., Huang, Y., Gu, L., Hural, J. & Self, S. G. Statistical positivity criteria for the analysis of ELISpot assay data in HIV-1 vaccine trials. J. Immunol. Methods 315, 121–132 (2006).

    Article  CAS  PubMed  Google Scholar 

  18. Oliveira, G. et al. Phenotype, specificity and avidity of antitumour CD8(+) T cells in melanoma. Nature 596, 119–125 (2021).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  19. Caushi, J. X. et al. Transcriptional programs of neoantigen-specific TIL in anti-PD-1-treated lung cancers. Nature 596, 126–132 (2021).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  20. Anadon, C. M. et al. Ovarian cancer immunogenicity is governed by a narrow subset of progenitor tissue-resident memory T cells. Cancer Cell 40, 545–557 e513 (2022).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  21. Martinez-Jimenez, F. et al. Genetic immune escape landscape in primary and metastatic cancer. Nat. Genet. 55, 820–831 (2023).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  22. Filip, I. et al. Pervasiveness of HLA allele-specific expression loss across tumor types. Genome Med. 15, 8 (2023).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  23. Takaya, H., Nakai, H., Takamatsu, S., Mandai, M. & Matsumura, N. Homologous recombination deficiency status-based classification of high-grade serous ovarian carcinoma. Sci Rep. 10, 2757 (2020).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  24. Knijnenburg, T. A. et al. Genomic and molecular landscape of DNA damage repair deficiency across the cancer genome atlas. Cell Rep. 23, e236 (2018).

    Article  Google Scholar 

  25. Broad GDAC Firehose. stddata__2016_01_28 https://gdac.broadinstitute.org/runs/stddata__2016_01_28/ (Broad Institute of MIT and Harvard, 2016).

  26. Sethna, Z. et al. RNA neoantigen vaccines prime long-lived CD8(+) T cells in pancreatic cancer. Nature 639, 1042–1051 (2025).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  27. Huber, F. & Bassani-Sternberg, M. Defects in antigen processing and presentation: mechanisms, immune evasion and implications for cancer vaccine development. Nat. Rev. Immunol. https://doi.org/10.1038/s41577-025-01208-8 (2025).

  28. Huber, F. et al. A comprehensive proteogenomic pipeline for neoantigen discovery to advance personalized cancer immunotherapy. Nat. Biotechnol. 43, 1360–1372 (2025).

    Article  CAS  PubMed  Google Scholar 

  29. Blass, E. et al. A multi-adjuvant personal neoantigen vaccine generates potent immunity in melanoma. Cell (2025).

  30. Cancer Genome Atlas Research Network. Comprehensive molecular characterization of clear cell renal cell carcinoma. Nature 499, 43–49 (2013).

    Article  Google Scholar 

  31. Lawrence, M. S. et al. Discovery and saturation analysis of cancer genes across 21 tumour types. Nature 505, 495–501 (2014).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  32. Hill, S. J. et al. Prediction of DNA repair inhibitor response in short-term patient-derived ovarian cancer organoids. Cancer Discov. 8, 1404–1421 (2018).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  33. Lee, P. C. et al. Reversal of viral and epigenetic HLA class I repression in Merkel cell carcinoma. J. Clin. Invest. https://doi.org/10.1172/jci151666 (2022).

  34. Keskin, D. B. et al. Direct identification of an HPV-16 tumor antigen from cervical cancer biopsy specimens. Front. Immunol. 2, 75 (2011).

    PubMed  PubMed Central  Google Scholar 

  35. Hao, Y. et al. Dictionary learning for integrative, multimodal and scalable single-cell analysis. Nat. Biotechnol. 42, 293–304 (2024).

    Article  CAS  PubMed  Google Scholar 

  36. Hippen, A. A. et al. miQC: an adaptive probabilistic framework for quality control of single-cell RNA-sequencing data. PLoS Comput. Biol. 17, e1009290 (2021).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  37. Korsunsky, I. et al. Fast, sensitive and accurate integration of single-cell data with Harmony. Nat. Methods 16, 1289–1296 (2019).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  38. Germain, P. L., Lun, A., Garcia Meixide, C., Macnair, W. & Robinson, M. D. Doublet identification in single-cell sequencing data using scDblFinder. F1000Res. 10, 979 (2021).

