Abstract
Cell therapy with tumor-infiltrating lymphocytes (TILs) has yielded durable responses for multiple cancer types, but the causes of therapeutic resistance remain largely unknown. Here multidimensional analysis was performed on time-serial tumor and blood in a lung cancer TIL therapy trial. Using T cell receptor sequencing on both functionally expanded T cells and neoantigen-loaded tetramer-sorted T cells, we identified tumor antigen-specific T cell receptors. We then mapped clones into individual transcriptomes and found that tumor-reactive clonotypes expressed a dysfunctional program and lacked stem-like features among patients who lacked clinical benefit. Tracking tumor-reactive clonotypes over time, decay of antigen-reactive peripheral T cell clonotypes was associated with the emergence of progressive disease. Further, subclonal neoantigens previously targeted by infused T cells were subsequently absent within tumors at progression, suggesting potential adaptive resistance. Our findings suggest that targeting clonal antigens and circumventing dysfunctional states may be important for conferring clinical responses to TIL therapy.
This is a preview of subscription content, access via your institution
Access options
Access Nature and 54 other Nature Portfolio journals
Get Nature+, our best-value online-access subscription
$32.99 / 30 days
cancel any time
Subscribe to this journal
Receive 12 digital issues and online access to articles
$119.00 per year
only $9.92 per issue
Buy this article
- Purchase on SpringerLink
- Instant access to full article PDF
Prices may be subject to local taxes which are calculated during checkout
Similar content being viewed by others
Data availability
Genomic data containing the tumor whole-exome sequencing and RNA-seq raw data files have been uploaded for public use at the NIH dbGAP under accession code phs002486.v2.p1. TCRVβ sequencing data derived from infused TIL, time-serial tumors and peripheral blood can be accessed at NIH Gene Expression Omnibus under accession code GSE291406. Proteomics data for tumors have been made available through the ProteomeXchange via the PRIDE database under accession code PXD049438. Single-cell sequencing data for patients’ TILs have been made available at the NIH Gene Expression Omnibus under accession code GSE255549. Source data for Figures and Extended Data Figures have been provided as Source Data files, with each.xls file labeled by figure. Source data are provided with this paper.
Code availability
No unique code was developed for this study.
Change history
10 June 2025
In the version of the article initially published, in the Data availability section, the GEO accession code GSE255549 was correctly listed but the attached hyperlink was incorrect and has now been amended in the HTML and PDF versions of the article.
References
Gadgeel, S. et al. Updated analysis from KEYNOTE-189: pembrolizumab or placebo plus pemetrexed and platinum for previously untreated metastatic nonsquamous non-small-cell lung cancer. J. Clin. Oncol. 38, 1505–1517 (2020).
Goff, S. L. et al. Randomized, prospective evaluation comparing intensity of lymphodepletion before adoptive transfer of tumor-infiltrating lymphocytes for patients with metastatic melanoma. J. Clin. Oncol. 34, 2389–2397 (2016).
Creelan, B. C. et al. Tumor-infiltrating lymphocyte treatment for anti-PD-1-resistant metastatic lung cancer: a phase 1 trial. Nat. Med. 27, 1410–1418 (2021).
Rohaan, M. W., van den Berg, J. H., Kvistborg, P. & Haanen, J. B. A. G. Adoptive transfer of tumor-infiltrating lymphocytes in melanoma: a viable treatment option. J. Immunother. Cancer 6, 102 (2018).
Shah, N. N. & Fry, T. J. Mechanisms of resistance to CAR T cell therapy. Nat. Rev. Clin. Oncol. 16, 372–385 (2019).
Sade-Feldman, M. et al. Resistance to checkpoint blockade therapy through inactivation of antigen presentation. Nat. Commun. 8, 1136 (2017).
Restifo, N. P., Smyth, M. J. & Snyder, A. Acquired resistance to immunotherapy and future challenges. Nat. Rev. Cancer 16, 121–126 (2016).
Winkler, J., Abisoye-Ogunniyan, A., Metcalf, K. J. & Werb, Z. Concepts of extracellular matrix remodelling in tumour progression and metastasis. Nat. Commun. 11, 5120 (2020).
Lauss, M. et al. Mutational and putative neoantigen load predict clinical benefit of adoptive T cell therapy in melanoma. Nat. Commun. 8, 1738 (2017).
Kristensen, N. P. et al. Neoantigen-reactive CD8+ T cells affect clinical outcome of adoptive cell therapy with tumor-infiltrating lymphocytes in melanoma. J. Clin. Invest. 132, e150535 (2022).
Krishna, S. et al. Stem-like CD8 T cells mediate response of adoptive cell immunotherapy against human cancer. Science 370, 1328–1334 (2020).
Chen, B., Khodadoust, M. S., Liu, C. L., Newman, A. M. & Alizadeh, A. A. in Cancer Syst. Biol. 243–259 (Springer, 2018).
