Computational biomarker predicts lung ICI response via deep learning-driven hierarchical spatial modelling from H&E
Open accessAbstract
Determining which lung cancer patients are likely to respond to immune checkpoint inhibitors (ICI) remains a crucial challenge. Existing FDA-approved biomarkers lack sensitivity and specificity for identifying treatment candidates. To overcome this problem, we present a computational biomarker for predicting ICI response directly from routine H&E stained whole slide images of the initial biopsy. To achieve this, we developed an end-to-end deep learning system (EPL-GNN) that performs hierarchical spatial modeling on whole slide images to learn both spatial and morphological features from 2.1 billion cells and output a response score for each patient. The computational biomarker was trained and evaluated on the largest reported cohort of stage 4 lung cancer patients with ICI treatment response (N=583), resulting in an AUC of 0.69 and sensitivity of 91% on the independent test cohort, which compares favorably to PD-L1 immunohistochemistry (IHC) with an AUC of 0.68 and sensitivity of 57%, and tumor mutation burden (TMB) with an AUC of 0.62. The EPL-GNN model correctly identified 81% of the patients with a negative PD-L1 IHC result as responders. Visualizations of the hierarchical spatial model revealed potential cellular patterns that correspond to ICI treatment response. In addition to the increased sensitivity achieved by the EPL-GNN model, H&E-based Computational Biomarkers offer a faster, less expensive, more objective and reproducible alternative or adjunct to existing IHC or sequencing based biomarkers.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
There is evidence that descriptors of local cellular networks that better model the tumor microenvironment can achieve AUCs of around 0.7 for ICI response prediction from H&E slides, on par with PD-L1 IHC, the current clinical gold standard.
“ICI response prediction from H&E slides is a challenging task, yet there is evidence that descriptors of local cellular networks42, that better model the tumor microenvironment (TME) can achieve AUCs of around 0.7, on-par with PD-L1 IHC, the current clinical gold standard.”
From A clinical benchmark of public self-supervised pathology foundation models (2025) · cites this paper as [42]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Sung, H., Ferlay, J., Siegel, R. L., Laversanne, M., Soerjomataram, I., Jemal, A., & Bray, F. (2021) · CA: A Cancer Journal for Clinicians · doi:10.3322/caac.21660 · Cited by 108,699 · see citing works
DNA sequencing with chain-terminating inhibitors
Sanger, F., Nicklen, S., & Coulson, A. R. (1977) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.74.12.5463 · Cited by 68,572 · see citing works
New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1)
Eisenhauer, E. A., Therasse, P., Bogaerts, J., Schwartz, L. H., Sargent, D., Ford, R., Dancey, J., Arbuck, S., Gwyther, S., Mooney, M., et al. (2009) · European Journal of Cancer · doi:10.1016/j.ejca.2008.10.026 · Cited by 29,534 · see citing works
Pembrolizumab versus Chemotherapy for PD-L1-Positive Non-Small-Cell Lung Cancer
Reck, M., Rodríguez-Abreu, D., Robinson, A. G., Hui, R., Csőszi, T., Fülöp, A., Gottfried, M., Peled, N., Tafreshi, A., Cuffe, S., et al. (2016) · New England Journal of Medicine · doi:10.1056/nejmoa1606774 · Cited by 10,089 · see citing works
Understanding the tumor immune microenvironment (TIME) for effective therapy
Binnewies, M., Roberts, E. W., Kersten, K., Chan, V., Fearon, D. F., Merad, M., Coussens, L. M., Gabrilovich, D. I., Ostrand-Rosenberg, S., Hedrick, C. C., et al. (2018) · Nature Medicine · doi:10.1038/s41591-018-0014-x · Cited by 6,145 · see citing works
Elements of cancer immunity and the cancer–immune set point
Chen, D. S., & Mellman, I. (2017) · Nature · doi:10.1038/nature21349 · Cited by 5,352 · see citing works
Fast Graph Representation Learning with PyTorch GeometricPreprint
Fey, M., & Lenssen, J. E. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1903.02428 · Cited by 3,403 · see citing works