    Article  PubMed  PubMed Central  Google Scholar 

  39. Oliveira, G. et al. Preexisting tumor-resident T cells with cytotoxic potential associate with response to neoadjuvant anti-PD-1 in head and neck cancer. Sci. Immunol. 8, eadf4968 (2023).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  40. Zappia, L. & Oshlack, A. Clustering trees: a visualization for evaluating clusterings at multiple resolutions. Gigascience https://doi.org/10.1093/gigascience/giy083 (2018).

  41. Oliveira, G. et al. Landscape of helper and regulatory antitumour CD4(+) T cells in melanoma. Nature 605, 532–538 (2022).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  42. Liberzon, A. et al. The Molecular Signatures Database (MSigDB) hallmark gene set collection. Cell Syst. 1, 417–425 (2015).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  43. Korotkevich, G. et al. Fast gene set enrichment analysis. Preprint at bioRxiv https://doi.org/10.1101/060012 (2021).

  44. Wickham, H. ggplot2: Elegant Graphics for Data Analysis (Springer, 2016).

  45. Afeyan, A. B., Wu, C. J. & Oliveira, G. Rapid parallel reconstruction and specificity screening of hundreds of T cell receptors. Nat. Protoc. https://doi.org/10.1038/s41596-024-01061-4 (2024).

  46. Parsons, H. A. et al. Circulating tumor DNA association with residual cancer burden after neoadjuvant chemotherapy in triple-negative breast cancer in TBCRC 030. Ann. Oncol. 34, 899–906 (2023).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  47. Chen, X. et al. Manta: rapid detection of structural variants and indels for germline and cancer sequencing applications. Bioinformatics 32, 1220–1222 (2016).

    Article  CAS  PubMed  Google Scholar 

Download references

Acknowledgements

We are appreciative of S. Hill for input and assistance with tissue processing; A. Afeyan for computational assistance with TCR identification; E. Lin for technical assistance; L. Pomerance for input on immune analyses; B. Igoresceu for input on cfDNA; S. Pyle for artwork; staff at Almac Sciences for vaccine peptide manufacture; Bristol Myers Squibb for providing nivolumab; and the Dana-Farber Cancer Institute (DFCI) Cell Manipulation Core Facility for sterility and endotoxin testing.

Funding

This work was supported by National Cancer Institute (NCI) Dana-Farber/Harvard Cancer Center Ovarian Cancer SPORE (P50-CA240243-05; to U.A.M., P.A.K. and C.J.W.), the Mark Foundation Endeavor Award (to C.J.W.), National Institutes of Health (NIH)/NCI award 5R01CA276865 (to C.J.W.), Breast Cancer Research Fund award (to U.A.M. and P.A.K.), Friends of Dana-Farber (to M.C.S. and R.J.P.), the Fund in Memory of Patty Klein, the Feldmann Ovarian Cancer Research Fund, and the Blavatnik Family Foundation. M.C.S. received support from the Lubin Family Foundation, the Dana-Farber NIH Medical Oncology training grant (NIH T32CA009172) and the Ovarian Cancer Research Association. A.S. was supported by a Wellcome Trust Early Career Award (227000/Z/23/Z). S.L. is supported by the NCI Research Specialist Award (R50CA251956). E.H.S. is supported in part by NCI K08CA237871. G.O. received support from the Claudia Adams Barr Program for Innovative Cancer Research and an American Cancer Society Institutional Research Grant. D.B.K. is supported by U54-CA272688, R01-HL157174, R01-CA285308, R01-CA279391, R01-NS140967 and Alex’s Lemonade Stand. P.A.O. is supported in part by NCI-1R01CA229261, the Francis and Adele Kittredge Family Immuno-Oncology and Melanoma Research Fund, the Fisher Family Fund, The Morgan Family Fund for Melanoma Research, the Susana and Bruce Hampton Research Fund and the Faircloth Family Research Fund. C.J.W. is a member of the Parker Institute for Cancer Immunotherapy at DFCI. Her work is supported, in part, by the Parker Institute for Cancer Immunotherapy. C.J.W. is also the Lavine Family Chair for Preventative Cancer Therapies at DFCI. The funders had no role in the conduct of the trial or in the analysis of clinical or immunological end points.

