Abstract
An important challenge in the real-world management of patients with advanced clear-cell renal cell carcinoma (aRCC) is determining who might benefit from immune checkpoint blockade (ICB). Here we performed a comprehensive multiomics mapping of aRCC in the context of ICB treatment, involving discovery analyses in a real-world data cohort followed by validation in independent cohorts. We cross-connected bulk-tumor transcriptomes across >1,000 patients with validations at single-cell and spatial resolutions, revealing a patient-specific crosstalk between proinflammatory tumor-associated macrophages and (pre-)exhausted CD8+ T cells that was distinguished by a human leukocyte antigen repertoire with higher preference for tumoral neoantigens. A cross-omics machine learning pipeline helped derive a new tumor transcriptomic footprint of neoantigen-favoring human leukocyte antigen alleles. This machine learning signature correlated with positive outcome following ICB treatment in both real-world data and independent clinical cohorts. In experiments using the RENCA-tumor mouse model, CD40 agonism combined with PD1 blockade potentiated both proinflammatory tumor-associated macrophages and CD8+ T cells, thereby achieving maximal antitumor efficacy relative to other tested regimens. Thus, we present a new multiomics and spatial map of the immune-community architecture that drives ICB response in patients with aRCC.
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
$29.99 / 30 days
cancel any time
Subscribe to this journal
Receive 12 print issues and online access
$209.00 per year
only $17.42 per issue
Buy this article
- Purchase on Springer Link
- 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
All relevant newly created data for this study (that is, bulk RNA-seq of inhouse immunotherapy-treated RCC patients) are available on Synapse (syn53162048). Due to GDPR sensitivities, the HLA haplotyping data and raw RNA-seq data are not publicly available but can be provided for research purposes upon reasonable request. Other publicly available datasets used are available elsewhere: (1) clinical and transcriptomics data of the JAVELIN 101 cohort as reported in Supplementary Data by Motzer et al.20 (https://doi.org/10.1038/s41591-020-1044-8); (2) survival and transcriptomics data from cohorts included in the TCGA PanCancerAtlas Immune Response Working Group’s Cancer Research Institute (CRI) iAtlas Explorer67; (3) TCGA KIRC data, publicly available from Xena (https://xenabrowser.net/datapages/?cohort=TCGA%20TARGET%20GTEx&removeHub=https%3A%2F%2Fxena.treehouse.gi.ucsc.edu%3A443), FireBrowse portal (a Broad Institute GDAC Firehose analyses pipeline: http://firebrowse.org/?cohort=KIRC); (4) single-cell cohort of Bi et al.15; (5) public spatial transcriptomics data from clear-cell renal cell carcinoma samples through GEO (GSE175540); (6) public transcriptomics data from various murine preclinical tumor models obtained through GEO (GSE85509); (7) single-cell transcriptomics and metadata of public datasets of Krishna et al.42 and Braun et al.14 retrieved from BioTuring database (Talk2Data v. 4, accessed on 17 November 2023); (8) transcriptomics and clinical data from the SuMR trial and SCOTRRCC study39 retrieved from GEO (GSE67818). Source data are provided with this paper.
Code availability
All relevant code for this study is available on Synapse (syn53162048).
References
Motzer, R. J. et al. Molecular subsets in renal cancer determine outcome to checkpoint and angiogenesis blockade. Cancer Cell 38, 803–817.e4 (2020).
Hsieh, J. J. et al. Renal cell carcinoma. Nat. Rev. Dis. Primers 3, 17009 (2017).
Borràs, D. M. et al. Single cell dynamics of tumor specificity vs bystander activity in CD8+ T cells define the diverse immune landscapes in colorectal cancer. Cell Discov. 9, 114 (2023).
Sprooten, J. et al. Peripherally-driven myeloid NFkB and IFN/ISG responses predict malignancy risk, survival, and immunotherapy regime in ovarian cancer. J. Immunother. Cancer 9, e003609 (2021).
Naulaerts, S. et al. Multiomics and spatial mapping characterizes human CD8+ T cell states in cancer. Sci. Transl. Med. 15, eadd1016 (2023).
de Velasco, G. et al. Tumor mutational load and immune parameters across metastatic renal cell carcinoma risk groups. Cancer Immunol. Res. 4, 820–822 (2016).
Alexandrov, L. B. et al. Signatures of mutational processes in human cancer. Nature 500, 415–421 (2013).
Fridman, W. H., Zitvogel, L., Sautès-Fridman, C. & Kroemer, G. The immune contexture in cancer prognosis and treatment. Nat. Rev. Clin. Oncol. 14, 717–734 (2017).
Ross-Macdonald, P. et al. Molecular correlates of response to nivolumab at baseline and on treatment in patients with RCC. J. Immunother. Cancer 9, e001506 (2021).
Cristescu, R. et al. Pan-tumor genomic biomarkers for PD-1 checkpoint blockade-based immunotherapy. Science 362, eaar3593 (2018).
Au, L. et al. Determinants of anti-PD-1 response and resistance in clear cell renal cell carcinoma. Cancer Cell 39, 1497–1518.e11 (2021).
Braun, D. A. et al. Interplay of somatic alterations and immune infiltration modulates response to PD-1 blockade in advanced clear cell renal cell carcinoma. Nat. Med. 26, 909–918 (2020).
