CoChat
Sign Up
Cited by 83AI in cancer detectionRadiomics and Machine Learning in Medical ImagingBRCA gene mutations in cancer

Deep Learning Artificial Intelligence Predicts Homologous Recombination Deficiency and Platinum Response From Histologic Slides

Open access
Bergstrom, Erik N., Abbasi, Ammal, Díaz-Gay, Marcos, Galland, Loïck, Ladoire, Sylvain, Lippman, Scott M., Alexandrov, Ludmil B.·2024·Journal of Clinical Oncology
DOI 10.1200/jco.23.02641PMC PMC11469627
Open this paper in CoChat — free account

Abstract

PURPOSE: Cancers with homologous recombination deficiency (HRD) can benefit from platinum salts and poly(ADP-ribose) polymerase inhibitors. Standard diagnostic tests for detecting HRD require molecular profiling, which is not universally available. METHODS: We trained DeepHRD, a deep learning platform for predicting HRD from hematoxylin and eosin (H&E)–stained histopathological slides, using primary breast (n = 1,008) and ovarian (n = 459) cancers from The Cancer Genome Atlas (TCGA). DeepHRD was compared with four standard HRD molecular tests using breast (n = 349) and ovarian (n = 141) cancers from multiple independent data sets, including platinum-treated clinical cohorts with RECIST progression-free survival (PFS), complete response (CR), and overall survival (OS) endpoints. RESULTS: DeepHRD predicted HRD from held-out H&E-stained breast cancer slides in TCGA with an AUC of 0.81 (95% CI, 0.77 to 0.85). This performance was confirmed in two independent primary breast cancer cohorts (AUC, 0.76 [95% CI, 0.71 to 0.82]). In an external platinum-treated metastatic breast cancer cohort, samples predicted as HRD had higher complete CR (AUC, 0.76 [95% CI, 0.54 to 0.93]) with 3.7-fold increase in median PFS (14.4 v 3.9 months; P = .0019) and hazard ratio (HR) of 0.45 ( P = .0047). There were no significant differences in nonplatinum treatment outcome by predicted HRD status in three breast cancer cohorts, including CR (AUC, 0.39) and PFS (HR, 0.98, P = .95) in taxane-treated metastatic breast cancer. Through transfer learning to high-grade serous ovarian cancer, DeepHRD-predicted HRD samples had better OS after first-line (HR, 0.46; P = .030) and neoadjuvant (HR, 0.49; P = .015) platinum therapy in two cohorts. CONCLUSION: DeepHRD can predict HRD in breast and ovarian cancers directly from routine H&E slides across multiple external cohorts, slide scanners, and tissue fixation variables. When compared with molecular testing, DeepHRD classified 1.8- to 3.1-fold more patients with HRD, which exhibited better OS in high-grade serous ovarian cancer and platinum-specific PFS in metastatic breast cancer.

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.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1200/jco.23.02641

Cites

5 of 5 references verified0 citation errors · 0 not checkable

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.

A few free checks a day · papers already verified open instantly · results are public

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?

  • SpaHE-Infil pioneers in situ resolution of immune cell spatial distributions within the TME using routine H&E-stained histopathological slides.

    “The SpaHE-Infil tool developed in this study pioneers in situ resolution of immune cell spatial distributions within the TME using routine H&E-stained histopathological slides.15”

    From SpaHE-Infil: A spatial heterogeneity framework for decoding TME infiltration from H&E-stained slides (2026) · cites this paper as [15]

    The Methods (Study Design; page unavailable) identify the tool implemented in this study as DeepHRD, which predicts homologous recombination deficiency from H&E slides—not SpaHE-Infil. The Methods (Training workflow; page unavailable) describe mapping regions important to HRD predictions, not resolving immune-cell spatial distributions. The cited paper therefore contradicts the attribution that this study developed SpaHE-Infil. (LLM, conf=0.87)

    Checked by a reasoning model — may be wrong.