Author information

Author notes

  1. These authors contributed equally: Panagiotis A. Konstantinopoulos, Catherine J. Wu.

Authors and Affiliations

  1. Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA, USA

    MacLean C. Sellars, Jessica A. Frank, Cleo Forman, Chloe R. Tu, Kyohei K. Tabata, Haley E. Sax, Amit Sud, Savni Sarolkar, Martin Hayes, Oriol Olive, Haley Greenslade, Edward F. Fritsch, Elizabeth H. Stover, Carolyn Krasner, Rebecca Porter, Ursula Matulonis, Giacomo Oliveira, Patrick A. Ott, Derin B. Keskin, Panagiotis A. Konstantinopoulos & Catherine J. Wu

  2. Harvard Medical School, Boston, MA, USA

    MacLean C. Sellars, Jessica A. Frank, Antonia Kowelewski, Nir Hacohen, Elizabeth H. Stover, Carolyn Krasner, Rebecca Porter, Ursula Matulonis, Giacomo Oliveira, Patrick A. Ott, Derin B. Keskin, Panagiotis A. Konstantinopoulos & Catherine J. Wu

  3. Department of Data Science, Dana-Farber Cancer Institute, Boston, MA, USA

    Bohoon Shim, Cleo Forman, Chloe R. Tu, Su-Chun Cheng, Nabihah Tayob & Jeremy M. Simon

  4. Translational Immunogenomics Laboratory, Dana-Farber Cancer Institute, Boston, MA, USA

    Allison Vanasse, Vipheaviny Chea, Wesley Lu, J. Thomas Janes 3rd, Shuqiang Li, Kenneth J. Livak & Derin B. Keskin

  5. Broad Institute of MIT and Harvard, Cambridge, MA, USA

    Ruolin Liu, Antonia Kowelewski, Catherine Song, Kan Xiong, Pia Jain, Andjela Crnjac, Micah Rickles-Young, Carrie Cibulskis, Nir Hacohen, Edward F. Fritsch, Shuqiang Li, Gad Getz, Patrick A. Ott, Viktor A. Adalsteinsson, Derin B. Keskin & Catherine J. Wu

  6. Department of Biostatistics, Harvard T.H. Chan School of Public Health, Boston, MA, USA

    Su-Chun Cheng, Nabihah Tayob & Jeremy M. Simon

  7. Centre for Immuno-Oncology, Nuffield Department of Medicine, University of Oxford, Oxford, UK

    Amit Sud

  8. Krantz Family Center for Cancer Research, Massachusetts General Hospital, Boston, MA, USA

    Nir Hacohen

  9. Department of Pathology, Massachusetts General Hospital, Boston, MA, USA

    Gad Getz

  10. Department of Pathology, Harvard Medical School, Boston, MA, USA

    Gad Getz

  11. Center for Immuno-Oncology, Dana-Farber Cancer Institute, Boston, MA, USA

    Patrick A. Ott

  12. Department of Medicine, Brigham and Women’s Hospital, Boston, MA, USA

    Patrick A. Ott & Catherine J. Wu

  13. Section of Computational Biomedicine, Department of Medicine, Boston University School of Medicine, Boston, MA, USA

    Derin B. Keskin

  14. Department of Medical Education and Biomedical Sciences, Methodist University CFVH School of Medicine, Fayetteville, NC, USA