Formal Semantics for Kolmogorov-Arnold Network Representations of Operational GamesPreprint
Zhukov, G. A. (2025) · Zenodo (CERN European Organization for Nuclear Research) · doi:10.5281/zenodo.15206559 · Cited by 3,385 · see citing works
Clinical-grade computational pathology using weakly supervised deep learning on whole slide images
Campanella, G., Hanna, M. G., Geneslaw, L., Miraflor, A., Werneck Krauss Silva, V., Busam, K. J., Brogi, E., Reuter, V. E., Klimstra, D. S., & Fuchs, T. J. (2019) · Nature Medicine · doi:10.1038/s41591-019-0508-1 · Cited by 2,799 · see citing works
Self-citationPan-tumor genomic biomarkers for PD-1 checkpoint blockade–based immunotherapy
Cristescu, R., Mogg, R., Ayers, M., Albright, A., Murphy, E., Yearley, J., Sher, X., Liu, X. Q., Lu, H., Nebozhyn, M., et al. (2018) · Science · doi:10.1126/science.aar3593 · Cited by 2,301 · see citing works
Reck, M., Rodríguez–Abreu, D., Robinson, A. G., Hui, R., Csőszi, T., Fülöp, A., Gottfried, M., Peled, N., Tafreshi, A., Cuffe, S., et al. (2019) · Journal of Clinical Oncology · doi:10.1200/jco.18.00149 · Cited by 1,670 · see citing works
Atezolizumab for First-Line Treatment of PD-L1–Selected Patients with NSCLC
Herbst, R. S., Giaccone, G., de Marinis, F., Reinmuth, N., Vergnenégre, A., Barrios, C. H., Morise, M., Felip, E., Andric, Z. G., Geater, S. L., et al. (2020) · New England Journal of Medicine · doi:10.1056/nejmoa1917346 · Cited by 1,668 · see citing works
Tissue Processing and Hematoxylin and Eosin Staining
Feldman, A. T., & Wolfe, D. (2014) · Methods in molecular biology · doi:10.1007/978-1-4939-1050-2_3 · Cited by 1,071 · see citing works
Intratumoral heterogeneity in cancer progression and response to immunotherapy
Vitale, I., Shema, E., Loi, S., & Galluzzi, L. (2021) · Nature Medicine · doi:10.1038/s41591-021-01233-9 · Cited by 823 · see citing works
An Intriguing Failing of Convolutional Neural Networks and the CoordConv Solution
Liu, R., Lehman, J., Molino, P., Such, F. P., Frank, E., Sergeev, A., & Yosinski, J. (2018) · arXiv · doi:10.48550/arxiv.1807.03247 · Cited by 784 · see citing works
Ström, P., Kartasalo, K., Olsson, H., Solorzano, L., Delahunt, B., Berney, D. M., Bostwick, D. G., Evans, A. J., Grignon, D. J., Humphrey, P. A., et al. (2020) · The Lancet Oncology · doi:10.1016/s1470-2045(19)30738-7 · Cited by 620 · see citing works
Diao, J. A., Wang, J. K., Chui, W. F., Mountain, V., Gullapally, S. C., Srinivasan, R., Mitchell, R. N., Glass, B., Hoffman, S., Rao, S. K., et al. (2021) · Nature Communications · doi:10.1038/s41467-021-21896-9 · Cited by 281 · see citing works
AI-based pathology predicts origins for cancers of unknown primary
Lu, M. Y., Chen, T., Williamson, D. F. K., Zhao, M. M., Shady, M., Lipková, J., & Mahmood, F. (2021) · RePEc: Research Papers in Economics · Cited by 263 · see citing works
Stenzinger, A., Allen, J. D., Maas, J., Stewart, M. D., Merino, D. M., Wempe, M. M., & Dietel, M. (2019) · Genes Chromosomes and Cancer · doi:10.1002/gcc.22733 · Cited by 239 · see citing works
Whole slide imaging equivalency and efficiency study: experience at a large academic center
Hanna, M. G., Reuter, V. E., Hameed, M., Tan, L. K., Chiang, S., Sigel, C., Hollmann, T. J., Giri, D. D., Samboy, J., Moradel, C., et al. (2019) · Modern Pathology · doi:10.1038/s41379-019-0205-0 · Cited by 215 · see citing works
Self-citationGutierrez, M. E., Choi, K., Lanman, R. B., Licitra, E. J., Skrzypczak, S. M., Benito, R. P., Wu, T., Arunajadai, S. G., Kaur, S., Harper, H. D., et al. (2017) · Clinical Lung Cancer · doi:10.1016/j.cllc.2017.04.004 · Cited by 206 · see citing works
Brunnström, H., Johansson, A., Westbom-Fremer, S., Backman, M., Djureinovic, D., Patthey, A., Isaksson‐Mettävainio, M., Gulyás, M., & Micke, P. (2017) · Modern Pathology · doi:10.1038/modpathol.2017.59 · Cited by 200 · see citing works
Validation of a digital pathology system including remote review during the COVID-19 pandemic