Chevrier, S. et al. An immune atlas of clear cell renal cell carcinoma. Cell 169, 736–749.e18 (2017).
Braun, D. A. et al. Progressive immune dysfunction with advancing disease stage in renal cell carcinoma. Cancer Cell 39, 632–648.e8 (2021).
Bi, K. et al. Tumor and immune reprogramming during immunotherapy in advanced renal cell carcinoma. Cancer Cell 39, 649–661.e5 (2021).
Abernethy, A. Time for real-world health data to become routine. Nat. Med. 29, 1317 (2023).
Motzer, R. J. et al. Nivolumab versus everolimus in advanced renal-cell carcinoma. N. Engl. J. Med. 373, 1803–1813 (2015).
Motzer, R. J. et al. Nivolumab plus ipilimumab versus sunitinib in advanced renal-cell carcinoma. N. Engl. J. Med. 378, 1277–1290 (2018).
McDermott, D. F. et al. Clinical activity and molecular correlates of response to atezolizumab alone or in combination with bevacizumab versus sunitinib in renal cell carcinoma. Nat. Med. 24, 749–757 (2018).
Motzer, R. J. et al. Avelumab plus axitinib versus sunitinib in advanced renal cell carcinoma: biomarker analysis of the phase 3 JAVELIN Renal 101 trial. Nat. Med. 26, 1733–1741 (2020).
Beuselinck, B. et al. Sarcomatoid dedifferentiation in metastatic clear cell renal cell carcinoma and outcome on treatment with anti-vascular endothelial growth factor receptor tyrosine kinase inhibitors: a retrospective analysis. Clin. Genitourin. Cancer 12, e205–e214 (2014).
Blum, K. A. et al. Sarcomatoid renal cell carcinoma: biology, natural history and management. Nat. Rev. Urol. 17, 659–678 (2020).
Carretero-González, A. et al. The value of PD-L1 expression as predictive biomarker in metastatic renal cell carcinoma patients: a meta-analysis of randomized clinical trials. Cancers (Basel) 12, 1945 (2020).
Ernst, M. S. et al. Outcomes for International metastatic renal cell carcinoma database consortium prognostic groups in contemporary first-line combination therapies for metastatic renal cell carcinoma. Eur. Urol. 84, 109–116 (2023).
Newman, A. M. et al. Determining cell type abundance and expression from bulk tissues with digital cytometry. Nat. Biotechnol. 37, 773–782 (2019).
Thorsson, V. et al. The immune landscape of cancer. Immunity 48, 812–830.e14 (2018).
Lee, C.-H. et al. High response rate and durability driven by HLA genetic diversity in patients with kidney cancer treated with lenvatinib and pembrolizumab. Mol. Cancer Res. 19, 1510–1521 (2021).
Naranbhai, V. et al. HLA-A*03 and response to immune checkpoint blockade in cancer: an epidemiological biomarker study. Lancet Oncol. 23, 172–184 (2022).
Manczinger, M. et al. Negative trade-off between neoantigen repertoire breadth and the specificity of HLA-I molecules shapes antitumor immunity. Nat. Cancer 2, 950–961 (2021).
Barber, D. L. et al. Restoring function in exhausted CD8 T cells during chronic viral infection. Nature 439, 682–687 (2006).
Li, S. et al. Bioinformatics and in vitro-based comprehensive analysis of EVI2A expression and its immunological and prognostic significance in Kidney Renal Clear Cell Carcinoma. Res. Sq. https://doi.org/10.21203/rs.3.rs-2917863/v1 (2023).
Tan, Q., Liu, H., Xu, J., Mo, Y. & Dai, F. Integrated analysis of tumor-associated macrophage infiltration and prognosis in ovarian cancer. Aging (Albany NY) 13, 23210–23232 (2021).
Brumell, J. H. et al. Expression of the protein kinase C substrate pleckstrin in macrophages: association with phagosomal membranes. J. Immunol. 163, 3388–3395 (1999).
Zakrzewska, A. et al. Macrophage-specific gene functions in Spi1-directed innate immunity. Blood 116, e1–e11 (2010).
MacParland, S. A. et al. Single cell RNA sequencing of human liver reveals distinct intrahepatic macrophage populations. Nat. Commun. 9, 4383 (2018).
Ihim, S. A. et al. Interleukin-18 cytokine in immunity, inflammation, and autoimmunity: biological role in induction, regulation, and treatment. Front. Immunol. 13, 919973 (2022).
Miao, D. et al. Genomic correlates of response to immune checkpoint therapies in clear cell renal cell carcinoma. Science 359, 801–806 (2018).
Choueiri, T. K. et al. Immunomodulatory activity of nivolumab in metastatic renal cell carcinoma. Clin. Cancer Res. 22, 5461–5471 (2016).
Stewart, G. D. et al. Sunitinib treatment exacerbates intratumoral heterogeneity in metastatic renal cancer. Clin. Cancer Res. 21, 4212–4223 (2015).
Turajlic, S. et al. Tracking cancer evolution reveals constrained routes to metastases: tracerx renal. Cell 173, 581–594.e12 (2018).
Laureano, R. S. et al. The cell stress and immunity cycle in cancer: toward next generation of cancer immunotherapy. Immunol. Rev. 321, 71–93 (2024).
