{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,16]],"date-time":"2026-09-16T15:07:56Z","timestamp":1789571276292,"version":"build-2803163510"},"reference-count":24,"publisher":"Ovid Technologies (Wolters Kluwer Health)","issue":"18","content-domain":{"domain":["lww.com","ovid.com"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2026,9,16]]},"abstract":"<jats:sec>\n                    <jats:title>Background:<\/jats:title>\n                    <jats:p>Osteosarcoma treatment decisions require accurate prognostic assessment, yet utilization of machine learning (ML) models is problematic because performance degrades consistently when models are applied across data sets. Current single-data-set models (single models) learn population-specific patterns rather than generalizable disease characteristics, limiting their clinical implementation. We developed a multicomponent-model (multi-model) ML framework using domain-adversarial training across 2 national registries, integrating structured clinical variables with text-based patient data to learn generalizable disease patterns and achieve reliable survival prediction across diverse populations.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods:<\/jats:title>\n                    <jats:p>We conducted a retrospective study using data from 2 national cancer registries: SEER (Surveillance, Epidemiology, and End Results; n = 4,278 patients, 2004 to 2015) and NCDB (National Cancer Database; n = 4,049 patients, 2004 to 2018). We compared the cross-data-set performance of single models versus the multi-models. Primary outcomes were performance metrics for 2-year and 5-year overall survival predictions, measured by the area under the receiver operating characteristic curve (AUC), precision, recall, F1-score, and Brier score.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results:<\/jats:title>\n                    <jats:p>Single models achieved strong internal validation performance (AUC, 0.898 to 0.927) but performance declined substantially in cross-data-set validation (AUC, 0.563 to 0.665). The multi-model approach achieved cross-data-set AUCs of 0.708 to 0.843 for 2-year survival and 0.648 to 0.798 for 5-year survival, with improvements of 0.085 to 0.199 over single models across all evaluation metrics.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions:<\/jats:title>\n                    <jats:p>The multi-model ML approach demonstrated improved osteosarcoma prognostic ability across health-care data sets, addressing the generalizability challenges of models based on a single data set. Enhanced cross-data-set performance suggests the potential for consistent risk stratification to guide surgical planning, adjuvant therapy selection, and patient counseling. Prospective validation is needed to evaluate clinical impact.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Level of Evidence:<\/jats:title>\n                    <jats:p>\n                      Prognostic\n                      <jats:underline>Level III<\/jats:underline>\n                      . See Instructions for Authors for a complete description of levels of evidence.\n                    <\/jats:p>\n                  <\/jats:sec>","DOI":"10.2106\/jbjs.25.01327","type":"journal-article","created":{"date-parts":[[2026,5,26]],"date-time":"2026-05-26T18:00:18Z","timestamp":1779818418000},"page":"e17","update-policy":"https:\/\/doi.org\/10.1097\/lww.0000000000001000","source":"Crossref","is-referenced-by-count":0,"title":["Enhancing Osteosarcoma Survival Predictions"],"prefix":"10.2106","volume":"108","author":[{"ORCID":"https:\/\/orcid.org\/0009-0000-5627-8982","authenticated-orcid":false,"given":"Bishoy M.","family":"Galoaa","sequence":"first","affiliation":[{"name":"Orthopaedic Oncology Service, Department of Orthopaedic Surgery, Massachusetts General Hospital, Boston, Massachusetts"},{"name":"Northeastern University College of Engineering, Boston, Massachusetts"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6201-6230","authenticated-orcid":false,"given":"Sonia E.","family":"Ubong","sequence":"additional","affiliation":[{"name":"Orthopaedic Oncology Service, Department of Orthopaedic Surgery, Massachusetts General Hospital, Boston, Massachusetts"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-5170-0144","authenticated-orcid":false,"given":"Andrew G.","family":"Girgis","sequence":"additional","affiliation":[{"name":"Orthopaedic Oncology Service, Department of Orthopaedic Surgery, Massachusetts General