    Experimental — accuracy not yet measured

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

Flagged first, then by citations

Research works printed but not linked: 50

  1. [1]Schilsky RL, Longo DL. Closing the Gap in Cancer Genomic Testing. N Engl J Med 2022; 387(23): 2107– 2110. DOI: 10.1056/NEJMp2210638. 36472952 · DOI
  2. [2]Prakash R, Zhang Y, Feng W, Jasin M. Homologous recombination and human health: the roles of BRCA1, BRCA2, and associated proteins. Cold Spring Harb Perspect Biol 2015; 7(4): a016600. DOI: 10.1101/cshperspect.a016600. PMC4382744 25833843 · DOI · Paper page
  3. [3]Mangogna A, Munari G, Pepe F, Homologous Recombination Deficiency in Ovarian Cancer: from the Biological Rationale to Current Diagnostic Approaches. J Pers Med 2023; 13(2). DOI: 10.3390/jpm13020284. PMC9968181 36836518 · DOI
  4. [4]Polak P, Kim J, Braunstein LZ, A mutational signature reveals alterations underlying deficient homologous recombination repair in breast cancer. Nat Genet 2017; 49(10): 1476– 1486. DOI: 10.1038/ng.3934. PMC7376751 28825726 · DOI · Paper page
  5. [5]Stronach EA, Paul J, Timms KM, Biomarker Assessment of HR Deficiency, Tumor BRCA1/2 Mutations, and CCNE1 Copy Number in Ovarian Cancer: Associations with Clinical Outcome Following Platinum Monotherapy. Mol Cancer Res 2018; 16(7): 1103– 1111. DOI: 10.1158/1541-7786.MCR-18-0034. 29724815 · DOI
  6. [7]Peng G, Chun-Jen Lin C, Mo W, Genome-wide transcriptome profiling of homologous recombination DNA repair. Nat Commun 2014; 5: 3361. DOI: 10.1038/ncomms4361. PMC4017859 24553445 · DOI
  7. [8]Tate JG, Bamford S, Jubb HC, COSMIC: the Catalogue Of Somatic Mutations In Cancer. Nucleic Acids Res 2019; 47(D1): D941– D947. DOI: 10.1093/nar/gky1015. PMC6323903 30371878 · DOI · Paper page
  8. [9]Ngoi NYL, Tan DSP. The role of homologous recombination deficiency testing in ovarian cancer and its clinical implications: do we need it? ESMO Open 2021; 6(3): 100144. DOI: 10.1016/j.esmoop.2021.100144. PMC8141874 34015643 · DOI
  9. [10]Telli ML, Timms KM, Reid J, Homologous Recombination Deficiency (HRD) Score Predicts Response to Platinum-Containing Neoadjuvant Chemotherapy in Patients with Triple-Negative Breast Cancer. Clin Cancer Res 2016; 22(15): 3764– 73. DOI: 10.1158/1078-0432.CCR-15-2477. PMC6773427 26957554 · DOI · Paper page
  10. [12]Davies H, Glodzik D, Morganella S, HRDetect is a predictor of BRCA1 and BRCA2 deficiency based on mutational signatures. Nat Med 2017; 23(4): 517– 525. DOI: 10.1038/nm.4292. PMC5833945 28288110 · DOI · Paper page
  11. [13]Gulhan DC, Lee JJ, Melloni GEM, Cortes-Ciriano I, Park PJ. Detecting the mutational signature of homologous recombination deficiency in clinical samples. Nat Genet 2019; 51(5): 912– 919. DOI: 10.1038/s41588-019-0390-2. 30988514 · DOI
  12. [14]Perez-Villatoro F, Oikkonen J, Casado J, Optimized detection of homologous recombination deficiency improves the prediction of clinical outcomes in cancer. NPJ Precis Oncol 2022; 6(1): 96. (In eng). DOI: 10.1038/s41698-022-00339-8. PMC9800569 36581696 · DOI
  13. [15]Rodler E, Sharma P, Barlow WE, Cisplatin with veliparib or placebo in metastatic triple-negative breast cancer and BRCA mutation-associated breast cancer (S1416): a randomised, double-blind, placebo-controlled, phase 2 trial. Lancet Oncol 2023; 24(2): 162– 174. (In eng). DOI: 10.1016/s1470-2045(22)00739-2. PMC9924094 36623515 · DOI
  14. [16]Leith A, Ribbands A, Kim J, Real-world homologous recombination repair mutation testing in metastatic castration-resistant prostate cancer in the USA, Europe and Japan. Future Oncol 2022; 18(8): 937– 951. (In eng). DOI: 10.2217/fon-2021-1113. 35043687 · DOI
  15. [17]Park W, Chen J, Chou JF, Genomic Methods Identify Homologous Recombination Deficiency in Pancreas Adenocarcinoma and Optimize Treatment Selection. Clin Cancer Res 2020; 26(13): 3239– 3247. DOI: 10.1158/1078-0432.CCR-20-0418. PMC7380542 32444418 · DOI
  16. [18]Ascierto PA, Bifulco C, Palmieri G, Peters S, Sidiropoulos N. Preanalytic Variables and Tissue Stewardship for Reliable Next-Generation Sequencing (NGS) Clinical Analysis. J Mol Diagn 2019; 21(5): 756– 767. DOI: 10.1016/j.jmoldx.2019.05.004. 31251989 · DOI