Danilova, L. et al. The mutation-associated neoantigen functional expansion of specific T cells (MANAFEST) assay: a sensitive platform for monitoring antitumor immunity. Cancer Immunol. Res. 6, 888–899 (2018).
Andreatta, M. et al. Semi-supervised integration of single-cell transcriptomics data. Nat. Commun. 15, 872 (2024).
Andreatta, M. et al. Interpretation of T cell states from single-cell transcriptomics data using reference atlases. Nat. Commun. 12, 2965 (2021).
Chen, G. M. et al. Integrative bulk and single-cell profiling of premanufacture T-cell populations reveals factors mediating long-term persistence of CAR T-cell therapy. Cancer Discov. 11, 2186–2199 (2021).
Zheng, L. et al. Pan-cancer single-cell landscape of tumor-infiltrating T cells. Science 374, abe6474 (2021).
Chu, Y. et al. Pan-cancer T cell atlas links a cellular stress response state to immunotherapy resistance. Nat. Med. 29, 1550–1562 (2023).
Kienzler, A., Hargreaves, C. & Patel, S. The role of genomics in common variable immunodeficiency disorders. Clin. Exp. Immunol. 188, 326–332 (2017).
Caushi, J. X. et al. Transcriptional programs of neoantigen-specific TIL in anti-PD-1-treated lung cancers. Nature 596, 126–132 (2021).
Kaech, S. M., Hemby, S., Kersh, E. & Ahmed, R. Molecular and functional profiling of memory CD8 T cell differentiation. Cell 111, 837–851 (2002).
Luckey, C. J. et al. Memory T and memory B cells share a transcriptional program of self-renewal with long-term hematopoietic stem cells. Proc. Natl Acad. Sci. USA 103, 3304–3309 (2006).
Chevalier, N. et al. CXCR5 expressing human central memory CD4 T cells and their relevance for humoral immune responses. J. Immunol. 186, 5556–5568 (2011).
Alban, T. J. et al. Neoantigen immunogenicity landscapes and evolution of tumor ecosystems during immunotherapy with nivolumab. Nat. Med. 30, 3209–3222 (2024).
Gao, J. et al. Loss of IFN-γ pathway genes in tumor cells as a mechanism of resistance to anti-CTLA-4 therapy. Cell 167, 397–404 e399 (2016).
Dersh, D. et al. Genome-wide screens identify lineage- and tumor-specific genes modulating MHC-I- and MHC-II-restricted immunosurveillance of human lymphomas. Immunity 54, 116–131 e110 (2021).
McGranahan, N. et al. Allele-specific HLA loss and immune escape in lung cancer evolution. Cell 171, 1259–1271 e1211 (2017).
Roth, A. et al. PyClone: statistical inference of clonal population structure in cancer. Nat. Methods 11, 396–398 (2014).
Adzhubei, I. A. et al. A method and server for predicting damaging missense mutations. Nat. Methods 7, 248–249 (2010).
Parkhurst, M. R. et al. Unique neoantigens arise from somatic mutations in patients with gastrointestinal cancers. Cancer Discov. 9, 1022–1035 (2019).
Lu, Y. C. et al. Efficient identification of mutated cancer antigens recognized by T cells associated with durable tumor regressions. Clin. Cancer Res. 20, 3401–3410 (2014).
McLane, L. M., Abdel-Hakeem, M. S. & Wherry, E. J. CD8 T cell exhaustion during chronic viral infection and cancer. Ann. Rev. Immunol. 37, 457–495 (2019).
Biasco, L. et al. Clonal expansion of T memory stem cells determines early anti-leukemic responses and long-term CAR T cell persistence in patients. Nat. Cancer 2, 629–642 (2021).
Duhen, T. et al. Co-expression of CD39 and CD103 identifies tumor-reactive CD8 T cells in human solid tumors. Nat. Commun. 9, 2724 (2018).
Bremnes, R. M. et al. High-throughput tissue microarray analysis used to evaluate biology and prognostic significance of the E-cadherin pathway in non-small-cell lung cancer. J. Clin. Oncol. 20, 2417–2428 (2002).
Floc’h, A. L. et al. αEβ7 integrin interaction with E-cadherin promotes antitumor CTL activity by triggering lytic granule polarization and exocytosis. J. Exp. Med. 204, 559–570 (2007).
Kitakaze, M. et al. Cancer-specific tissue-resident memory T-cells express ZNF683 in colorectal cancer. Br. J. Cancer 128, 1828–1837 (2023).
Qiu, M. Z. et al. Dynamic single-cell mapping unveils Epstein‒Barr virus-imprinted T-cell exhaustion and on-treatment response. Signal Transduct. Target Ther. 8, 370 (2023).