Krishna, C. et al. Single-cell sequencing links multiregional immune landscapes and tissue-resident T cells in ccRCC to tumor topology and therapy efficacy. Cancer Cell 39, 662–677.e6 (2021).
Jin, S. et al. Inference and analysis of cell-cell communication using CellChat. Nat. Commun. 12, 1088 (2021).
Melero, I., Hirschhorn-Cymerman, D., Morales-Kastresana, A., Sanmamed, M. F. & Wolchok, J. D. Agonist antibodies to TNFR molecules that costimulate T and NK cells. Clin. Cancer Res. 19, 1044–1053 (2013).
Meylan, M. et al. Tertiary lymphoid structures generate and propagate anti-tumor antibody-producing plasma cells in renal cell cancer. Immunity 55, 527–541.e5 (2022).
Bolognesi, M. M. et al. Multiplex staining by sequential immunostaining and antibody removal on routine tissue sections. J. Histochem. Cytochem. 65, 431–444 (2017).
Philip, M. & Schietinger, A. CD8+ T cell differentiation and dysfunction in cancer. Nat. Rev. Immunol. 22, 209–223 (2022).
Larionova, I. et al. Tumor-associated macrophages in human breast, colorectal, lung, ovarian and prostate cancers. Front. Oncol. 10, 566511 (2020).
Vanmeerbeek, I. et al. Early memory differentiation and cell death resistance in T cells predicts melanoma response to sequential anti-CTLA4 and anti-PD1 immunotherapy. Genes Immun. 22, 108–119 (2021).
Sprooten, J. et al. Lymph node and tumor-associated PD-L1+ macrophages antagonize dendritic cell vaccines by suppressing CD8+ T cells. Cell Rep. Med. 5, 101377 (2024).
Vanmeerbeek, I., Naulaerts, S. & Garg, A. D. Reverse translation: the key to increasing the clinical success of immunotherapy? Genes Immun. 24, 217–219 (2023).
Mosely, S. I. S. et al. Rational selection of syngeneic preclinical tumor models for immunotherapeutic drug discovery. Cancer Immunol. Res. 5, 29–41 (2017).
Murgaski, A. et al. Efficacy of CD40 agonists is mediated by distinct cDC subsets and subverted by suppressive macrophages. Cancer Res. 82, 3785–3801 (2022).
Chauvin, J.-M. & Zarour, H. M. TIGIT in cancer immunotherapy. J. Immunother. Cancer 8, e000957 (2020).
Huo, J.-L., Wang, Y.-T., Fu, W.-J., Lu, N. & Liu, Z.-S. The promising immune checkpoint LAG-3 in cancer immunotherapy: from basic research to clinical application. Front. Immunol. 13, 956090 (2022).
Vanmeerbeek, I. et al. The interface of tumour-associated macrophages with dying cancer cells in immuno-oncology. Cells 11, 3890 (2022).
Motzer, R. J. et al. Biomarker analysis from CheckMate 214: nivolumab plus ipilimumab versus sunitinib in renal cell carcinoma. J. Immunother. Cancer 10, e004316 (2022).
Seymour, L. et al. iRECIST: guidelines for response criteria for use in trials testing immunotherapeutics. Lancet Oncol. 18, e143–e152 (2017).
Heng, D. Y. C. et al. Prognostic factors for overall survival in patients with metastatic renal cell carcinoma treated with vascular endothelial growth factor-targeted agents: results from a large, multicenter study. J. Clin. Oncol. 27, 5794–5799 (2009).
Chowell, D. et al. Evolutionary divergence of HLA class I genotype impacts efficacy of cancer immunotherapy. Nat. Med. 25, 1715–1720 (2019).
Sidney, J., Peters, B., Frahm, N., Brander, C. & Sette, A. HLA class I supertypes: a revised and updated classification. BMC Immunol. 9, 1 (2008).
Roussel, E. et al. Molecular subtypes and gene expression signatures as prognostic features in fully resected clear cell renal cell carcinoma: a tailored approach to adjuvant trials. Clin. Genitourin. Cancer 19, e382–e394 (2021).
Johnson, W. E., Li, C. & Rabinovic, A. Adjusting batch effects in microarray expression data using empirical Bayes methods. Biostatistics 8, 118–127 (2007).
Zhang, Y., Parmigiani, G. & Johnson, W. E. ComBat-seq: batch effect adjustment for RNA-seq count data. NAR Genom. Bioinform. 2, lqaa078 (2020).
Roussel, E. et al. Molecular underpinnings of glandular tropism in metastatic clear cell renal cell carcinoma: therapeutic implications. Acta Oncol. 60, 1499–1506 (2021).
Goldman, M. J. et al. Visualizing and interpreting cancer genomics data via the Xena platform. Nat. Biotechnol. 38, 675–678 (2020).
Eddy, J. A. et al. CRI iAtlas: an interactive portal for immuno-oncology research. F1000Res. 9, 1028 (2020).
Jiang, P. et al. Signatures of T cell dysfunction and exclusion predict cancer immunotherapy response. Nat. Med. 24, 1550–1558 (2018).
Vita, R. et al. The Immune Epitope Database (IEDB): 2018 update. Nucleic Acids Res. 47, D339–D343 (2019).