  17. [19]Fitzgerald J, Higgins D, Mazo Vargas C, Future of biomarker evaluation in the realm of artificial intelligence algorithms: application in improved therapeutic stratification of patients with breast and prostate cancer. J Clin Pathol 2021; 74(7): 429– 434. DOI: 10.1136/jclinpath-2020-207351. 34117103 · DOI
  18. [20]Heitz F, Ataseven B, Staniczok C, Implementing HRD Testing in Routine Clinical Practice on Patients with Primary High-Grade Advanced Ovarian Cancer. Cancers 2023; 15(3): 818. (https://www.mdpi.com/2072-6694/15/3/818). PMC9913091 36765776
  19. [21]Diao JA, Wang JK, Chui WF, Human-interpretable image features derived from densely mapped cancer pathology slides predict diverse molecular phenotypes. Nat Commun 2021; 12(1): 1613. DOI: 10.1038/s41467-021-21896-9. PMC7955068 33712588 · DOI
  20. [22]Campanella G, Hanna MG, Geneslaw L, Clinical-grade computational pathology using weakly supervised deep learning on whole slide images. Nat Med 2019; 25(8): 1301– 1309. DOI: 10.1038/s41591-019-0508-1. PMC7418463 31308507 · DOI · Paper page
  21. [23]Couture HD. Deep Learning-Based Prediction of Molecular Tumor Biomarkers from H&E: A Practical Review. J Pers Med 2022; 12(12). DOI: 10.3390/jpm12122022. PMC9784641 36556243 · DOI
  22. [24]Bilal M, Nimir M, Snead D, Taylor GS, Rajpoot N. Role of AI and digital pathology for colorectal immuno-oncology. Br J Cancer 2023; 128(1): 3– 11. DOI: 10.1038/s41416-022-01986-1. PMC9814831 36183010 · DOI
  23. [25]Kather JN, Pearson AT, Halama N, Deep learning can predict microsatellite instability directly from histology in gastrointestinal cancer. Nat Med 2019; 25(7): 1054– 1056. DOI: 10.1038/s41591-019-0462-y. PMC7423299 31160815 · DOI · Paper page
  24. [26]Schirris Y, Gavves E, Nederlof I, Horlings HM, Teuwen J. DeepSMILE: Contrastive self-supervised pre-training benefits MSI and HRD classification directly from H&E whole-slide images in colorectal and breast cancer. Med Image Anal 2022; 79: 102464. (In eng). DOI: 10.1016/j.media.2022.102464. 35596966 · DOI
  25. [27]Lazard T, Bataillon G, Naylor P, Deep learning identifies morphological patterns of homologous recombination deficiency in luminal breast cancers from whole slide images. Cell Rep Med 2022; 3(12): 100872. (In eng). DOI: 10.1016/j.xcrm.2022.100872. PMC9798078 36516847 · DOI
  26. [28]Nero C, Boldrini L, Lenkowicz J, Deep-Learning to Predict BRCA Mutation and Survival from Digital H&E Slides of Epithelial Ovarian Cancer. Int J Mol Sci 2022; 23(19) (In eng). DOI: 10.3390/ijms231911326. PMC9570450 36232628 · DOI
  27. [29]Valieris R, Amaro L, Osório C, Deep Learning Predicts Underlying Features on Pathology Images with Therapeutic Relevance for Breast and Gastric Cancer. Cancers (Basel) 2020; 12(12) (In eng). DOI: 10.3390/cancers12123687. PMC7763049 33316873 · DOI
  28. [30]Cancer Genome Atlas N. Comprehensive molecular portraits of human breast tumours. Nature 2012; 490(7418): 61– 70. DOI: 10.1038/nature11412. PMC3465532 23000897 · DOI · Paper page
  29. [31](CPTAC) NCICPTAC. The Clinical Proteomic Tumor Analysis Consortium Breast Invasive Carcinoma Collection (CPTAC-BRCA) (Version 1) [Data set]. The Cancer Imaging Archive 2020. DOI: 10.7937/TCIA.CAEM-YS80. · DOI
  30. [33]Galland L, Ballot E, Mananet H, Efficacy of platinum-based chemotherapy in metastatic breast cancer and HRD biomarkers: utility of exome sequencing. NPJ Breast Cancer 2022; 8(1): 28. DOI: 10.1038/s41523-022-00395-0. PMC8897409 35246547 · DOI
  31. [34]Cancer Genome Atlas Research N. Integrated genomic analyses of ovarian carcinoma. Nature 2011; 474(7353): 609– 15. DOI: 10.1038/nature10166. PMC3163504 21720365 · DOI · Paper page
  32. [36]Sztupinszki Z, Diossy M, Krzystanek M, Migrating the SNP array-based homologous recombination deficiency measures to next generation sequencing data of breast cancer. NPJ Breast Cancer 2018; 4: 16. DOI: 10.1038/s41523-018-0066-6. PMC6028448 29978035 · DOI
  33. [37]Takaya H, Nakai H, Takamatsu S, Mandai M, Matsumura N. Homologous recombination deficiency status-based classification of high-grade serous ovarian carcinoma. Sci Rep 2020; 10(1): 2757. DOI: 10.1038/s41598-020-59671-3. PMC7026096 32066851 · DOI
  34. [38]Li J, Li W, Sisk A, A multi-resolution model for histopathology image classification and localization with multiple instance learning. Comput Biol Med 2021; 131: 104253. DOI: 10.1016/j.compbiomed.2021.104253. PMC7984430 33601084 · DOI