    Derin B. Keskin

  15. Mellen and Eisenson Family Center for BRCA and Related Genes, Dana-Farber Cancer Institute, Boston, MA, USA

    Panagiotis A. Konstantinopoulos

Authors

  1. MacLean C. Sellars
  2. Jessica A. Frank
  3. Bohoon Shim
  4. Allison Vanasse
  5. Ruolin Liu
  6. Cleo Forman
  7. Chloe R. Tu
  8. Kyohei K. Tabata
  9. Antonia Kowelewski
  10. Su-Chun Cheng
  11. Catherine Song
  12. Kan Xiong
  13. Pia Jain
  14. Andjela Crnjac
  15. Haley E. Sax
  16. Vipheaviny Chea
  17. Wesley Lu
  18. J. Thomas Janes 3rd
  19. Micah Rickles-Young
  20. Amit Sud
  21. Savni Sarolkar
  22. Martin Hayes
  23. Oriol Olive
  24. Haley Greenslade
  25. Carrie Cibulskis
  26. Nir Hacohen
  27. Edward F. Fritsch
  28. Shuqiang Li
  29. Kenneth J. Livak
  30. Nabihah Tayob
  31. Elizabeth H. Stover
  32. Carolyn Krasner
  33. Rebecca Porter
  34. Ursula Matulonis
  35. Gad Getz
  36. Giacomo Oliveira
  37. Patrick A. Ott
  38. Jeremy M. Simon
  39. Viktor A. Adalsteinsson
  40. Derin B. Keskin
  41. Panagiotis A. Konstantinopoulos
  42. Catherine J. Wu

Contributions

P.A.K. was the principal investigator and directed the overall study. P.A.O. was the Investigational New Drug application holder. M.C.S., P.A.K. and C.J.W. designed and performed experimental and data analysis together with J.A.F., B.S., A.V., R.L., C.F., C.R.T., K.T., S.C., C.S., K.X., P.J., A.C., H.E.S., V.C., W.L., T.J., M.R.Y., K.J.L., S.L., A.S., G.O. and D.B.K. C.F., M.Y.R., C.C., D.B.K. and P.A.K. analyzed sequencing data and selected neoantigen targets. S.C., N.T. and J.M.S. designed and performed statistical analyses. S.S., M.H., O.O. and H.G. coordinated clinical research. R.P., E.H.S., C.K. and U.A.M. were co-investigators on the clinical trial and provided patient samples. M.C.S., R.L., C.S., K.X., P.J., A.C., R.P. and V.A. performed cfDNA analyses. J.A.F., B.S., C.R.T., A.S., A.K., G.G. and J.M.S. performed computational analysis. M.C.S., K.T., H.E.S. and G.O. performed TCR/organoid reactivity assays. M.C.S., P.A.K. and C.J.W. wrote the manuscript. All authors discussed and interpreted results and edited the manuscript.

Corresponding authors

Correspondence to Panagiotis A. Konstantinopoulos or Catherine J. Wu.

Ethics declarations

Competing interests

M.C.S., J.A.F., B.S., A.V., R.L., C.F., C.R.T., K.T., A.K., S.C., C.S., K.X., P.J., A.C., H.E.S., V.C., W.L., J.T.J., M.R.Y., A.S., S.S., M.H., O.O., H.G., C.C., S.L., N.T., E.H.S., C.K., R.P., G.O. and J.S.M. declare no competing interests. J.K.L. holds equity in Standard BioTools and is on the scientific advisory board for MBQ Pharma. N.H. holds equity in and advises Repertoire Immune Medicines, CytoReason and Danger Bio/Related Sciences, owns equity and has licensed patents to BioNtech, and receives research funding from Bristol Myers Squibb, Moderna, ResolveM/JJDC, Takeda and Calico Life Sciences. U.A.M. declares participation in scientific advisory boards: NextCure, AbbVie, Immunogen, Profound Bio, Eisai, the Ovarian Cancer Research Alliance, Tango Therapeutics, Novartis, GSK, Daiichi Sankyo, DayOne Bio and Whitehawk Therapeutics; U.A.M. also reports participation in a data safety-monitoring board: Mural Oncology, Macrogenics, Daiichi Sankyo, AstraZeneca and Symphogen, all unrelated to this work. V.A.A. is a co-inventor of the MAESTRO MRD test, which has been licensed to Exact Sciences and receives research funding from Exact Sciences, which was not involved in this study. V.A.A. is also a co-founder and advisor to Amplifyer Bio. G.G. receives research funds from IBM, Pharmacyclics/AbbVie, Bayer, Genentech, Calico, Ultima Genomics, Inocras, Google, Kite, AstraZeneca and Novartis and is also an inventor on patent applications filed by the Broad Institute, including those related to MSMuTect, MSMutSig, POLYSOLVER, SignatureAnalyzer-GPU, MSEye, MinimuMM-seq, DLBclass, HapASeg, Tonly2, MSIDetect, TuFEst and ABSOLUTE2. A full list of G.G.’s publicly available patent applications can be requested from the Broad’s legal team at patents@broadinstitute.org. G.G is a founder, consultant, and holds privately held equity in Scorpion Therapeutics; he is also a founder of, and holds privately held equity in Predicta Biosciences; and a consultant and holds privately held equity in Antares Therapeutics. D.B.K. owns equity in Affimed N.V., Agenus, Armata Pharmaceuticals, Breakbio, BioMarin Pharmaceuticals, Celldex Therapeutics, Editas Medicine, Immunitybio, Lexicon Pharmaceuticals, Moderna, PDS Biotechnology, Prime Medicine, Sana Biotechnologies and Viking Therapeutics. D.B.K. is a scientific advisor for Immunitrack, a wholly owned subsidiary of Eli Lilly and Company and Breakbio. P.A.O. has received research funding from and/or has advised Agenus, Arcturus, Amgen, Armo BioSciences, Array, AstraZeneca/MedImmune, Bristol-Meyers Squibb, Celldex, CytomX, Evaxion, Immunetune, Imunon, LGChem, Merck, Neon Therapeutics (now BioNTechUS), Novartis, Pharmajet, Phio, Pfizer, Oncorus, Roche/Genentech, Servier and Xencor. P.A.K. has received grants or contracts from Novartis, AstraZeneca, Pfizer, Eli Lilly, Bayer, Merck, GlaxoSmithKline, Tesaro and Merck KGaA and has received consulting fees for serving on advisory boards from AstraZeneca, Bayer, GlaxoSmithKline, Alkermes, Kadmon, Bristol Myers Squibb, IMV, Repare, Artios, Mersana, Novartis, Cardiff Oncology, Schrodinger, Mural Oncology, Scorpion and Nimbus. C.J.W. is an equity holder of BioNtech and is a scientific advisory board member of Repertoire, Aethon Therapeutics, Nature’s Toolbox and Adventris.

