{"doi":"10.1158/1078-0432.ccr-22-1663","title":"Predicting Molecular Subtype and Survival of Rhabdomyosarcoma Patients Using Deep Learning of H&E Images: A Report from the Children's Oncology Group","abstract":"PURPOSE: 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. EXPERIMENTAL DESIGN: 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. RESULTS: 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. CONCLUSIONS: 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.","journal":"Clinical Cancer Research","year":2022,"id":244630,"datarank":0.0,"base_score":0.0,"endowment":0.0,"self_citation_contribution":0.0,"citation_network_contribution":0.0,"self_endowment_contribution":0.0,"citer_contribution":0.0,"corpus_percentile":null,"corpus_rank":null,"citation_count":28,"citer_count":0,"citers_with_citation_signal":0,"citers_with_endowment":0,"datacite_reuse_total":0,"is_dataset":false,"is_dataset_confidence":0.9592,"is_data_producer":false,"deposit_databanks":null,"is_oa":true,"file_count":0,"downloads":0,"has_version_chain":false,"published_date":"2022-01-01","fair_score":null,"fair_percentile":null,"algorithm_id":"datarank_citation_only_1hop_v6","ranking_scope":"data_only","authors":[{"id":257837,"name":"Hyun Jung","orcid":"0000-0001-7467-8189","position":1,"is_corresponding":false},{"id":308598,"name":"G. Thomas Brown","orcid":"0000-0001-7383-3116","position":2,"is_corresponding":false},{"id":648100,"name":"Yanling Liu","orcid":"0000-0002-5173-2426","position":3,"is_corresponding":false},{"id":878426,"name":"Ben Somerville","orcid":"0000-0002-3222-4979","position":4,"is_corresponding":false},{"id":878427,"name":"Curtis Lisle","orcid":"0000-0003-4056-3038","position":5,"is_corresponding":false},{"id":1869,"name":"Marc Ladanyi","orcid":"0000-0001-9055-7213","position":6,"is_corresponding":false},{"id":335000,"name":"Erin R. Rudzinski","orcid":"0000-0002-7591-956X","position":7,"is_corresponding":false},{"id":291439,"name":"Hyoyoung Choo‐Wosoba","orcid":"0000-0002-8756-1016","position":8,"is_corresponding":false},{"id":277181,"name":"Donald A. Barkauskas","orcid":"0000-0002-2339-719X","position":9,"is_corresponding":false},{"id":277182,"name":"Tammy Lo","orcid":"0000-0002-7980-9595","position":10,"is_corresponding":false},{"id":277183,"name":"David Hall","orcid":"0000-0003-1257-2316","position":11,"is_corresponding":false},{"id":434429,"name":"Corinne M. Linardic","orcid":"0000-0002-3257-2885","position":12,"is_corresponding":false},{"id":250438,"name":"Jun S. Wei","orcid":"0000-0002-4812-0250","position":13,"is_corresponding":false},{"id":506189,"name":"Hsien-Chao Chou","orcid":"0000-0002-4870-9663","position":14,"is_corresponding":false},{"id":277184,"name":"Stephen X. Skapek","orcid":"0000-0002-4136-8384","position":15,"is_corresponding":false},{"id":335001,"name":"Rajkumar Venkatramani","orcid":"0000-0002-4785-106X","position":16,"is_corresponding":false},{"id":482470,"name":"Peter K. Bode","orcid":"0000-0002-9633-4042","position":17,"is_corresponding":false},{"id":109203,"name":"Seth M. Steinberg","orcid":"0000-0002-8280-551X","position":18,"is_corresponding":false},{"id":404282,"name":"George Zaki","orcid":"0000-0002-2740-3307","position":19,"is_corresponding":false},{"id":509891,"name":"Igor B. Kuznetsov","orcid":"0000-0003-3741-8964","position":20,"is_corresponding":false},{"id":277185,"name":"Douglas S. Hawkins","orcid":"0000-0003-3602-1375","position":21,"is_corresponding":false},{"id":309026,"name":"Jack F. Shern","orcid":"0000-0001-5579-7625","position":22,"is_corresponding":false},{"id":546124,"name":"Jack Collins","orcid":"0000-0001-8281-2085","position":23,"is_corresponding":false},{"id":250453,"name":"Javed Khan","orcid":"0000-0002-5858-0488","position":24,"is_corresponding":false},{"id":347040,"name":"David Milewski","orcid":"0000-0002-9778-8548","position":0,"is_corresponding":true}],"reference_count":56,"raw_metadata":null,"created_at":"2026-07-19T00:23:30.364338Z","pmid":"36346688","pmcid":null,"fwci":null,"citation_percentile":null,"influential_citations":0,"oa_status":null,"license":null,"views":0,"total_file_size_bytes":0,"version_count":0,"fair_f":null,"fair_a":null,"fair_i":null,"fair_r":null,"fair_zscore":null,"fair_rationale":null,"fair_model":null,"fair_agent_version":null,"fair_fulltext_source":null,"fair_has_llm":null,"fair_computed_at":null,"clinical_trials":[],"software_tools":[],"db_accessions":[],"linked_datasets":[],"topics":[]}