Vano, Y.-A. et al. Nivolumab, nivolumab-ipilimumab, and VEGFR-tyrosine kinase inhibitors as first-line treatment for metastatic clear-cell renal cell carcinoma (BIONIKK): a biomarker-driven, open-label, non-comparative, randomised, phase 2 trial. Lancet Oncol. 23, 612–624 (2022).
Kleshchevnikov, V. et al. Cell2location maps fine-grained cell types in spatial transcriptomics. Nat. Biotechnol. 40, 661–671 (2022).
Hunter, J. D. Matplotlib: a 2D graphics environment. Comput. Sci. Eng. 9, 90–95 (2007).
Waskom, M. seaborn: statistical data visualization. JOSS 6, 3021 (2021).
Kuleshov, M. V. et al. Enrichr: a comprehensive gene set enrichment analysis web server 2016 update. Nucleic Acids Res. 44, W90–W97 (2016).
Antoranz, A. et al. Mapping the immune landscape in metastatic melanoma reveals localized cell-cell interactions that predict immunotherapy response. Cancer Res. 82, 3275–3290 (2022).
Kask, P., Palo, K., Hinnah, C. & Pommerencke, T. Flat field correction for high-throughput imaging of fluorescent samples. J. Microsc. 263, 328–340 (2016).
Reddy, B. S. & Chatterji, B. N. An FFT-based technique for translation, rotation, and scale-invariant image registration. IEEE Trans. Image Process. 5, 1266–1271 (1996).
Weigert, M., Schmidt, U., Haase, R., Sugawara, K. & Myers, G. Star-convex polyhedra for 3D object detection and segmentation in microscopy. In Proc. 2020 IEEE Winter Conference on Applications of Computer Vision (WACV) 3655–3662 (IEEE, 2020); https://doi.org/10.1109/WACV45572.2020.9093435
Caicedo, J. C. et al. Data-analysis strategies for image-based cell profiling. Nat. Methods 14, 849–863 (2017).
Motzer, R. J. et al. Nivolumab versus everolimus in patients with advanced renal cell carcinoma: Updated results with long-term follow-up of the randomized, open-label, phase 3 CheckMate 025 trial. Cancer 126, 4156–4167 (2020).
Motzer, R. J. et al. Conditional survival and long-term efficacy with nivolumab plus ipilimumab versus sunitinib in patients with advanced renal cell carcinoma. Cancer 128, 2085–2097 (2022).
Rini, B. I. et al. Pembrolizumab plus axitinib versus sunitinib as first-line therapy for advanced clear cell renal cell carcinoma: 5-year analysis of KEYNOTE-426. J. Clin. Oncol. 41, LBA4501 (2023).
Burotto, M. et al. Nivolumab plus cabozantinib vs sunitinib for first-line treatment of advanced renal cell carcinoma (aRCC): 3-year follow-up from the phase 3 CheckMate 9ER trial. J. Clin. Oncol. 41, 603 (2023).
Motzer, R. J. et al. Nivolumab plus cabozantinib versus sunitinib in first-line treatment for advanced renal cell carcinoma (CheckMate 9ER): long-term follow-up results from an open-label, randomised, phase 3 trial. Lancet Oncol. 23, 888–898 (2022).
Choueiri, T. K. et al. Lenvatinib plus pembrolizumab versus sunitinib as first-line treatment of patients with advanced renal cell carcinoma (CLEAR): extended follow-up from the phase 3, randomised, open-label study. Lancet Oncol. 24, 228–238 (2023).
Motzer, R. J. et al. Final prespecified overall survival (OS) analysis of CLEAR: 4-year follow-up of lenvatinib plus pembrolizumab (L+P) vs sunitinib (S) in patients (pts) with advanced renal cell carcinoma (aRCC). J. Clin. Oncol. 41, 4502 (2023).
Rody, A. et al. T-cell metagene predicts a favorable prognosis in estrogen receptor-negative and HER2-positive breast cancers. Breast Cancer Res. 11, R15 (2009).
Rody, A. et al. A clinically relevant gene signature in triple negative and basal-like breast cancer. Breast Cancer Res. 13, R97 (2011).
Fan, C. et al. Building prognostic models for breast cancer patients using clinical variables and hundreds of gene expression signatures. BMC Med. Genomics 4, 3 (2011).
Beck, A. H. et al. The macrophage colony-stimulating factor 1 response signature in breast carcinoma. Clin. Cancer Res. 15, 778–787 (2009).
Wolf, D. M., Lenburg, M. E., Yau, C., Boudreau, A. & van ’t Veer, L. J. Gene co-expression modules as clinically relevant hallmarks of breast cancer diversity. PLoS ONE 9, e88309 (2014).
Zhang, W. et al. Inhibition of respiratory syncytial virus infection with intranasal siRNA nanoparticles targeting the viral NS1 gene. Nat. Med. 11, 233 (2005).