Peer review

Peer review information

Nature Medicine thanks James Gulley and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available. Primary Handling Editor: Ulrike Harjes, in collaboration with the Nature Medicine team.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data

Extended Data Fig. 1 Vaccine selection.

Example of epitope selection for Patients 106 (a) and 108 (b). For each missense mutation, the best personal HLA binding peptide is indicated (far left) with columns indicating if a synthetic long peptide for that gene was chosen for manufacture. Factors taken into account include HLA binding rank, RNA expression level, variant detection in RNA-seq, variant detection in normal exome data, and Cancer cell fraction (CCF). Mutations are classified as green (detected by RNA and WES), yellow (WES detection only), and red (potential artifacts). See Supplementary Tables 1 and 3 for associated source data.

Extended Data Fig. 2 cfDNA parts per million (PPM) and limits of detection.

Tumor cfDNA parts per million (PPM) was calculated at each time point, as well as the cfDNA limit of detection (LOD). Time starts at the first vaccine dose (Day 0).

Source data

Extended Data Fig. 3 Peripheral ex vivo ELISpot responses and TCRβ clonotype expansion after NeoVax and Nivolumab.

a. Ex vivo IFNγ ELISpots with PBMC from the indicated time points post-vaccination, stimulated with the indicated vaccine pools. PBMC were stimulated in triplicate on anti-IFNγ antibody coated plates, and IFNγ producing cells were counted as spots. Data represents mean (+/- S.D) spot forming units (SFU) per 106 PBMC after background subtraction (mean of DMSO control). Immune response induction (*) is defined as (i) > limit of detection (3x the median DMSO SFU across all patients and time points), (ii) 3x > DMSO control and, (iii) P < 0.05 by DFR(EQ)17, a validated statistical method for evaluating ELISpot responses in vaccine trials. b. CD3+ cells were isolated from Leukapheresis samples (if available) at W00 (pre-vaccine), W14 (post-Booster 1, pre-Nivolumab) and W22 (post-Booster 2 and Nivolumab), as well as at late timepoints for P108 and P116), and targeted TCRB RNA sequencing libraries were generated and sequenced via nano-pore long read sequencing; A single sample was sequenced from each timepoint for a given patient. The numbers of significantly expanded TCRB clonotypes after Booster 1 and after Booster 2 + Nivolumab, vs. W00, are graphed (top; one-sided Fisher’s Exact tests). Box and whisker plots (bottom) indicate median (line), 25th-75th percentiles (box) and maxima/minima (whiskers). c. Mean (+/- S.E.M) UMI for significantly expanded TCRB transcripts at the indicated time points; the number of TCR clonotypes contributing to mean UMI at each point (for example n) is shown in (b).