  35. [39]Tovey H, Sipos O, Parker JS, Integrated Multimodal Analyses of DNA Damage Response and Immune Markers as Predictors of Response in Metastatic Triple-Negative Breast Cancer in the TNT Trial (NCT00532727). Clin Cancer Res 2023; 29(18): 3691– 3705. DOI: 10.1158/1078-0432.CCR-23-0370. PMC10502473 37574209 · DOI
  36. [40]Sipos O, Tovey H, Quist J, Assessment of structural chromosomal instability phenotypes as biomarkers of carboplatin response in triple negative breast cancer: the TNT trial. Ann Oncol 2021; 32(1): 58– 65. DOI: 10.1016/j.annonc.2020.10.475. PMC7784666 33098992 · DOI
  37. [41]Chen Y, Wang X, Du F, Association between homologous recombination deficiency and outcomes with platinum and platinum-free chemotherapy in patients with triple-negative breast cancer. Cancer Biol Med 2023; 20(2): 155– 68. DOI: 10.20892/j.issn.2095-3941.2022.0525. PMC9978893 36861447 · DOI
  38. [42]Chai Y, Chen Y, Zhang D, Homologous Recombination Deficiency (HRD) and BRCA 1/2 Gene Mutation for Predicting the Effect of Platinum-Based Neoadjuvant Chemotherapy of Early-Stage Triple-Negative Breast Cancer (TNBC): A Systematic Review and Meta-Analysis. J Pers Med 2022; 12(2). DOI: 10.3390/jpm12020323. PMC8876589 35207810 · DOI
  39. [43]Graham S, Vu QD, Raza SEA, Hover-Net: Simultaneous segmentation and classification of nuclei in multi-tissue histology images. Med Image Anal 2019; 58: 101563. DOI: 10.1016/j.media.2019.101563. 31561183 · DOI
  40. [45]Swisher EM, Aghajanian C, O’Malley DM, Impact of homologous recombination status and responses with veliparib combined with first-line chemotherapy in ovarian cancer in the Phase 3 VELIA/GOG-3005 study. Gynecol Oncol 2022; 164(2): 245– 253. DOI: 10.1016/j.ygyno.2021.12.003. 34906376 · DOI
  41. [46]Tsang ES, Csizmok V, Williamson LM, Homologous recombination deficiency signatures in gastrointestinal and thoracic cancers correlate with platinum therapy duration. NPJ Precis Oncol 2023; 7(1): 31. DOI: 10.1038/s41698-023-00368-x. PMC10039042 36964191 · DOI
  42. [47]Casolino R, Paiella S, Azzolina D, Homologous Recombination Deficiency in Pancreatic Cancer: A Systematic Review and Prevalence Meta-Analysis. J Clin Oncol 2021; 39(23): 2617– 2631. DOI: 10.1200/JCO.20.03238. PMC8331063 34197182 · DOI
  43. [48]Stewart MD, Merino Vega D, Arend RC, Homologous Recombination Deficiency: Concepts, Definitions, and Assays. Oncologist 2022; 27(3): 167– 174. DOI: 10.1093/oncolo/oyab053. PMC8914493 35274707 · DOI · Paper page
  44. [49]Ektefaie Y, Yuan W, Dillon DA, Integrative multiomics-histopathology analysis for breast cancer classification. NPJ Breast Cancer 2021; 7(1): 147. DOI: 10.1038/s41523-021-00357-y. PMC8630188 34845230 · DOI
  45. [50]Nemunaitis J, Stanbery L, Walter A, Rocconi R, Stephens P. Rationale for the Use of Homologous Recombination Proficient Molecular Profile as a Biomarker for Therapeutic Targeting in Ovarian Cancer. Oncol Rev 2023; 17: 11471. DOI: 10.3389/or.2023.11471. PMC10547877 37799595 · DOI
  46. [51]Chen KT, Madison R, Moore J, A Novel HRD Signature Is Predictive of FOLFIRINOX Benefit in Metastatic Pancreatic Cancer. Oncologist 2023; 28(8): 691– 698. DOI: 10.1093/oncolo/oyad178. PMC10400136 37354528 · DOI
  47. [52]Rocconi RP, Monk BJ, Walter A, Gemogenovatucel-T (Vigil) immunotherapy demonstrates clinical benefit in homologous recombination proficient (HRP) ovarian cancer. Gynecol Oncol 2021; 161(3): 676– 680. DOI: 10.1016/j.ygyno.2021.03.009. 33715892 · DOI
  48. [53]Creeden JF, Nanavaty NS, Einloth KR, Homologous recombination proficiency in ovarian and breast cancer patients. BMC Cancer 2021; 21(1): 1154. DOI: 10.1186/s12885-021-08863-9. PMC8555001 34711195 · DOI
  49. [54]Scott JA, Lennerz J, Johnson ML, Compromised Outcomes in Stage IV Non-Small-Cell Lung Cancer With Actionable Mutations Initially Treated Without Tyrosine Kinase Inhibitors: A Retrospective Analysis of Real-World Data. JCO Oncol Pract 2023: OP2200611. DOI: 10.1200/OP.22.00611. PMC10827288 37556776 · DOI
  50. [55]Bruno DS, Hess LM, Li X, Su EW, Patel M. Disparities in Biomarker Testing and Clinical Trial Enrollment Among Patients With Lung, Breast, or Colorectal Cancers in the United States. JCO Precis Oncol 2022; 6: e2100427. DOI: 10.1200/PO.21.00427. 35737912 · DOI