Acknowledgements
We thank Y. Vano (Hôpital Européen Georges Pompidou, Université de Paris Cité, France) and M. Meylan (Centre de Recherche des Cordeliers, INSERM, Université de Paris Cité, Sorbonne Université, Paris, France) for providing outcome data associated with the Visium spatial transcriptomics samples from BIONIKK. This study is supported by Kom op tegen Kanker (Stand up to Cancer), the Flemish cancer society via Emmanuel van der Schueren (EvDS) PhD fellowship (projectID: 3328; to L.K.), Research Foundation Flanders (FWO) (Fundamental Research Grant, G0B4620N to A.D.G.; Excellence of Science/EOS grant, 30837538, for ‘DECODE’ consortium, to A.D.G.), KU Leuven (C1 grant, C14/19/098 to A.D.G.; C3 grant, C3/21/037 or C3/23/067 to A.D.G.), and VLIR-UOS (iBOF grant, iBOF/21/048, for ‘MIMICRY’ consortium to A.D.G.). I.V. and R.S.L. are supported by FWO-SB PhD Fellowship (1S06821N to I.V. and 1S44123N to R.S.L.). J.S. is funded by KU Leuven’s Postdoctoral Mandate (PDM) fellowship (PDMT2/23/071). S.N. is funded by Stichting tegen Kanker Postdoctoral fellowship (2023-046). B. Beuselinck is supported by FWO Senior Clinical Investigator Fellowship (1801520N). F.F. was supported by the Austrian Science Fund (FWF) (no. T 974-B30 and FG 2500-B) and by the Oesterreichische Nationalbank (OeNB) (no. 18496). The computational results presented here have been achieved in part using the LEO HPC infrastructure of the University of Innsbruck. The results shown here are in part based upon data generated by TCGA Research Network (https://cancergenome.nih.gov).
Author information
Authors and Affiliations
Contributions
L.K. was involved in sample collection and RNA extraction, performed the bulk- and scRNA-seq bioinformatics analyses, coordinated and managed the research efforts, created the figures and cowrote the paper. S.N. designed the ML model and MILAN interaction analysis workflow, performed the VISIUM analyses and provided bioinformatical guidance and critical data interpretation and statistics discussion. J.G., I.V., J.S., R.S.L. performed the cell culture, in vivo experiments, and flow cytometry analysis, and helped with paper revision and figure preparation. N.D. performed the MILAN staining and image acquisition. G.S. performed the MILAN image analysis. F.M.B. coordinated the MILAN pipeline. E.R., A.V., B. Beuselinck, L.K., M.B., M. Albersen and A.W. helped with sample collection of the Leuven cohort. E.R., A.V. and L.K. performed bulk RNA extraction. M.B. performed pathological review of samples. D.L. and B. Boeckx helped with bulk RNA sequencing and DNA extraction of the Leuven cohort. L.B. helped with antibody analyses. M. Ausserhofer and F.F. performed the genomic analysis on the TCGA KIRC dataset. J.K. helped with HLA analyses. J.Z.-R. helped with bulk RNA sequencing of the Leuven cohort. M. Albersen, A.V., E.R., L.B., J.K., G.S., B. Boeckx, D.L., F.F. and B. Beuselinck revised the paper. B. Beuselinck performed patient recruitment and clinical data collection of the Leuven cohort, as well as critical data interpretation and paper revision. A.D.G. was the lead investigator of the project, conceptualized the overall project, supervised the research design, and cowrote the paper.
Corresponding authors
Ethics declarations
Competing interests
B. Beuselinck: speaker’s bureau: BMS, Pfizer, MSD, Ipsen; unrestricted research grants: BMS; advisory board: BMS, MSD, Ipsen. A.D.G. received consulting/advisory/lecture honoraria from Boehringer Ingelheim (Germany), Miltenyi Biotec (Germany), Novigenix (Switzerland), SOTIO (Czech Republic) and IsoPlexis (United States) and received R&D project funding from SOTIO. F.F. consults for iOnctura. The other authors declare no competing interests.
Peer review
Peer review information
Nature Medicine thanks A. Ari Hakimi, Jiyang Yu and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. 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 Positioning and sample availability of Leuven RWD cohort.
Positioning of Leuven RWD cohort relative to ICB-treated arms of relevant phase III clinical RCC trials and specific validation cohorts used in this study. Data shown are as per latest report (before August 2023) from the Checkmate025 trial (CM025, nivolumab arm)80, Checkmate214 (CM214, ipilimumab-nivolumab arm)81, Keynote-426 (KN-426, axitinib-pembrolizumab arm)82, Checkmate-9ER (CM9ER, cabozantinib-nivolumab arm)83,84, CLEAR (lenvatinib-pembrolizumab arm)85,86, Javelin101 (avelumab-axitinib arm, PFS and ORR as reported in Motzer et al.20), Immotion150 (atezolizumab±bevacizumab arms; data as reported in CRI iAtlas survival data67), Miao et al. (data as reported in CRI iAtlas survival data), Choueiri et al.38 (data as reported in CRI iAtlas survival data). Colours are indicating the ICB type used in the study, brown represents a mixed cohort of both anti-PD1 and anti-PD1/anti-CTLA4. N numbers represent number of patients. a, Bar chart showing median progression-free survival (mPFS) with 95% confidence interval as error bars. mPFS is indicated above the bar. b, Bar chart showing median overall survival (mOS) with 95% confidence interval as error bars. Cohorts with median OS not reached were omitted (that is, Javelin101, Immotion150). mOS is indicated above the bar. c, Stacked bar chart showing proportion of categories of best response (complete response (CR), partial response (PR), non-responder (NR)). d, Bar chart showing median follow-up time. In case the upper 95% CI value was ‘not reached’, the upper side of the error bar is omitted. e, Flowchart indicating sample availability or loss in the Leuven RWD cohort (created with Biorender.com).