Source data

Extended Data Fig. 4 In vitro elicited vaccine response deconvolution.

a. PBMC from Leukapharesis samples at W00 (pre-vaccine), W14 (post-Booster 1, pre-Nivolumab) and W22 (post-Booster 2 and Nivolumab) were stimulated in vitro for 10 days with vaccine pools, then re-stimulated (in triplicate) with individual IMP peptides and ELISpots were enumerated. Data represents mean (+/- S.D.) spot forming units (SFU) per 106 PBMC for the indicated peptide or pool, after background subtraction (average of DMSO triplicate was subtracted from each peptide replicate). Responses were considered positive if the mean spot count was > 6 (LOD: 3x the median DMSO; equivalent to 30 SFU/106), 3x matched DMSO control, and significant by DFR(EQ) (P < 0.05)17, a validated statistical method for ELISpot interpretation in vaccine trials. Induced immune responses are defined as 2.5x induction over W00. Peptides with immune response are labeled with an asterisk for those time points meeting response definition. b. The number of targeted mutations, total and with an induced immune response, is graphed for each patient (patient symbols match those in Figs. 2 and 3).

Source data

Extended Data Fig. 5 Characterization of TIL and selection of expanded TIL TCR clonotypes.

a. Clustering of all cells from NeoVax patient biopsy scRNA/TCR sequencing (n = 25,103 cells). b. Re-clustering of CD4 and CD8 T cells from NeoVax patient Biopsy scRNA/TCR sequencing. c. Summary of re-clustered tumor infiltrating T cell subsets identified by scRNA-seq. d. Clonotype expansion by exhaustion state. Graphs represent the Log(rank) vs. Log10(frequency) of TCR clonotypes, based on whether they were sequenced from exhausted (TTE1-4; red) or non-exhausted (blue) cells from scRNA/TCR sequencing of the indicated biopsies. Rank and frequency are defined by the frequency of a clonotype amongst all TCRs sequenced within a sample. When red curves are shifted to the right vs. blue, it is consistent with the more expanded clonotypes also being more exhausted. e. For P108, P113 and P116, graphs represent for each TCR clonotype, expansion (total count; left), exhaustion (fraction of that clonotype with exhausted or not, phenotype; middle), and fraction CD4 vs. CD8 lineage (right). f. Projection on the CD8 T cell clustering UMAP of the tested P113 TCR clonotypes that were tumor-reactive (left), vaccine reactive (middle) and TEM TCR clonotypes (right). g. Fraction of the indicated DPP3 p.L558R specific TCR expressing T cells within the indicated scRNA-seq T cell clusters for patient 113.

Extended Data Fig. 6 TIL TCR tumor and vaccine reactivity.