Hanna, M. G., Reuter, V. E., Ardon, O., Kim, D., Sirintrapun, S. J., Schüffler, P. J., BUSAM, K. J., Sauter, J. L., Brogi, E., Tan, L. K., et al. (2020) · Modern Pathology · doi:10.1038/s41379-020-0601-5 · Cited by 187 · see citing works
Understanding Attention and Generalization in Graph Neural NetworksPreprint
Knyazev, B. A., Taylor, G. W., & Amer, M. R. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1905.02850 · Cited by 174 · see citing works
Rehman, J., Han, G., Carvajal‐Hausdorf, D., Wasserman, B. E., Pelekanou, V., Mani, N. L., McLaughlin, J. F., Schalper, K. A., & Rimm, D. L. (2016) · Modern Pathology · doi:10.1038/modpathol.2016.186 · Cited by 164 · see citing works
Intra- and Interobserver Reproducibility Assessment of PD-L1 Biomarker in Non–Small Cell Lung Cancer
Cooper, W. A., Russell, P. A., Cherian, M., Duhig, E., Godbolt, D., Jessup, P. J., Khoo, C., Leslie, C., Mahar, A. M., Moffat, D. F. C., et al. (2017) · Clinical Cancer Research · doi:10.1158/1078-0432.ccr-17-0151 · Cited by 138 · see citing works
Sha, L., Osinski, B. L., Ho, I. Y., Tan, T. L., Willis, C., Weiss, H., Beaubier, N., Mahon, B. M., Taxter, T. J., & Yip, S. S. F. (2019) · Journal of Pathology Informatics · doi:10.4103/jpi.jpi_24_19 · Cited by 130 · see citing works
Building(s and) cities: Delineating urban areas with a machine learning algorithm
Arribas‐Bel, D., Garcia–López, M.-À., & Viladecans‐Marsal, E. (2019) · Journal of Urban Economics · doi:10.1016/j.jue.2019.103217 · Cited by 105 · see citing works
Schüffler, P. J., Geneslaw, L., Yarlagadda, D. V. K., Hanna, M. G., Samboy, J., Stamelos, E., Vanderbilt, C., Philip, J., Jean, M.-H., Corsale, L., et al. (2021) · Journal of the American Medical Informatics Association · doi:10.1093/jamia/ocab085 · Cited by 83 · see citing works
Self-citationBeyond PD-L1 testing-emerging biomarkers for immunotherapy in non-small cell lung cancer
Voong, K. R., Feliciano, J. L., Becker, D. J., & Levy, B. P. (2017) · Annals of Translational Medicine · doi:10.21037/atm.2017.06.48 · Cited by 82 · see citing works
Yang, Y. A., Yang, J., Shen, L., Chen, J., Xia, L., Ni, B., Ge, L., Wang, Y., & Lü, S. (2021) · PubMed Central · Cited by 76 · see citing works
RNA Sequencing of the Tumor Microenvironment in Precision Cancer Immunotherapy
Lau, D., Bobe, A. M., & Khan, A. A. (2019) · Trends in cancer · doi:10.1016/j.trecan.2019.02.006 · Cited by 51 · see citing works
Fang, C., Xu, D. H., Su, J., Dry, J. R., & Linghu, B. (2021) · npj Digital Medicine · doi:10.1038/s41746-021-00381-z · Cited by 50 · see citing works
Bao, X., Shi, R., Zhao, T., & Wang, Y. (2020) · Journal of Molecular Medicine · doi:10.1007/s00109-020-01908-9 · Cited by 31 · see citing works
AI-BASED PATHOLOGY PREDICTS ORIGINS FOR CANCERS OF UNKNOWN PRIMARY
Lu, M. Y., Chen, T., Williamson, D. F. K., et al. (2022) · Journal of Pathology Informatics · doi:10.1016/j.jpi.2022.100045 · Cited by 31 · see citing works
Beyond Classification: Whole Slide Tissue Histopathology Analysis By End-To-End Part Learning
Xie, C., Muhammad, H., Vanderbilt, C., et al. (2020) · Medical Imaging with Deep Learning · Cited by 15 · see citing works
Self-citationBuilding(s and) Cities: Delineating Urban Areas with a Machine Learning AlgorithmPreprint
de Barcelona, I. E., Arribas‐Bel, D., Garcia–López, M.-À., et al. (2020) · SSRN Electronic Journal · doi:10.2139/ssrn.3685818 · Cited by 11 · see citing works
VOCA: Cell Nuclei Detection In Histopathology Images By Vector Oriented Confidence Accumulation
Xie, C., Vanderbilt, C., Grabenstetter, A., et al. (2018) · Cited by 4 · see citing works
Self-citationHassan, M., Xie, C., Sigel, C., et al. (2021) · Medical Imaging with Deep Learning · Cited by 4 · see citing works
Self-citationNuc2Vec: Learning Representations of Nuclei in Histopathology Images with Contrastive Loss
Feng, C., Vanderbilt, C., & Fuchs, T. J. (2021) · Medical Imaging with Deep Learning · Cited by 3 · see citing works
Self-citation
Flagged first, then by citations
Research works printed but not linked: 3
- C. U. Blank, J. B. Haanen, A. Ribas and T. N. Schumacher, "The "cancer immunogram"," Science, vol. 352, no. 6286, 2016.