Lu, Y. C. et al. Single-cell transcriptome analysis reveals gene signatures associated with T-cell persistence following adoptive cell therapy. Cancer Immunol. Res. 7, 1824–1836 (2019).
Hall, M. S. et al. Neoantigen-specific CD4(+) tumor-infiltrating lymphocytes are potent effectors identified within adoptive cell therapy products for metastatic melanoma patients. J. Immunother. Cancer 11, e007288 (2023).
Landsberg, J. et al. Melanomas resist T-cell therapy through inflammation-induced reversible dedifferentiation. Nature 490, 412–416 (2012).
Jensen, S. M. et al. Increased frequency of suppressive regulatory T cells and T cell-mediated antigen loss results in murine melanoma recurrence. J. Immunol. 189, 767–776 (2012).
Verdegaal, E. M. et al. Neoantigen landscape dynamics during human melanoma-T cell interactions. Nature 536, 91–95 (2016).
Rosenthal, R. et al. Neoantigen-directed immune escape in lung cancer evolution. Nature 567, 479–485 (2019).
Schrors, B. et al. HLA class I loss in metachronous metastases prevents continuous T cell recognition of mutated neoantigens in a human melanoma model. Oncotarget 8, 28312–28327 (2017).
Jamal-Hanjani, M. et al. An open-label, multicenter phase I/IIa study evaluating the safety and clinical activity of clonal neoantigen reactive T cells in patients with advanced non-small cell lung cancer (CHIRON). J. Clin. Oncol. https://doi.org/10.1200/JCO.2021.39.15_suppl.TPS9138 (2021).
Hong, D. S. et al. Autologous T cell therapy for MAGE-A4+ solid cancers in HLA-A* 02+ patients: a phase 1 trial. Nat. Med. 29, 104–114 (2023).
Parkhurst, M. et al. Isolation of T-cell receptors specifically reactive with mutated tumor-associated antigens from tumor-infiltrating lymphocytes based on CD137 expression. Clin. Cancer Res. 23, 2491–2505 (2017).
Becker, C. et al. Adoptive tumor therapy with T lymphocytes enriched through an IFN-γ capture assay. Nat. Med. 7, 1159–1162 (2001).
Zacharakis, N. et al. Breast cancers are immunogenic: immunologic analyses and a phase II pilot clinical trial using mutation-reactive autologous lymphocytes. J. Clin. Oncol. https://doi.org/10.1200/jco.21.02170 (2022).
Kim, S. P. et al. Adoptive cellular therapy with autologous tumor-infiltrating lymphocytes and T-cell receptor-engineered T cells targeting common p53 neoantigens in human solid tumors. Cancer Immunol. Res. 10, 932–946 (2022).
Blumenthal, G. et al. Analysis of time-to-treatment discontinuation of targeted therapy, immunotherapy, and chemotherapy in clinical trials of patients with non-small-cell lung cancer. Ann. Oncol.30, 830–838 (2019).
Boyle, T. A. et al. A community-based lung cancer rapid tissue donation protocol provides high-quality drug-resistant specimens for proteogenomic analyses. Cancer Med. 9, 225–237 (2020).
Degasperi, A. et al. A practical framework and online tool for mutational signature analyses show inter-tissue variation and driver dependencies. Nat. Cancer 1, 249–263 (2020).
Favero, F. et al. Sequenza: allele-specific copy number and mutation profiles from tumor sequencing data. Ann. Oncol. 26, 64–70 (2015).
Stewart, P. A. et al. Proteogenomic landscape of squamous cell lung cancer. Nat. Commun. 10, 3578 (2019).
Welsh, E. A., Eschrich, S. A., Berglund, A. E. & Fenstermacher, D. A. Iterative rank-order normalization of gene expression microarray data. BMC Bioinform. 14, 153 (2013).
Harro, C. M. et al. Sezary syndrome originates from heavily mutated hematopoietic progenitors. Blood Adv. 7, 5586–5602 (2023).
Sharma, R. & Sharma, S. Physiology, Blood Volume (StatPearls Publishing, 2023).