a. Gating strategy for TCR screening. The displayed sample is P113 TIL TCRs incubated with autologous B1 PDO pulsed with IFNγ (Row 3 in the upper heatmap of c). In the CellTrace Far Red by CellTrace Violet plots, numbers within gates indicate tested TCRs. TCR5 is used to provide representative gating for mTCRβ and CD137 expression. b, c & d: Expanded, exhausted (TEX) and effector memory (TEM) TIL TCR from P108 (b) and P113 (c&d) tumor biopsies were cloned, transduced into healthy donor cells, labeled with cell trace dyes and incubated overnight with either patient-derived organoids (PDO) (top) or autologous EBV-LCL organoids incubated with individual vaccine pools (bottom). PDO were pre-stimulated for 24 h with IFNγ as indicated. Reactivity was read out by CD137 staining and the mean CD137 expression from triplicate wells are graphed. TIL TCR phenotype and reactivity calls to IFNγ-stimulated PDO and EBV-LCL are shown in tracks at the bottom (see Methods and Supplementary Table 4; summarized in Fig. 3h, i). Representative of two (P108 and P113 B1 PDO) or three (P113 B2 PDO, EBV-LCL, vaccine pools) experiments. N = no TCR; C = Irrelevant control TCR. d. Testing of TCR2 as in (c) with no TCR, control TCR and TCR5 included for comparison (There were not enough TCR2 transduced cells for analysis in the experiment shown in (c)). e. P113 TCRs 5, 6, 19 and 20 were found to be reactive against EBV-LCL pulsed with vaccine Pool C (c) and TCR reactivity was subsequently tested against the individual Pool C IMP peptides (IMP-14 targets the DPP3 neoantigen p.L658R). Representative of two experiments. f. Four short epitopes from IMP-14 were predicted to bind to the patient’s personal HLA alleles (HLAthena or NetPanHMC4.2 rank < 2.0): LTRRDTVLL (rank 0.793 [NetMHCpan4.2] and 1.14 [HLAthena] for HLA-B*57:01 and 1.957 [NetMHCpan4.2] for HLA-C*03:04), FLTRRDTVL (rank 1.76 [HLAthena] for HLA-C*03:04), LTRRDTVLLR (rank 1.9 [NetMHCpan4.2] for HLA-A*03:01) and TRRDTVLLR (rank 1.79 [NetMHCpan4.2] for HLA-C*06:02). 9-mer predicted binders, along with their paired wild-type sequences, were pulsed on to EBV-LCL cells at the indicated concentrations (1 well per concentration), and reactivity was tested against TCRs 5, 6, 19 and 20 transduced into healthy donor T cells. Peptides were pulsed in three concentrations (10x dilutions) onto EBV-LCL cells. The HLA-B*57:01 restricted LTRRDTVLL was found to be the strongest interacting epitope for all four P113 vaccine reactive TCRs. Representative of two experiments.

Source data

Extended Data Fig. 7 JAK1 structural variant at progression in P113 abrogates IFNγ signaling and decreases antigen presentation.

a. WGS of the post-vaccine B2 biopsy revealed a structural variant in Chromosome 1, leading to inversion of the JAK1 intron 3 to an intergenic region ~ 1.6 Mb away. b. Schematic of the putative inversion event, as well as primers used in (c) to confirm the translocation by amplifying chromosomal break and fusion points: c. PCR amplification of a common JAK1 region (“Common”), the JAK1 break point (JAK1 break), the intergenic break point (“intergenic break”) and the fusion point (“inversion”) (See Supplementary information). Genomic DNA was isolated from two independently derived Pre-Vax PDO lines (B1 and B1b [the latter derived from cryopreserved tumor cell suspension at the time of scRNA-seq preparation (Fig. 3)]), the post-Vax PDO (B2), and autologous EBV-LCL and used for PCR. Note that the donor break points are not identified in the post-Vax B2 PDO, and the inversion fusion point is only identified in the B2 PDO. Representative of two independent experiments (independent gDNA preparations and PCRs). The uncropped gel images (wells to below the ladder [<100 bp]) are shown. d. To confirm the inversion fusion point, the “inversion” PCR product was gel purified and submitted for Sanger sequencing. e. Cell Titer Glow (CTG) viability analysis of pre-vax PDO lines (B1 and B1b) and post-Vax B2 PDO treated for 7 days in triplicate, as indicated, with carboplatin, doxorubicin or both. Viability is normalized to untreated PDO cells (Mean +/- S.D. of triplicate wells). B1 PDO was not tested with combined carboplatin and doxorubcin. Representative of four (carboplatin, carboplatin/doxorubicin) and three (doxorubicin; due to one experiment failure) experiments. f-h. Post-Vax B2 PDO was transduced with a GFP-only Lentiviral vector, or a JAK1-GFP lentiviral vector and assessed for HLA-I and intracellular P-STAT1. f. Flow cytometry gating scheme for PDO staining for HLA-I (left) and intracellular P-STAT1 (right). JAK1 cDNA rescued HLA-I expression (g) and STAT1 phosphorylation (h) following 24 h and 45 min of IFNγ stimulation, respectively, in a dose dependent manner (representative of three (HLA) or two (P-STAT1) independent experiments). i-k. Health donor T cells were transduced with select P113 TIL TCRs, including four vaccine antigen specific TIL (5, 6, 19 and 20; See Fig. 3 and Extended Data Fig. 6), and incubated overnight with Pre-Vax or Post-Vax PDO that had been transduced to express GFP alone or JAK1-GFP. CD137 staining was used to quantify antigen specific tumor PDO recognition. i. Representative flow plots of CD137 induction by T cells transduced with TCR 5 (representative of 2 independent experiments). j. Heat map depicting CD137 upregulation for T cells expressing the indicated TCRs (top; N: no TCR; C: control irrelevant TCR) when incubated with the indicated target cells (representative of 2 independent experiments). k. Data from (j) is graphed as fold CD137 upregulation with pre-vax PDO treated as indicated, normalized to GFP-PDO without IFNγ; Data for post-vax PDO is shown in Fig. 4d.