Extended Data Fig. 2 Antigenic TAMs-T cell hub as the core of ICB-driven clinical benefit.
a, b, Forest plots showing HR (centre of error bar) with 95%CI (error bar) and two-sided p-values of multivariate [MVA, adjusted for age and IMDC risk group (IMDC adjusted for age only)] Cox proportional hazards models correlating biomarkers with PFS (a) and OS (b) after ICB, with n as indicated on figure for continuous (cont.) and categorical (cat.) variables. P values unadjusted for multiple testing. c, Boxplot comparing expression of biomarkers in responders (R; that is complete/partial response (CR/PR), n = 32 patients) vs. non-responders (NR; that is stable/progressive disease (SD/PD), n = 62 patients), with two-sided P-values from Wilcoxon’s tests (uncorrected for multiple testing). d, Stacked bar chart showing categorical biomarkers by responder status (p value as calculated by two-sided Fisher’s Exact test, n patients as indicated on figure). e, Schematic overview of standard bioinformatics biomarker mining approach (created with Biorender.com). f, g, Forestplots showing HR (centre of error bar) with 95%CI (error bar) and two-sided p-values of UVA Cox proportional hazard regression models correlating deconvoluted immune cell densities (CIBERSORTx) with PFS (f) and OS (g), n = 98 independent patient samples. HRs were log-transformed for visual representation. P values unadjusted for multiple testing. h, Boxplots of deconvoluted immune cell fractions by responder status. Two-sided p-values by Wilcoxon’s test, unadjusted for multiple testing. i, Forestplot displaying HR (centre of error bar), 95% CI (error bar) and two-sided p-values as calculated through MVA Cox proportional hazard regression model including age, sex, IMDC risk groups, ICB treatment, Fuhrman grade and sarcomatoid differentiation. Covariates significantly correlated to OS in this MVA model are highlighted in bold. P values uncorrected for multiple testing. For boxplots in c and h, boxes represent median (centre) and first/third quartile (bottom and top, resp.) values; whiskers show most extreme values within 1,5x interquartile range (IQR). Outliers extending beyond 1.5x IQR above/below the median are plotted individually. Non-significant results are abbreviated as ns.
Extended Data Fig. 3 Immune landscape clusters.
a-d, Forest plots displaying HR (centre of error bar), 95%CI (error bar) and two-sided p-values as calculated through either univariate (UVA) or MVA (adjusted for age and IMDC risk group) Cox proportional hazard regression models of whole patients’ cohort and subgroup analysis per treatment type (with n representing number of patients per group), for PFS (a-b) and OS (c-d). P values are not corrected for multiple testing. e, f, Bar chart representing distribution of Fuhrman tumour grade (e) and sarcomatoid differentiation (f) stratified by ILS clusters. Two-sided p-value as calculated by Fisher’s Exact test. g, Co-expression network based on TCGA KIRC and further based on 95 genes associated with worse ORR, PFS and OS. Only spearman correlations with coefficient > 0.7 are shown. Node size represents betweenness.
Extended Data Fig. 4 Quantitative HLA metrics fail to predict ICB outcome in Leuven cohort.
a, Bar chart representing overall response rate (ORR), showing complete response (CR), partial response (PR), stable disease (SD), progressive disease (PD) and not evaluable (NA), by HLA class I expression (dichotomized by optimal statistical cutoff for overall survival (OS), also for panel b and c). Two-sided p-value as calculated by Fisher Exact test. b-c, Kaplan-Meier curves showing progression-free survival (PFS) (b) and OS (c) stratified by HLA class I expression. d, Bar chart representing ORR by HLA class II expression (dichotomized by optimal statistical cutoff for OS, also for panel e and f). Two-sided p-value as calculated by Fisher Exact test. e, f, Kaplan-Meier curve showing PFS (e), and OS (f) stratified by HLA class II expression. g, Bar chart representing ORR by HLA heterozygosity (dichotomized by <= 10 vs. 11-12, also for panel h and I). Two-sided p-value as calculated by Fisher Exact test. h, i, Kaplan-Meier curve showing PFS (h), and OS (i) stratified by HLA heterozygosity. j, Bar chart representing ORR by HLA A*03 allele carrier status. Two-sided p-value as calculated by Fisher Exact test. k-l, Kaplan-Meier curve showing PFS (k), and OS (l) stratified by HLA A*03 allele carrier status. m, Bar chart representing ORR by HLA evolutionary divergence (HED), dichotomized by highest quartile (also for panel n and o). Two-sided p-value as calculated by Fisher Exact test. n-o, Kaplan-Meier curve showing PFS (n), and OS (o) stratified by HLA HED. Hazard ratio (HR), 95% confidence interval and two-sided p-values as calculated by Cox proportional hazard models, both univariate (UVA) as well as multivariate (MVA) adjusting for age and IMDC risk group.