Acknowledgements
This work was supported by a Catalyst Award Clinical Trial grant through the Stand Up 2 Cancer Foundation (SU2C-AACR-CT04-17) to E.B.H. and S.J.A.; Barbara Bauer Prelude to a Cure Foundation grant to C.W. and B.C.C.; ER Squibb & Sons for nivolumab supply to B.C.C.; an Iovance Biotherapeutics research grant to B.C.C.; Clinigen Group for aldesleukin supply to B.C.C.; a National Cancer Institute (NCI) Cancer Center support grant (P30-CA076292) to E.B.H.; and Clinic and Laboratory Integration Program grants from the Cancer Research Institute to B.C.C., and a grant from Troper Wojnicki Philanthropies to B.C.C. Stand Up 2 Cancer is a program of the Entertainment Industry Foundation administered by the American Association for Cancer Research. The companies played no other role in the study or report. We thank the patients and their families who participated in this research; C. Zhang, L. Zhang, A. Smith and T. Mesa at the Moffitt Molecular Genomics Facility; G. Zhang for assistance with clinical specimens; and other members of our clinical and laboratory research teams. Cell sorting for this project was performed on instruments in the Moffitt Flow Cytometry Core; Sequencing was performed by the Moffitt Molecular Genomics and Proteomics & Metabolomics Core Facilities. This work has been supported in part by the Biostatistics & Bioinformatics Shared Resource, Cell Therapies Core, Flow Cytometry Core, Molecular Genomics Core, Proteomics & Metabolomics, and Tissue Core at the H. Lee Moffitt Cancer Center & Research Institute, a comprehensive cancer center designated by the NCI and funded in part by Moffitt’s Cancer Center support grant (P30-CA076292). We also thank N. Riaz of Memorial Sloan Kettering Cancer Center, W. Geese of Bristol-Myers Squibb and Checkmate 153 coauthors for providing the exon mutation dataset on time-serial tumors from an anti-PD-1 cohort. We recognize crucial funding support from R. Staley and J. Goetze of Florida.
Author information
Authors and Affiliations
Contributions
C.W.: conceptualization, data curation, formal analysis, supervision, funding acquisition, validation, investigation, visualization, methodology, writing–original draft, writing–review and editing. X.Y.: software, formal analysis, supervision, investigation, visualization, methodology, writing–original draft, writing–review and editing. J.K.T.: software, formal analysis, supervision, investigation, visualization, methodology, writing–original draft, writing–review and editing. J.Y.: software, formal analysis, investigation, methodology. D.D.: software, formal analysis, investigation, methodology, writing–review and editing. X.L.: software, formal analysis, investigation, methodology. Z.T.: software, formal analysis, investigation, visualization, methodology. M.W.: data curation, methodology. E.W.: software, formal analysis, methodology, writing–original draft, writing–review and editing. D.M.: investigation, methodology, writing–review and editing. T.C.: Data curation, methodology, writing–review and editing. V.M.: data curation, methodology. C.A.: methodology, writing–review and editing. L.S.: data curation, formal analysis, investigation, methodology. T.B.: formal analysis, supervision, methodology. B.F.: data curation, formal analysis, investigation, methodology, writing–original draft, writing–review and editing. J.M.K.: data curation, formal analysis, supervision, investigation, methodology, writing–original draft, writing–review and editing. C.C.: supervision, investigation, methodology. A.M.L.: data curation, investigation, methodology, writing–review and editing. S.Y.: data curation, methodology, writing–original draft. S.K.: supervision, investigation, methodology, writing–review and editing. D.C.: formal analysis, supervision, investigation, methodology, writing–review and editing. S.A.P.T.: supervision, investigation, writing–review and editing. J.R.C.G.: supervision, investigation, writing–review and editing. S.J.A.: conceptualization, supervision, funding acquisition, investigation, writing–review and editing. E.B.H.: conceptualization, supervision, funding acquisition, investigation, writing–review and editing. B.C.C.: conceptualization, data curation, formal analysis, supervision, funding acquisition, validation, investigation, visualization, methodology, writing–original draft, writing–review and editing.
Corresponding authors
Ethics declarations
Competing interests
B.C.C. has received speaking fees from AstraZeneca and Hoffmann-La Roche and consultant fees; Regeneron, AbbVie, Iovance Biotherapeutics, DAVA Oncology, MJH Lifesciences, Bayer Pharma, Boehringer Ingelheim, Johnson & Johnson, CDR-Life, Achilles, E.R. Squibb and AstraZeneca. B.C.C., C.W., S.J.A. and E.B.H have patent applications filed (US201962865697P, US202062976867P and US202263415674P) related to use of neoantigen-specific T cells, outside the submitted work. J.R.C.G. received consultant fees from Alloy Theraputics, has intellectual property (IP) with Compass Therapeutics and Anixa Biosciences; and is co-founder of Cellepus Therapeutics. E.B.H. has research funding from Revolution Medicines and consultant fees with Amgen, Janssen, Ellipses, Kanaph Therapeutics and ORI Capital II, outside the submitted work. S.K. reports nonfinancial and research financial support from BMS and AstraZeneca. T.C. has patent applications outside the submitted work related to determinants of response to immunotherapy in cancer: US20210308241A1, US20160326597A1, US20200232040A1, US20140113286A1 and US20190092864A1. A.M.L. received financial support from Iovance Biotherapeutics. J.M.K. has received research funding from Bristol-Myers Squibb outside of the submitted work. Moffitt Cancer Center has licensed IP related to the proliferation and expansion of TILs to Iovance Biotherapeutics. Moffitt has also licensed IP to Tuhura Biopharma. S.A.P.T. is an inventor on such IP. S.A.P.T. is listed as a co-inventor on a patent application with Provectus Biopharmaceuticals and participates in sponsored research agreements with Provectus Biopharmaceuticals, Iovance Biotherapeutics, Intellia Therapeutics, Dyve Biosciences, Turnstone Biologics and Celgene, which are not related to this research. S.A.P.T. has also received research support that is not related to this research from the following entities: NIH-NCI (U01 CA244100-01, R01 CA259387, R43 CA257552-01A1 and R01 CA239219-01A1), DOD, Swim Across America and V Foundation; and has received consulting fees from Seagen, Morphogenesis and KSQ Therapeutics. J.K.T. has grant support for this work through P30-CA76292 and SU2C and has a patent awarded for Large Data Set Negative Information Storage Model. J.K.T. has also been funded through grants from Turnstone Biologics outside the submitted work. S.J.A. has received advisor fees from Achilles, Amgen, AstraZeneca, E.R. Squibb, Caris Life Sciences, Celsius Therapeutics, G1 Therapeutics, GlaxoSmithKline, Memgen, Merck & Co., Nektar Therapeutics, RAPT Therapeutics, Venn Therapeutics, Glympse, EMD Serano and Samyang Biopharm USA and research funding from Cellular Biomedicine Group. The remaining authors declare no competing interests.