- K. AbdulJabbar, S. E. Raza, R. Rosenthal, M. Jamal-Hanjani, S. Veeriah, A. Akarca, T. Lund, D. A. Moore, R. Salgado, M. Al-Bakir, L. Zapata, C. T. Hiley, L. Officer, M. Sereno, C. R. Smith, S. Loi, A. Hackshaw, T. Marafioti, S. A. Quezada, N. McGranahan, J. Le Quesne, C. Swanton, M. Jamal-Hanjani, J. Le Quesne, A. Hackshaw, S. A. Quezada, N. McGranahan, R. Rosenthal, C. T. Hiley, S. Veeriah, D. A. Moore, T. Marafioti, R. Salgado, Y. Ngai, A. Sharp, C. Rodrigues, O. Pressey, S. Smith, N. Gower, H. Dhanda, J. Riley, L. Primrose, L. Martinson, N. Carey, J. A. Shaw, D. Fennell, G. A. Wilson, N. J. Birkbak, T. B. Watkins, M. Escudero, A. Stewart, A. Rowan, J. Goldman, P. Van Loo, R. K. Stone, T. Denner, E. Nye, S. Ward, E. L. Lim, S. Boeing, M. Greco, K. Litchfield, J. Nicod, C. Puttick, K. Enfield, E. Colliver, B. Campbell, C. Abbosh, Y. Wu, M. Skrzypski, R. E. Hynds, A. Georgiou, M. W. Sunderland, J. L. Reading, K. S. Peggs, J. A. Hartley, P. Gorman, H. L. Lowe, L. Ensell, V. Spanswick, A. Karamani, D. Biswas, M. Razaq, S. Beck, A. Huebner, M. Dietzen, C. Naceur-Lombardelli, M. A. Akther, H. Zhai, N. Kannu, E. Manzano, S. K. Bola, E. Ghorani, M. R. de Massy, E. Hoxha, E. Hatipoglu, S. Ogwuru, B. Chain, G. Price, S. Dubois-Marshall, K. Kerr, S. Palmer, H. Cheyne, J. Miller, K. Buchan, M. Chetty, M. Khalil, V. Ezhil, V. Prakash, G. Anand, S. Khan, K. Lau, M. Sheaff, P. Schmid, L. Lim, J. Conibear, R. Schwarz, J. Tugwood, J. Pierce, C. Dive, G. Brady, D. G. Rothwell, F. Chemi, E. Kilgour, F. Blackhall, L. Priest, M. G. Krebs, P. Crosbie, A. Nakas, S. Rathinam, L. Nelson, K. Ryanna, M. Tuffail, A. Bajaj, J. Brozik, F. Morgan, M. Kornaszewska, R. Attanoos, H. Adams, H. Davies, M. Carter, C. R. Lindsay, F. Gomes, Z. Szallasi, I. Csabai, M. Diossy, H. Aerts, A. Kirk, M. Asif, J. Butler, R. Bilanca, N. Kostoulas, M. MacKenzie, M. Wilcox, S. Busacca, A. Dawson, M. R. Lovett, M. Shackcloth, S. Feeney, J. Asante- Siaw, J. Gosney, A. Leek, N. Totten, J. D. Hodgkinson, R. Waddington, J. Rogan, K. Moore, W. Monteiro, H. Marshall, K. G. Blyth, C. Dick, A. Kidd, E. Lim, P. De Sousa, S. Jordan, A. Rice, H. Raubenheimer, H. Bhayani, M. Hamilton, L. Ambrose, A. Devaraj, H. Chavan, S. Begum, A. Mani, D. Kaniu, M. Malima, S. Booth, A. G. Nicholson, N. Fernandes, J. E. Wallen, P. Shah, S. Danson, J. Bury, J. Edwards, J. Hill, S. Matthews, Y. Kitsanta, J. Rao, S. Tenconi, L. Socci, K. Suvarna, F. Kibutu, P. Fisher, R. Young, J. Barker, F. Taylor, K. Lloyd, T. Light, T. Horey, D. Papadatos-Pastos, P. Russell, S. Lock, K. Gilbert, D. Lawrence, M. Hayward, N. Panagiotopoulos, R. George, D. Patrini, M. Falzon, E. Borg, R. Khiroya, A. Ahmed, M. Taylor, J. Choudhary, P. Shaw, S. M. Janes, M. Forster, T. Ahmad, S. M. Lee, J. Herrero, D. Carnell, R. Mendes, J. George, N. Navani, M. Scarci, E. Bertoja, R. C. Stephens, E. M. Hoogenboom, J. W. Holding, S. Bandula, B. Naidu, G. Langman, A. Robinson, H. Bancroft, A. Kerr, S. Kadiri, C. Ferris, G. Middleton, M. Djearaman, A. Patel, C. Ottensmeier, S. Chee, B. Johnson, A. Alzetani, E. Shaw, J. Lester, Y. Summers, R. Califano, P. Taylor, R. Shah, P. Krysiak, K. Rammohan, E. Fontaine, R. Booton, M. Evison, S. Moss, J. Novasio, L. Joseph, P. Bishop, A. Chaturvedi, H. Doran, F. Granato, V. Joshi, E. Smith, A. Montero, C. Swanton and Y. Yuan, "Geospatial immune variability illuminates differential evolution of lung adenocarcinoma," Nature Medicine, vol. 26, no. 7, 2020.
