Prediction of recurrence risk in endometrial cancer with multimodal deep learning

Sarah Volinsky-Fremond; Nanda Horeweg; Sonali Andani; Jurriaan Barkey Wolf; Maxime W Lafarge; Cor D de Kroon; Gitte Ørtoft; Estrid Høgdall; Jouke Dijkstra; Jan J Jobsen; Ludy C H W Lutgens; Melanie E Powell; Linda R Mileshkin; Helen Mackay; Alexandra Leary; Dionyssios Katsaros; Hans W Nijman; Stephanie M de Boer; Remi A Nout; Marco de Bruyn; David Church; Vincent T H B M Smit; Carien L Creutzberg; Viktor H Koelzer; Tjalling Bosse

doi:10.1038/s41591-024-02993-w

Prediction of recurrence risk in endometrial cancer with multimodal deep learning

Nat Med. 2024 May 24. doi: 10.1038/s41591-024-02993-w. Online ahead of print.

Authors

Sarah Volinsky-Fremond¹, Nanda Horeweg², Sonali Andani^{3

4

5}, Jurriaan Barkey Wolf¹, Maxime W Lafarge⁴, Cor D de Kroon⁶, Gitte Ørtoft⁷, Estrid Høgdall⁸, Jouke Dijkstra⁹, Jan J Jobsen¹⁰, Ludy C H W Lutgens¹¹, Melanie E Powell¹², Linda R Mileshkin¹³, Helen Mackay¹⁴, Alexandra Leary¹⁵, Dionyssios Katsaros¹⁶, Hans W Nijman¹⁷, Stephanie M de Boer², Remi A Nout¹⁸, Marco de Bruyn¹⁷, David Church^{19

20}, Vincent T H B M Smit¹, Carien L Creutzberg², Viktor H Koelzer^{4

21}, Tjalling Bosse²²

Affiliations

¹ Department of Pathology, Leiden University Medical Center, Leiden, The Netherlands.
² Department of Radiation Oncology, Leiden University Medical Center, Leiden, The Netherlands.
³ Department of Computer Science, ETH Zurich, Zurich, Switzerland.
⁴ Department of Pathology and Molecular Pathology, University Hospital, University of Zurich, Zurich, Switzerland.
⁵ Swiss Institute of Bioinformatics, Lausanne, Switzerland.
⁶ Department of Gynecology and Obstetrics, Leiden University Medical Center, Leiden, The Netherlands.
⁷ Department of Gynecology, Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark.
⁸ Department of Pathology, Herlev University Hospital, Herlev, Denmark.
⁹ Department of Radiology, Leiden University Medical Center, Leiden, The Netherlands.
¹⁰ Department of Radiation Oncology, Medisch Spectrum Twente, Enschede, The Netherlands.
¹¹ Maastricht Radiation Oncology, MAASTRO, Maastricht, The Netherlands.
¹² Department of Clinical Oncology, Barts Health NHS Trust, London, UK.
¹³ Department of Medical Oncology, Peter MacCallum Cancer Center, Melbourne, Victoria, Australia.
¹⁴ Department of Medical Oncology and Hematology, Odette Cancer Center Sunnybrook Health Sciences Center, Toronto, Ontario, Canada.
¹⁵ Department Medical Oncology, Gustave Roussy Institute, Villejuif, France.
¹⁶ Department of Surgical Sciences, Gynecologic Oncology, Città della Salute and S Anna Hospital, University of Turin, Turin, Italy.
¹⁷ Department of Obstetrics and Gynecology, University Medical Center Groningen, University of Groningen, Groningen, The Netherlands.
¹⁸ Department of Radiotherapy, Erasmus MC Cancer Institute, University Medical Center Rotterdam, Rotterdam, The Netherlands.
¹⁹ Wellcome Centre for Human Genetics, University of Oxford, Oxford, UK.
²⁰ Oxford NIHR Comprehensive Biomedical Research Centre, Oxford University Hospitals NHS Foundation Trust, Oxford, UK.
²¹ Institute of Medical Genetics and Pathology, University Hospital Basel, Basel, Switzerland.
²² Department of Pathology, Leiden University Medical Center, Leiden, The Netherlands. t.bosse@lumc.nl.

PMID: 38789645
DOI: 10.1038/s41591-024-02993-w

Abstract

Predicting distant recurrence of endometrial cancer (EC) is crucial for personalized adjuvant treatment. The current gold standard of combined pathological and molecular profiling is costly, hampering implementation. Here we developed HECTOR (histopathology-based endometrial cancer tailored outcome risk), a multimodal deep learning prognostic model using hematoxylin and eosin-stained, whole-slide images and tumor stage as input, on 2,072 patients from eight EC cohorts including the PORTEC-1/-2/-3 randomized trials. HECTOR demonstrated C-indices in internal (n = 353) and two external (n = 160 and n = 151) test sets of 0.789, 0.828 and 0.815, respectively, outperforming the current gold standard, and identified patients with markedly different outcomes (10-year distant recurrence-free probabilities of 97.0%, 77.7% and 58.1% for HECTOR low-, intermediate- and high-risk groups, respectively, by Kaplan-Meier analysis). HECTOR also predicted adjuvant chemotherapy benefit better than current methods. Morphological and genomic feature extraction identified correlates of HECTOR risk groups, some with therapeutic potential. HECTOR improves on the current gold standard and may help delivery of personalized treatment in EC.