Peer review
Peer review information
Nature Cancer thanks the anonymous reviewers for their contribution to the peer review of this work.
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 Transcriptomic and proteomic features associated with TIL clinical benefits.
a-b, Heatmap showing differentially expressed genes (a) and proteins (b) in pretreatment tumors comparing CB (n = 6) versus NB (n = 10) patients. Differentially expressed genes and proteins with fold-difference > 1.5 & p value < 0.05 (two-sided) were selected for KEGG pathway enrichment. Top 3 enriched pathways and their contained gene and protein expression intensities were plotted. Differentially expressed tumor antigens, T cell activation markers, and antigen presentation–associated genes/proteins are shown on the right. Tumor source and histology are classified on the top. c, Localization of CD3+ T cell infiltrate in the baseline tumor. Formalin-fixed baseline tumor samples were sectioned for multiplex immunohistochemistry. Nine regions of interest per tumor were selected for automated scoring to determine T cell counts in stroma vs tumor based on cytokeratin colocalization. Upper panel, T cell counts in tumor (brown) or stroma (purple) regions between Clinical Benefit vs No-Benefit patients. Lower panel, Ratios of stroma T cell counts to tumor T cell counts between Clinical Benefit vs No-Benefit patients. Each symbol indicates one region from the indicated patient. Two-sided Mann–Whitney U-test was used. ns, not significant. d, Imputed fractional abundance of tumor-infiltrating immune subtypes using CIBERSORTx algorithm were shown in CB (n = 6 tumor samples) versus NB (n = 8 tumor samples). Error bars indicate mean ± SEM. One-sided two-sample t-test with unequal variance was used. P = 0.03 for macrophages M1, while not significant for other subtypes. e, Identification of T cell recognized tumor antigens in TIL-treated NSCLC patients. Both TSA and MAGE antigens were screened. The number of TIL-recognized TA (Left), MAGE antigens (middle) and TSA (right) were plotted between Clinical Benefit (CB, n = 6) vs. No-Benefit (NB, n = 10). TA, Tumor Antigen; TSA, Tumor-Specific Antigen. Two-sided Mann–Whitney test was performed for the comparison. P = 0.03 for TA, P = 0.04 for MAGE, P = 0.13 for TSA. f, Synthesized neoantigen peptides for tetramer staining according to their affinity to available Flex-T tetramers (HLA-A*11:01 and HLA-A*03:01). One or two mutated peptides with the lowest “affinity_mut(nM)” were selected for each mutation site. g, Synthesized peptides were loaded on tetramers for staining validation by flow cytometry. Only loaded tetramers with a clear positive staining population were selected for further analysis.
Extended Data Fig. 2 Identification of neoantigen-specific TCRs for PT03 and PT31.