Extended Data Fig. 5 HLA promiscuity (HLApr).
a, Bar chart representing overall response rate (ORR) (complete response: CR, partial response: PR, stable disease: SD, progressive disease: PD, not evaluable: NA), stratified by HLApr (dichotomized by optimal statistical cutoff determined on Leuven RWD cohort, also for panel b, c, d, e and f). Two-sided p-value as calculated by Fisher’s Exact test. b, c, Kaplan-Meier curve showing progression-free survival (PFS) (b) and overall survival (OS) (c) stratified by HLApr. d, Bar chart representing ORR stratified by HLApr. Two-sided p-value as calculated by Fisher’s Exact test. e, f, Kaplan-Meier curve showing PFS (e), and OS (f) stratified by HLApr. Hazard ratio (HR), 95% confidence interval and two-sides p-values as calculated by Cox proportional hazard models, both univariate (UVA) as well as multivariate (MVA) adjusting for age and IMDC risk group. g, h, Bar chart representing distribution of Fuhrman tumour grade (g) and sarcomatoid differentiation (h) stratified by HLA promiscuity. Two-sided p-value as calculated by Fisher’s Exact test. i, Scatter pie plot. Pies represent HLA alleles, with x-axis position representing difference in ORR in patients with HLA allele vs. patients without, and y-axis position representing allele promiscuity value. Pie chart represents types of antigens presented by a particular allele, wherein the size of the pie-chart represents number of HLA-antigen pairs. j, Scatter pie plot. Pies represent HLA alleles, with x-axis position representing HR as calculated by UVA Cox proportional hazard regression model with PFS after start of ICB, and y-axis position representing allele promiscuity value. Pie chart represents types of antigens presented by a particular allele, wherein the size of the pie-chart represents number of HLA-antigen pairs.
Extended Data Fig. 6 Tumour transcriptomic footprint of HLAprLOW.
a, Averaged feature importance (Gini) of top 100 genes in ML-model b, Kaplan-Meier curves of PFS in Leuven cohort, by tLHP. c, ORR stratified by tLHP. Two-sided p-value by Fisher’s Exact test. d, e, Forestplot displaying UVA correlation of tLHP, as continuous signature and by optimal cutoff, with PFS (d) and OS (e) in Leuven cohort. f-i, Forestplots displaying UVA and MVA (adjusted for age/IMDC) correlation of tLHP with PFS (f–g) and OS (h-i) within Leuven cohort treatment subgroups. j, Boxplots of tLHP by IMDC (n independent patients). Two-sided p-value by Kruskal-Wallis test. k-l, Fuhrman tumour grade (k) and sarcomatoid differentiation (l) stratified by tLHP. Two-sided p-value by Fisher’s Exact test. m, Violinplot showing tLHP expression in cells by disease stage. Two-sided p-value by one-way ANOVA, effect size by Eta-squared. n, Venn diagram showing tLHP and ILS overlap. o, ORR stratified by tLHP. Two-sided p-value by Fisher’s Exact test. p, Forestplot displaying UVA correlation of tLHP with PFS in IMmotion150 subgroups. q, Kaplan-Meier curves showing PFS in Javelin101 sunitinib arm by tLHP. r, Forestplot displaying UVA correlation of tLHP (optimal statistical cutoff in combined cohort) with OS by cancer type in CRI iAtlas. s, Forestplot displaying correlation of tLHP (optimal statistical cutoff in combined cohort) with OS in CRI iAtlas melanoma subgroup. Ipilimumab+pembrolizumab treated patients are not displayed as the statistical model could not be constructed due to insufficient events. For b and q, HR, 95%CI and two-sided p-values by Cox proportional hazard regression (high vs. low) from UVA and MVA [adjusted for age/IMDC (Leuven RWD cohort) or age/sex (Javelin101 cohort)]. For boxplots in j and m, boxes represent median (centre) and first/third quartile (bottom and top, resp.) values; whiskers show extreme values within 1,5x interquartile range (IQR). Outliers extending beyond 1.5x IQR above/below median are plotted individually. Forestplots in d-i, p, r and s show HR, 95%CI (centre, error bar, resp.) and p-values as calculated through Cox proportional hazard regression (high vs. low). P-values unadjusted for multiple testing.
Extended Data Fig. 7 Genomic and transcriptomic characterization of the tLHP signature.