- K. Xu, S. Jegelka, W. Hu and J. Leskovec, "How powerful are graph neural networks?," 7th International Conference on Learning Representations, ICLR 2019, 2019.
Cited by
Sorted by citations
A foundation model for clinical-grade computational pathology and rare cancers detection
Vorontsov, E., Bozkurt, A., Casson, A., Shaikovski, G., Zelechowski, M., Severson, K., Zimmermann, E., Hall, J., Tenenholtz, N., Fusi, N., et al. (2024) · Nature Medicine · doi:10.1038/s41591-024-03141-0 · Cited by 572 · see citing works
Self-citationA clinical benchmark of public self-supervised pathology foundation models
Campanella, G., Chen, S., Singh, M., Verma, R., Muehlstedt, S., Zeng, J., Stock, A., Croken, M., Veremis, B., Elmas, A., et al. (2025) · Nature Communications · doi:10.1038/s41467-025-58796-1 · Cited by 126 · see citing works
Self-citationDeep learning in digital pathology for personalized treatment plans of cancer patients
Wen, Z., Wang, S., Yang, D. M., Xie, Y., Chen, M., Bishop, J., & Xiao, G. (2023) · Seminars in Diagnostic Pathology · doi:10.1053/j.semdp.2023.02.003 · Cited by 34 · see citing works
- Authors
- Xie, C., Vanderbilt, C., Feng, C., Ho, D., Campanella, G., Egger, J., Plodkowski, A., Girshman, J., Sawan, P., Arbour, K., Hellmann, M., & Fuchs, T.
- Affiliations
- Memorial Sloan Kettering Cancer Center (US)
- Icahn School of Medicine at Mount Sinai (US)
- AstraZeneca Oncology R&D
- ORCID
- Venue
- Research Square
- Published
- 2022-01-21
- OpenAlex
- W4205925879
- Cited by
- 8
- References
- 40
- Citations by year
- 2025: 2 · 2024: 2 · 2023: 3 · 2022: 1
- Type
- posted content
- Language
- en
- Open access
- green
- doi.org
- Locations
- Research Square (acceptedVersion, cc-by)
- Funders
- Damon Runyon Cancer Research Foundation (P30 CA008748, CI-98-18)
- Warren Alpert Foundation
- Memorial Sloan-Kettering Cancer Center (CA008748, PCT/US2015/062208, P30 CA008748, NIH/NCI Cancer Center Support grant P30CA008748, Support Grant P30 CA008748)
- National Institutes of Health (CA008748, NIH/NCI Cancer Center Support Grant P30 CA008748)
- Weill Cornell Graduate School of Medical Sciences
- Weill Cornell Medical College
- National Cancer Institute (Grant P30-CA008748, PCT/US2015/062208, CA008748, NIH/NCI Cancer Center Support Grant P30 CA008748., Support Grant P30 CA008748, P30CA008748)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › AI in cancer detection
- Health Sciences › Medicine › Oncology › Cancer Immunotherapy and Biomarkers
- Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Radiomics and Machine Learning in Medical Imaging
- Keywords
- lung cancer, immune checkpoint inhibitors, treatment response prediction, H&E whole-slide images, graph neural networks, PD-L1 immunohistochemistry, tumor mutational burden, deep learning
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