Hospital, Boston, Massachusetts"},{"name":"Boston University Chobanian and Avedisian School of Medicine, Boston, Massachusetts"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3352-6362","authenticated-orcid":false,"given":"Marcos R.","family":"Gonzalez","sequence":"additional","affiliation":[{"name":"Orthopaedic Oncology Service, Department of Orthopaedic Surgery, Massachusetts General Hospital, Boston, Massachusetts"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Santiago A.","family":"Lozano-Calderon","sequence":"additional","affiliation":[{"name":"Orthopaedic Oncology Service, Department of Orthopaedic Surgery, Massachusetts General Hospital, Boston, Massachusetts"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"276","published-online":{"date-parts":[[2026,5,26]]},"reference":[{"issue":"1","key":"bib2-20260916","doi-asserted-by":"crossref","first-page":"7327","DOI":"10.1038\/s41598-024-57527-8","article-title":"Identification of an early survival prognostic gene signature for localized osteosarcoma patients","volume":"14","author":"Patel","year":"2024","journal-title":"Sci Rep."},{"key":"bib3-20260916","article-title":"The epidemiology of osteosarcoma","volume":"152","author":"Ottaviani","year":"2009","journal-title":"Cancer Treat Res."},{"issue":"1","key":"bib4-20260916","doi-asserted-by":"crossref","first-page":"290","DOI":"10.1186\/s12885-025-13496-3","article-title":"Comparison of overall survival of adult and pediatric osteosarcoma patients using the national cancer database","volume":"25","author":"Boyland","year":"2025","journal-title":"BMC Cancer."},{"key":"bib5-20260916","doi-asserted-by":"crossref","first-page":"780264","DOI":"10.3389\/fonc.2021.780264","article-title":"Osteosarcoma and Metastasis","volume":"11","author":"Sheng","year":"2021","journal-title":"Front Oncol."},{"issue":"2","key":"bib6-20260916","doi-asserted-by":"crossref","first-page":"360","DOI":"10.3390\/cancers14020360","article-title":"Metastatic Progression of Osteosarcomas: A Review of Current Knowledge of Environmental versus Oncogenic Drivers","volume":"14","author":"Odri","year":"2022","journal-title":"Cancers (Basel)."},{"issue":"8","key":"bib7-20260916","doi-asserted-by":"crossref","first-page":"2130","DOI":"10.3390\/cancers12082130","article-title":"Molecular Biology of Osteosarcoma","volume":"12","author":"Czarnecka","year":"2020","journal-title":"Cancers (Basel)."},{"issue":"1","key":"bib8-20260916","doi-asserted-by":"crossref","first-page":"17386","DOI":"10.1038\/s41598-025-00179-z","article-title":"Feasibility of machine learning-based modeling and prediction to assess osteosarcoma outcomes","volume":"15","author":"Zhao","year":"2025","journal-title":"Sci Rep."},{"issue":"2","key":"bib9-20260916","doi-asserted-by":"crossref","DOI":"10.1093\/jjco\/hyy157","article-title":"New TNM classification (AJCC eighth edition) of bone and soft tissue sarcomas: JCOG Bone and Soft Tissue Tumor Study Group","volume":"49","author":"Tanaka","year":"2019","journal-title":"Jpn J Clin Oncol."},{"issue":"3","key":"bib10-20260916","doi-asserted-by":"crossref","DOI":"10.1245\/s10434-007-9747-3","article-title":"The National Cancer Data Base: a powerful initiative to improve cancer care in the United States","volume":"15","author":"Bilimoria","year":"2008","journal-title":"Ann Surg Oncol."},{"issue":"9","key":"bib11-20260916","doi-asserted-by":"crossref","DOI":"10.1002\/ijc.34409","article-title":"Incidence and survival of soft tissue sarcoma in England between 2013 and 2017, an analysis from the National Cancer Registration and Analysis Service","volume":"152","author":"Bacon","year":"2023","journal-title":"Int J Cancer."},{"issue":"3","key":"bib12-20260916","article-title":"Survival nomogram for osteosarcoma patients: SEER data retrospective analysis with external validation","volume":"13","author":"Liu","year":"2023","journal-title":"Am J Cancer Res."},{"issue":"2","key":"bib13-20260916","doi-asserted-by":"crossref","DOI":"10.1097\/CORR.0000000000003701","article-title":"Bias or Best Fit? A Comparative Analysis of the SEER and NCDB Data Sets in Single-model Machine Learning for Predicting Osteosarcoma Survival Outcomes","volume":"484","author":"Girgis","year":"2026","journal-title":"Clin Orthop Relat Res."},{"issue":"1","key":"bib14-20260916","doi-asserted-by":"crossref","DOI":"10.1080\/14737167.2023.2279107","article-title":"Systematic reviews of machine learning in healthcare: a literature review","volume":"24","author":"Kolasa","year":"2024","journal-title":"Expert Rev Pharmacoecon Outcomes Res."