Source data

Extended Data Fig. 8 HLA loss of heterozygosity in ovarian cancer.

a. Parental allele copy number across chromosome 6 in pre-vaccine biopsies (B1) for all treated NeoVax patients. A vertical bar represents the location of the HLA locus and a horizontal dotted line represents diploid heterozygosity. Asterisks denote patients already presented in Fig. 4e, and presented here for ease of comparison. Patients p111, P117 and P118 exhibit LOH across the HLA locus. b. Parental allele copy number across chromosome 6 in biopsies from 6 patients with available WES, who ultimately did not proceed with vaccination (see Fig. 1b). HLA locus LOH is present in 2 of 6 patients (P107, P110). c. Fraction of Ovarian TCGA (top) and all TCGA (bottom) cases with HLA Allele Specific Expression (ASE)22 loss (left; measured at the RNA level) or HLA locus LOH (right; genomic) is compared for tumors that are HR deficient vs. proficient. HLA-ASE and HLA-LOH calls are taken from Filip et al.22 (of note, only a subset of the overall cohort had matched WES data for LOH identification); HRD/HRP calls are taken from Knijnenburg et al.24. Across the full cohort of TCGA tumors, HRD status is associated with genomic HLA-LOH, but not RNA expression-based HLA-ASE. In contrast, HRP and HRD ovarian tumors are equally likely to exhibit HLA-LOH. d. There was no difference in the rate of HLA-ASE (top)22 or HLA-LOH (bottom)22 in CCNE1 amplified vs. unamplified ovarian tumors25, indicating that in this high need population, HLA-LOH remains an important consideration in antigen-directed therapies. c-d: P values reflect Fisher Exact Test with Benjamini-Hochberg correction; n for each group is represented in the figure.

Source data

Extended Data Fig. 9 HLA Haplotype Loss prior to vaccination.

For P111 (a), P117 (b) and P118 (c), tumor and normal WES VAFs for SNPs differentiating each A, B and C allele are graphed according to location in HLA gene cDNA (left), as well as in aggregate (right). In the event that the expected SNP bases were not identified in WES, the SNP VAF is not graphed. Purple represents the HLA alleles retained in tumor, and white represents the lost alleles. A single replicate of gDNA was sequenced for each patient tumor and normal sample as part of vaccine development.

Source data

Extended Data Fig. 10 cfDNA VAF for vaccine target and non-target missense SNVs.

As in Fig. 4f. (Graphs for P106, P108 and P113 are duplicated here for ease of comparison) cfDNA VAFs at the indicated timepoints after vaccination for vaccine target and non-target missense SNV mutations. Data is graphed as mean +/- S.E (blue: vaccine target mutations, n = 15 [P106], 9 [P108], 6 [P111], 8 [P113], 9 [P114], 13 [P116], 11 [P117], 13 [P118], and 13 [P119]; pink: non-target missense mutations, n = 36 [P106], 26 [P108], 35 [P111], 41 [P113], 32 [P114], 30 [P116], 61 [P117], 104 [P118], and 58 [P119]).

Source data

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Sellars, M.C., Frank, J.A., Shim, B. et al. Personalized neoantigen-targeting cancer vaccines in ovarian cancer: a phase 1 trial. Nat Med (2026). https://doi.org/10.1038/s41591-026-04717-8

Download citation

  • Received: 26 January 2026

  • Accepted: 17 September 2026

  • Published: 14 October 2026

  • Version of record: 14 October 2026

  • DOI: https://doi.org/10.1038/s41591-026-04717-8

来源:Nature 健康护理主题 · nature.com