Predicting Molecular Subtype and Survival of Rhabdomyosarcoma Patients Using Deep Learning of H&E Images: A Report from the Children's Oncology Group

• Published in: Clinical cancer research 2023-01, Vol.29 (2), p.364-378
• Main Authors: Milewski, David, Jung, Hyun, Brown, G Thomas, Liu, Yanling, Somerville, Ben, Lisle, Curtis, Ladanyi, Marc, Rudzinski, Erin R, Choo-Wosoba, Hyoyoung, Barkauskas, Donald A, Lo, Tammy, Hall, David, Linardic, Corinne M, Wei, Jun S, Chou, Hsien-Chao, Skapek, Stephen X, Venkatramani, Rajkumar, Bode, Peter K, Steinberg, Seth M, Zaki, George, Kuznetsov, Igor B, Hawkins, Douglas S, Shern, Jack F, Collins, Jack, Khan, Javed
• Format: Article
• Language: English
• Subjects: Child; Deep Learning; Eosine Yellowish-(YS); Hematoxylin; Humans; Paired Box Transcription Factors - genetics; Precision Medicine and Imaging; Prospective Studies; Rhabdomyosarcoma - diagnosis; Rhabdomyosarcoma - genetics; Rhabdomyosarcoma, Alveolar - genetics; Young Adult
• ISSN: 1078-0432; 1557-3265
• DOI: 10.1158/1078-0432.CCR-22-1663

Rhabdomyosarcoma (RMS) is an aggressive soft-tissue sarcoma, which primarily occurs in children and young adults. We previously reported specific genomic alterations in RMS, which strongly correlated with survival; however, predicting these mutations or high-risk disease at diagnosis remains a significant challenge. In this study, we utilized convolutional neural networks (CNN) to learn histologic features associated with driver mutations and outcome using hematoxylin and eosin (H&E) images of RMS. Digital whole slide H&E images were collected from clinically annotated diagnostic tumor samples from 321 patients with RMS enrolled in Children's Oncology Group (COG) trials (1998-2017). Patches were extracted and fed into deep learning CNNs to learn features associated with mutations and relative event-free survival risk. The performance of the trained models was evaluated against independent test sample data (n = 136) or holdout test data. The trained CNN could accurately classify alveolar RMS, a high-risk subtype associated with PAX3/7-FOXO1 fusion genes, with an ROC of 0.85 on an independent test dataset. CNN models trained on mutationally-annotated samples identified tumors with RAS pathway with a ROC of 0.67, and high-risk mutations in MYOD1 or TP53 with a ROC of 0.97 and 0.63, respectively. Remarkably, CNN models were superior in predicting event-free and overall survival compared with current molecular-clinical risk stratification. This study demonstrates that high-risk features, including those associated with certain mutations, can be readily identified at diagnosis using deep learning. CNNs are a powerful tool for diagnostic and prognostic prediction of rhabdomyosarcoma, which will be tested in prospective COG clinical trials.