a, Schematics showing neoantigen identification pipeline in a NSCLC TIL cohort. neoAg, neoantigen. b, An example showing a representative ELISpot data to identify T cell recognized neoantigens. Positive controls: CD3 Ab, T cells + 1 μg/mL anti-CD3; Viral peptide, T cells + APCs + 1 μg/mL CEF peptides. Negative controls: T cells only; T cells + APCs only. Orange columns show positive neoantigens that can be recognized by T cells; black columns show negative neoantigens that cannot be recognized by T cells. Bars indicate mean ± SD. Shown 2-sided p value calculated by repeated measures ANOVA with Dunnett’s multiple comparisons test. n = 3 biological replicates. APCs, antigen presenting cells. c-d, Identification of neoantigen-specific TCRs by FEST assay on neoantigen activated T cell expansion for PT03. In brief, post-TIL peripheral T cells were sorted and co-cultured with APCs loading neoantigen peptides. After 10 days of co-cultured, T cell expansion were enriched and sequenced (c). Compared to “T cells + APCs” controls, expansion fold for each TCR clonotype was shown at each condition (d). e, Identification of TIL-recognized neoantigens by IFN-γ ELISpot assay for PT31. Orange columns show positive neoantigens that can be recognized by T cells. Bars indicate mean ± SD. Shown two-sided p value calculated by repeated measures ANOVA with Dunnett’s multiple comparisons test. n = 3 biological replicates. f, Identification of TIL-recognized neoantigens by 41BB upregulation for PT31. PT31 TIL were co-cultured with dendritic cells loading 10 μg/mL neoantigen peptides in vitro for 24 hrs. Then co-cultures were stained with anti-41BB and anti-CD8 to detect the proportions of 41BB+CD8+ TILs. NEG peptides, negative neoantigen peptides (FOXF2D254A). g, Identification of TIL-recognized neoantigens by multiplex assays for PT31. Supernatants of PT31 TIL co-cultures in f were collected and multiplex assays was performed on Ella. Bars indicate mean ± SD. Shown two-sided p value calculated by repeated measures ANOVA with Dunnett’s multiple comparisons test. n = 3 technical replicates. h-i, Identification of neoantigen-specific CDR3s by FEST assay on neoantigen activated TIL expansion for PT31. Workflow schematics was shown in h. Compared to “T cells + APCs” controls, expansion fold for each TCR clonotype was shown at each condition. j-k, Identification of neoantigen-specific CDR3s by sequencing neoantigen-loaded tetramer positive T cells for PT03. A NUP133C119Y neoantigen epitope (DEGGWACLVY) loaded HLA-B*18:01 tetramer was requested from NIH Tetramer Core Facility. A MEX3DE477Q epitope (ATIWAPFQR) loaded HLA-A*11:01 Flex-TTM tetramer was constructed according to manufacture’s instructions. TILs were incubated with loaded tetramers, together with surface lineage markers. Tetramer staining positive TILs were sorted and their TCRs were sequenced. Frequencies of TCRs between tetramer positive and negative populations were compared and only those highly enriched in tetramer positive populations were identified to be neoantigen-specific TCRs. l-m, Identification of neoantigen-specific CDR3s by sequencing neoantigen-loaded tetramer positive T cells for PT31. A KRASG12V epitope (VVGAVGVGK) and a TP53G244C epitope (SSCMCGMNR) loaded HLA-A*11:01 Flex-TTM tetramer was constructed. Tetramer staining positive TILs were sorted and their TCRs were sequenced, and labeled as neoantigen-specific as above.
Extended Data Fig. 3 Identification of tumor-reactive TCRs for PT05, 07, 14 and 16.
a-d, Identification of tumor antigen-specific TCRVβ clonotypes by FEST assay on activated T cell expansion for PT05 (a), PT07 (b), PT14 (c) and PT16 (d). TIL (PT05 and PT16) or peripheral T cells (PT07 and PT14) were co-cultured with APCs loading neoantigen peptides. After 10 days of co-culture, TIL expansion was enriched and sequenced. Compared to “T cells + APCs” controls, expansion fold for each TCR clonotype was shown at each condition. CD4 or CD8 origin of each clonotype was labeled. ND: not detectable in CDR3Vβ sequencing data of flow sorted CD4+ or CD8+ TIL.
Extended Data Fig. 4 TIL reactivity against viral epitopes.
a, TIL reactivity against CMV/EBV/Flu viral epitopes determined by IFN-γ ELISpot assay. Positive control: TIL + anti-CD3 (OKT3). Negative controls: TIL, TIL + APC. Bars indicate mean ± SD. n = 3 biological replicates each for 16 patients total. b, Frequencies of viral-associated CDRVβ sequences in infused TIL. TIL CDRVβ sequences were compared with public virus reactive CDRVβ database (VDJdb) to identify the frequency of virus reactive T cells for each of 16 patients. Two-sided Mann–Whitney U-test was used. ns, not significant.