Cited by

Sorted by citations

Authors
Bergstrom, E. N., Abbasi, A., Díaz-Gay, M., Galland, L., Ladoire, S., Lippman, S. M., & Alexandrov, L. B.
Affiliations
  • University of California San Diego (US)
  • Moores Cancer Center, UC San Diego, La Jolla, CA
  • Inserm (FR)
Corresponding author
Alexandrov, Ludmil B.
ORCID
Venue
Journal of Clinical Oncology
Publisher
American Society of Clinical Oncology (ASCO)
Published
2024-10-20
Volume
42
Issue
30
Pages
3550-3560
ISSN
  • 0732-183X (ISSN-L)
  • 1527-7755
DOI
10.1200/jco.23.02641
PMC
PMC11469627
OpenAlex
W4401165910
Cited by
83
References
5
Citations by year
2026: 43 · 2025: 32 · 2024: 8
FWCI
12.25
Type
journal article
Language
en
Open access
Licence
cc-by-nc-nd
Locations
  • Journal of Clinical Oncology (publishedVersion, cc-by-nc-nd)
  • PubMed Central (publishedVersion, cc-by-nc-nd)
  • PubMed (publishedVersion)
Funders
  • National Cancer Institute (P30CA023100, U01CA290479)
  • National Institute of Environmental Health Sciences (R01ES032547)
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › AI in cancer detection
  • Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Radiomics and Machine Learning in Medical Imaging
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Genetics › BRCA gene mutations in cancer
Keywords
homologous recombination deficiency, platinum response, H&E-stained slides, breast cancer, ovarian cancer, high-grade serous ovarian cancer, transfer learning, progression-free survival, complete response, overall survival, fixation artifacts, slide scanners
MeSH terms
Poly(ADP-ribose) Polymerase Inhibitors, Deep Learning, Artificial Intelligence, Breast Neoplasms, Female, Humans, Middle Aged, Ovarian Neoplasms, Platinum, Homologous Recombination

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