},{"issue":"1","key":"bib15-20260916","doi-asserted-by":"crossref","first-page":"5","DOI":"10.1186\/s12911-024-02830-7","article-title":"External validation of AI-based scoring systems in the ICU: a systematic review and meta-analysis","volume":"25","author":"Rockenschaub","year":"2025","journal-title":"BMC Med Inform Decis Mak."},{"issue":"39","key":"bib16-20260916","doi-asserted-by":"crossref","first-page":"e39582","DOI":"10.1097\/MD.0000000000039582","article-title":"Machine learning-based individualized survival prediction model for prognosis in osteosarcoma: Data from the SEER database","volume":"103","author":"Cao","year":"2024","journal-title":"Medicine (Baltimore)."},{"issue":"1","key":"bib17-20260916","doi-asserted-by":"crossref","first-page":"5542","DOI":"10.1038\/s41598-021-85223-4","article-title":"Predictive model for the 5-year survival status of osteosarcoma patients based on the SEER database and XGBoost algorithm","volume":"11","author":"Jiang","year":"2021","journal-title":"Sci Rep."},{"key":"bib18-20260916","doi-asserted-by":"crossref","first-page":"972352","DOI":"10.3389\/fgene.2022.972352","article-title":"Prognostic significance of modified lung immune prognostic index in osteosarcoma patients","volume":"13","author":"He","year":"2022","journal-title":"Front Genet."},{"issue":"4","key":"bib19-20260916","doi-asserted-by":"crossref","DOI":"10.1007\/s11912-010-0106-3","article-title":"Stratifying osteosarcoma: minimizing and maximizing therapy","volume":"12","author":"Niswander","year":"2010","journal-title":"Curr Oncol Rep."},{"issue":"10","key":"bib20-20260916","doi-asserted-by":"crossref","DOI":"10.1016\/S1470-2045(16)30214-5","article-title":"Comparison of MAPIE versus MAP in patients with a poor response to preoperative chemotherapy for newly diagnosed high-grade osteosarcoma (EURAMOS-1): an open-label, international, randomised controlled trial","volume":"17","author":"Marina","year":"2016","journal-title":"Lancet Oncol."},{"key":"bib22-20260916","doi-asserted-by":"crossref","first-page":"548151","DOI":"10.1155\/2011\/548151","article-title":"Using epidemiology and genomics to understand osteosarcoma etiology","volume":"2011","author":"Savage","year":"2011","journal-title":"Sarcoma."},{"issue":"6","key":"bib23-20260916","doi-asserted-by":"crossref","DOI":"10.1097\/01.sla.0000250438.04393.a8","article-title":"Should soft tissue sarcomas be treated at high-volume centers? An analysis of 4205 patients","volume":"245","author":"Gutierrez","year":"2007","journal-title":"Ann Surg."},{"issue":"5","key":"bib24-20260916","doi-asserted-by":"crossref","DOI":"10.1093\/annonc\/mdq589","article-title":"The influence of tumor- and treatment-related factors on the development of local recurrence in osteosarcoma after adequate surgery. An analysis of 1355 patients treated on neoadjuvant Cooperative Osteosarcoma Study Group protocols","volume":"22","author":"Andreou","year":"2011","journal-title":"Ann Oncol."},{"issue":"3","key":"bib25-20260916","doi-asserted-by":"crossref","DOI":"10.1200\/JCO.2002.20.3.776","article-title":"Prognostic factors in high-grade osteosarcoma of the extremities or trunk: an analysis of 1,702 patients treated on neoadjuvant cooperative osteosarcoma study group protocols","volume":"20","author":"Bielack","year":"2002","journal-title":"J Clin Oncol."},{"key":"bib26-20260916","article-title":"Disparities in Cancer Care: The Example of Sarcoma-In Search of Solutions for a Global Issue","volume":"41","author":"Alcindor","year":"2021","journal-title":"Am Soc Clin Oncol Educ Book."}],"container-title":["Journal of Bone and Joint Surgery"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.ovid.com\/10.2106\/JBJS.25.01327","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,16]],"date-time":"2026-09-16T15:01:00Z","timestamp":1789570860000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.ovid.com\/10.2106\/JBJS.25.01327"}},"subtitle":["A Comparative Study of a Multicomponent-Model Machine Learning Approach Integrating SEER and NCDB Data Sets Versus Conventional Single-Data-Set Modeling"],"short-title":[],"issued":{"date-parts":[[2026,5,26]]},"references-count":24,"journal-issue":{"issue":"18","published-print":{"date-parts":[[2026]]}},"URL":"https:\/\/doi.org\/10.2106\/jbjs.25.01327","relation":{},"ISSN":["0021-9355","1535-1386"],"issn-type":[{"value":"0021-9355","type":"print"},{"value":"1535-1386","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,26]]}}}