Extended Data Fig. 5 Quality control for single-cell sequencing analysis on tumor-reactive TILs.
a, Batch effect quality control of single-cell RNA sequencing of TIL for each patient. nFeature_RNA: the number of genes; nCount_RNA: the number of reads; percent.mt: percentage of reads mapped to mitochondrial genes. b, A UMAP plot showing analyzed TIL cells colored by patients. c, Metrics to evaluate single-cell sequencing data integration quality. LISI, Local Inverse Simpson’s Index; iLISI, integration LISI; cLISI, cell type LISI; CiLISI, cell type integration LISI; ASW, Average Sihouette Width. d, Marker plots and QC for major cell types. e, UMAP projections of 52,830 T cells from 8 total TIL samples. f-g, Nonscaled gene expression UMAP for each marker was shown in CD4+TIL (f) and CD8+TIL (g), with their corresponding cell annotation UMAPs presented on the top left in each panel. Abbreviations: Tm: memory T; Tcytotoxic: cytotoxic T; Tstemlike: stem-like T; Teff: effector T; Tem: effector memory T; Tex: exhausted T; Tprolif: proliferating T; Tfh: follicular helper T; Treg: regulatory T.
Extended Data Fig. 6 Functional subset annotation of tumor-reactive TILs by single-cell sequencing analysis.
a, Signature gene expression of each functional subset for CD4+ and CD8+ TIL. b, Comparison of the utilized annotation method with a historical control. Red, matched or most similar to; gray: unmatched. c, Enrichment of CD4/CD8 T cell signature gene sets among functional subsets. d, Comparison of annotation markers in this manuscript with those previously reported in other T cell studies. e, Composition of TIL clones among annotated functional subsets for each patient. PT# denotes patient ID. f, Frequencies of tumor-reactive and bulk TIL clones among annotated functional subsets. g, Comparison of tumor-reactive TIL (left) and bulk TIL (right) phenotype in Clinical Benefit vs No-Benefit patients. Abbreviations: Tm: memory T; Tcytotoxic: cytotoxic T; Tstemlike: stem-like T; Teff: effector T; Tem: effector memory T; Tex: exhausted T; Tprolif: proliferating T; Tfh: follicular helper T; Treg: regulatory T. h-i, UMAP was generated separately for CD4+ TIL (h) and CD8+ TIL (i). Functional cluster annotation was performed (left). Tumor-reactive clones and their composition among annotated subsets were shown (right).
Extended Data Fig. 7 Comparison of transcriptomic features between patient subgroups and flow cytometry validation of differentially expressed markers.
a, Comparing tumor-reactive TIL transcriptomic features between CB vs NB patients, differentially expressed genes were selected by using a non-biased selection criteria: P < 0.05 & fold change ≥2× for CD8 T cells and P < 0.05 & fold change ≥1.7× for CD4 T cells. Transcriptomic expression of those genes in bulk CD4+ and CD8+ TIL was plotted for a side-by-side comparison. ns: not significant between CB vs NB patients. b, Tetramers loading MEX3D & NUP133 neoAg epitopes for PT03 and TP53 & KRAS neoAg epitopes for PT31 were used to gate neoantigen-specific TIL. Stained differentially expressed markers were labeled on the bottom. The number on the upper right corner of each histogram shows the median fluorescence intensity for the indicated marker. Corresponding FMOs are shown in the top row. #, CD8 BV605 was used instead of CD8 FITC for detecting perforin expression. c, Flow cytometry analysis detecting surface CD39/CD69 expression in CD4+ and CD8+ TIL. Two-sided Mann–Whitney U-test was performed to compare Clinical Benefit (n = 6 TIL samples) vs No-Benefit groups (n = 10 TIL samples) with no statistical significance. Bars indicate mean ± SD. Gating strategy was shown in Supplementary Information. d, Single-cell transcriptomic analysis showing expression of Tscm and Tex signature genes in tumor-reactive CD8+TIL between Clinical Benefit (CB) vs No-Benefit (NB) groups. e, Gene Set Enrichment Analysis (GSEA) raw data for Fig. 2f.
Extended Data Fig. 8 Tracking TIL-derived T cell clones in baseline and post-treatment progressive tumors.
a, Cumulative frequency of infused TIL clonotypes in tumors from Clinical Benefit (n = 6 for baseline tumors, n = 5 for post-treatment tumors) vs. No-Benefit patients (n = 10 for baseline tumors, n = 7 for progressive tumors). b, Frequency of tumor-reactive T cell clones in time-serial tumors from Clinical Benefit (n = 4 tumors with identified TIL-recognized tumor antigens) vs. No-Benefit patients (n = 2 tumors with identified TIL-recognized tumor antigens). One-sided Mann–Whitney U-test was performed for the comparison. ns, not significant. Bars indicate median with range. c, Composition of different functional T cell subsets among KRASG12V/TP53G244C neoantigen-specific TIL vs other tumor antigen-specific TIL vs non-specific TIL. d, Transcriptomic expression of CD8 T cell subset signatures comparing KRASG12V/TP53G244C neoantigen-specific TIL vs non-specific TIL. The bar value shows fold change for each gene. e, Peripheral tracking of tumor-reactive T cell clones over time. Tumor-reactive TCRVβ clonotypes were determined using both FEST assay and HLA multimer sorting. Absolute numbers of tumor-reactive T cells at day 0 (TIL infusion) were inferred using infused TIL number divided by estimated blood volume.