a, Heatmap showing tLHP expression in paired untreated and post-treatment samples. Rows represent individual patients and are hierarchically clustered by complete method. For multiple samples per treatment group, mean tLHP is displayed. b, Boxplots showing tLHP expression in paired untreated and post-treatment samples. P value as calculated with Kruskal-Wallis test. N numbers represent individual samples. c, Violinplots showing tLHP expression in treatment-naïve vs. post-VEGFR-TKI treated samples. Two-sided p-value as calculated by Mann-Whitney U test. d-f, Boxplots showing log10(TMB + 1) in TCGA-KIRC (d), Javelin101 (e) and IMmotion150 (f), by tLHP signature (dichotomized by optimal statistical cutoff). Two-sided p-value by Mann-Whitney U test. N numbers represent independent samples. g-i, Heatmap showing gene mutation status by sample in TCGA-KIRC (g), Javelin101 (h) and IMmotion150 (i). P values by Fisher’s Exact test are FDR-corrected. j, Boxplot showing TCR richness by tLHP signature (dichotomized by optimal statistical cutoff). Two-sided p-value as calculated by Mann-Whitney U test. N numbers represent individual samples. k, l, Dotplots showing correlation of CIBERSORTx cell types with tLHP signature in Leuven RWD cohort (k) and TCGA-KIRC (l) (Pearson or Spearman correlation depending on normality of cell type estimates. Two-sided p-values are FDR-corrected). m-q, Violinplot showing expression of tLHP signature by responder status for dendritic cells (m), monocytes (n), tumour cell type 2 (o), tumour cell type 1 (p) and regulatory T cells (q). N numbers represent independent cells. Two-sided p-values were calculated by Kolmogorov-Smirnov test. Effect sizes by Cohen’s D. r, Violin plot showing expression of tLHP signature per cell type (n = 152876 independent cells). s-u, Violinplot showing expression of tLHP signature in cells from untreated vs. ICB-treated patients, for all cells (s), TAM (t) and CD8+ T cells (u). N numbers represent independent cells. Two-sided p-values were calculated by Kolmogorov-Smirnov test, effect sizes by Cohen’s D. For all boxplots in this figure, boxes represent median (centre) and first/third quartile (bottom and top, resp.) values; whiskers show most extreme values within 1,5x interquartile range (IQR). Outliers extending beyond 1.5x IQR above/below the median are plotted individually.
Extended Data Fig. 8 Antigenicity-relevant TAM-CD8+ T cell interactions associate with ICB benefit in aRCC.
a, Barchart showing relative information flow along pathways between single cells from responder and non-responder patients, as determined through CellChat. Pathways relevant to co-stimulation (CD86, CD70), immune-inhibitory receptors relevant for the exhaustion state (BTLA, PDL2, PDL1, PVR, TIGIT, CD39), T cell activation (CD226, CD96, OX40, LCK), cytokine signalling (XCR, CSF, CXCL), complement signalling (CD46, COMPLEMENT) and antigen presentation (MHC-I, MHC-II) are highlighted in bold. b, Graph representation of correlation (Spearman) per spot between inferred cell population and normalized tLHP expression in ICB non-responder patients. Node colour visualizes graph betweenness centrality, while node size illustrates degree. Edge width represents Spearman correlation coefficient (shown at threshold > 0.48). c, Barplot representing enriched pathways in non-responders using Gene Ontology Biological Processes. Combined scores were used for visual representation (that is natural log of the p-value multiplied by the z-score, where the z-score is the deviation from the expected rank), after manual curation for immune-related pathways. P-values are adjusted for multiple testing with FDR.
Extended Data Fig. 9 Cell type identification through multiplex immunohistochemistry (MILAN).
a, Dotplot showing expression of cell type markers across final annotated cell types. b, Boxplots showing number of interactions of pre-exhausted CD8+ T cells with other cell types. N numbers indicate independent cells per HLApr/ICB-response group. Effect size metric was calculated as Cohen’s D (positive values indicating higher in HLAprLOW/ICB-responders) and two-sided p-values shown are extracted through Kolmogorov-Smirnov tests (multiple testing corrected with false discovery rate with Benjamini-Hochberg method). c, Boxplots showing number of interactions of exhausted CD8+ T cells with other cell types. N numbers indicate independent cells per HLApr/ICB-response group. Effect size metric is shown by Cohen’s D (positive values indicating higher in HLAprLOW/ICB-responders) and two-sided p-values shown are extracted through Kolmogorov-Smirnov tests (multiple testing corrected with false discovery rate). For all boxplots in this figure, boxes represent median (centre) and first/third quartile (bottom and top, resp.) values; whiskers show most extreme values within 1,5x interquartile range (IQR). Outliers extending beyond 1.5x IQR above/below the median were omitted from visualisation.
Extended Data Fig. 10 Immunophenotyping of RENCA tumors.
a, Violin plot of ratio of CD4+ T cells (CD4+CD3+) to tumour area in RENCA tumours after different treatments. P values are from Kruskal-Wallis tests and are not corrected for multiple testing. b, Violin plots of mature classic dendritic cells (DC) type 1 (cDC1) (MHC-II+CD86+ of XCR1+CD11b+CD11c+) as percentage of DCs in RENCA tumours after different treatments. P values are from Kruskal-Wallis tests and are not corrected for multiple testing. c, Violin plots of mature classic DC type 2 (cDC2) (MHCII+CD86+ of CD172a+CD11b+CD11c+) as percentage of DCs in RENCA tumours after different. P values are from Kruskal-Wallis tests and are not corrected for multiple testing. d, Heatmap showing correlation coefficient of Spearman correlation of cell populations with inverse of tumour area (at day 21) in RENCA tumours (columns and rows are clustering using hierarchical clustering with complete method). Displayed values are scaled by treatment type.
Supplementary information
Supplementary Information
Supplementary Note, Figs. 1 and 2 and Data 1, 4 and 6.
Supplementary Table 1
Supplementary Data 2, 3, 5 and 7.
Supplementary Data
Statistical source data.
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 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. 8
Statistical source data.
Source Data Extended Data Fig. 9
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
Kinget, L., Naulaerts, S., Govaerts, J. et al. A spatial architecture-embedding HLA signature to predict clinical response to immunotherapy in renal cell carcinoma. Nat Med (2024). https://doi.org/10.1038/s41591-024-02978-9
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41591-024-02978-9