Extended Data Fig. 9 Single-cell RNA sequencing analysis on T cells from baseline and progressive tumors.
a, Preparation procedure for single-cell sequencing on neoantigen-specific T cells from PT03 baseline and progressive tumors. CD103+ T cells were enriched using magnetic sorting from tumor digest. CD3+103+ T cells were then sorted and encapsulated for single-cell RNA-seq paired with TCR VDJ seq. To compare transcriptomic features of T cells reactive to antigens at baseline vs progression, tumor-reactive TCRs were determined using both FEST and tetramer-sorting methods. b, UMAP analysis shows T cell functional clusters for baseline (left) and progressive (right) tumor CD103+T cells. Overall 2781 T cells from baseline tumor and 6985 T cells from progressive tumor passed QC criteria for transcriptomic analysis. c, Distribution of T cell clones reactive to baseline (left) vs progressive tumor neoAgs (right). d, Violin plots comparing transcriptomic expression of markers linked to memory (KLF2, LEF1, IL7R), exhausted (CD69, ITGAE, LAG3, TIGIT) and effector (GZMB) T cells reactive to antigens at baseline vs progression. Baseline tumor neoAgs include those five only expressed in the baseline tumor: ASPMI1250S, NUP133C119Y, MEX3DE477Q, SMAD4D537E, BIRC6V1860I. Progressive tumor neoAgs are the two retained neoAgs: STIP1S45F, TUBBT238I.
Extended Data Fig. 10 Cellular prevalence of imputed clonal and subclonal neoantigens before versus after TIL therapy.
a, Surgical or core biopsies of tumor sites were shown at baseline and tumor progression. The bar on the right of each figure shows composition of each type of mutations corresponding to Fig. 6a. Organ sites are identified based on anatomic site on the representative body map for each patient. (7 total patients, 32 total tumor blocks). b, Each dot indicates the cellular prevalence of a TIL-reactive neoantigen derived from one tumor sample. Numbers of TIL-recognized neoantigens for each cluster were shown. Green bars show cellular frequency of the corresponding clonal neoantigen cluster, while purple bars show that of the subclonal neoantigen clusters. The mean ± SD is shown based on the sample size of total mutations within each cluster. BL: Baseline tumor; PD: Progressive tumor. For PT31, PD#1: progressive tumor biopsy; PD#2: autopsy liver metastasis; PD#3: autopsy lung metastasis; PD#4: autopsy lymph node metastasis. In total, PT03: n = 5 tumors, PT05: n = 4 tumors, PT07: n = 4 tumors, PT09, n = 5 tumors, PT14: n = 4 tumors, PT25: n = 5 tumors, Pt31, n = 5 tumors.
Supplementary information
Supplementary Information
Supplementary Fig. 1.
Supplementary Tables
Supplementary Table 1: Radiographic Outcomes of TIL therapy for NSCLC. Supplementary Table 2: Experimental assays performed by subject ID. Supplementary Table 3: Tumor mutation changes after TIL therapy. Supplementary Table 4: Tumor mutation changes after anti-PD-1 therapy. Supplementary Table 5: HLA loss of heterogeneity analysis for tumors. Supplementary Table 6: T cell subset markers used in single-cell sequencing analysis and annotation.
Source data
Source Data Fig. 1
Statistical Source Data.
Source Data Fig. 2
Statistical Source Data.
Source Data Fig. 3
Statistical Source Data.
Source Data Fig. 4
Statistical Source Data.
Source Data Fig. 5
Statistical Source Data.
Source Data Fig. 6
Statistical Source Data.
Source Data Fig. 7
Statistical Source Data.
Source Data Fig. 8
Statistical Source Data.
Source Data Extended Data Fig. 1
Statistical Source Data.
Source Data Extended Data Fig. 2
Statistical Source Data.
Source Data Extended Data Fig. 3
Statistical Source Data.
Source Data Extended Data Fig. 4
Statistical Source Data.
Source Data Extended Data Fig. 5
Statistical Source Data.
Source Data Extended Data Fig. 6
Statistical Source Data.
Source Data Extended Data Fig. 7
Statistical Source Data.
Source Data Extended Data Fig. 7
Image Data.
Source Data Extended Data Fig. 8
Statistical Source Data.
Source Data Extended Data Fig. 10
Statistical 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.
About this article
Cite this article
Wang, C., Yu, X., Teer, J.K. et al. Impaired T cell and neoantigen retention in time-serial analysis of metastatic non-small cell lung cancer in patients unresponsive to TIL cell therapy. Nat Cancer 6, 801–819 (2025). https://doi.org/10.1038/s43018-025-00946-x
Received:
Accepted:
Published:
Issue date:
DOI: https://doi.org/10.1038/s43018-025-00946-x
