{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,10]],"date-time":"2026-06-10T04:59:54Z","timestamp":1781067594221,"version":"3.54.1"},"reference-count":29,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,6,10]],"date-time":"2026-06-10T00:00:00Z","timestamp":1781049600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Med."],"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Accurate prediction of breast cancer recurrence is vital for optimizing adjuvant therapy intensity. However, the traditional TNM staging system often lacks precision in capturing individual risk due to biological heterogeneity. This study aimed to develop an interpretable machine learning model to refine risk stratification.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>\n                      We retrospectively analyzed clinical data from 578 breast cancer patients with a median follow-up duration of 62\u202fmonths. The cohort was randomly partitioned into a training set (\n                      <jats:italic>n<\/jats:italic>\n                      \u202f=\u202f405, 190 recurrence events) and an internal validation set (\n                      <jats:italic>n<\/jats:italic>\n                      \u202f=\u202f173, 72 recurrence events). An Extreme Gradient Boosting (XGBoost) model was developed using 15 clinicopathological features. Model interpretability was achieved using SHapley Additive exPlanations (SHAP), and a visual nomogram was constructed for clinical application.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      In the internal validation cohort, the XGBoost model demonstrated superior discriminative performance with an AUC of 0.877 (95% CI: 0.835\u20130.918), significantly outperforming logistic regression (AUC\u202f=\u202f0.693, 95% CI: 0.612\u20130.768) and the standard TNM system. SHAP analysis identified the Ki-67 index and positive lymph nodes as the most influential predictors, revealing non-linear risk associations. Crucially, the model successfully stratified patients within TNM Stages II and III into distinct high- and low-risk trajectories (Log-rank\n                      <jats:italic>p<\/jats:italic>\n                      \u202f&amp;lt;\u202f0.05), addressing intra stage heterogeneity.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>The proposed XGBoost-based framework provides a robust and interpretable tool for predicting 5-year recurrence, offering superior prognostic accuracy over standard anatomical staging. This approach holds promise for facilitating personalized clinical decision-making.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fmed.2026.1821410","type":"journal-article","created":{"date-parts":[[2026,6,10]],"date-time":"2026-06-10T04:29:37Z","timestamp":1781065777000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Development and validation of an interpretable machine learning model for predicting 5-year recurrence in breast cancer"],"prefix":"10.3389","volume":"13","author":[{"given":"Shaoda","family":"Meng","sequence":"first","affiliation":[{"name":"Department of Breast and Thyroid Surgery, The First People\u2019s Hospital of Yunnan Province","place":["Kunming, Yunnan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sicheng","family":"Liu","sequence":"additional","affiliation":[{"name":"Department of Medical Oncology, The First People\u2019s Hospital of Yunnan Province","place":["Kunming, Yunnan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Minghua","family":"Lai","sequence":"additional","affiliation":[{"name":"Department of Breast and Thyroid Surgery, The First People\u2019s Hospital of Yunnan Province","place":["Kunming, Yunnan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Li","family":"Li","sequence":"additional","affiliation":[{"name":"Department of Breast and Thyroid Surgery, The First People\u2019s Hospital of Yunnan Province","place":["Kunming, Yunnan, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,6,10]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","first-page":"40","DOI":"10.1186\/s12935-025-03678-1","article-title":"Rack1 promotes breast cancer stemness properties and tumorigenesis through the E2F1-SOX2 axis","volume":"25","author":"Jia","year":"2025","journal-title":"Cancer Cell Int"},{"key":"ref2","doi-asserted-by":"publisher","first-page":"e098846","DOI":"10.1136\/bmjopen-2025-098846","article-title":"Prognostic value of RANK signalling in breast cancer: a systematic review and meta-analysis","volume":"15","author":"Li","year":"2025","journal-title":"BMJ Open"},{"key":"ref3","doi-asserted-by":"publisher","first-page":"1357340","DOI":"10.3389\/fimmu.2024.1357340","article-title":"Human tissue-resident peritoneal macrophages reveal resistance towards oxidative cell stress induced by non-invasive physical plasma","volume":"15","author":"Schultze-Rhonhof","year":"2024","journal-title":"Front Immunol"},{"key":"ref4","doi-asserted-by":"publisher","first-page":"1402","DOI":"10.1097\/JS9.0000000000003581","article-title":"Deep learning for prognostic stratification and biomarker exploration in upper tract urothelial carcinoma: a multicenter retrospective cohort study","volume":"112","author":"Peng","year":"2026","journal-title":"Int J Surg"},{"key":"ref5","doi-asserted-by":"publisher","first-page":"117","DOI":"10.1186\/s12876-023-02749-9","article-title":"Development and validation of nomograms for predicting overall survival and cancer-specific survival in elderly patients with locally advanced gastric cancer: a population-based study","volume":"23","author":"Sun","year":"2023","journal-title":"BMC Gastroenterol"},{"key":"ref6","doi-asserted-by":"publisher","first-page":"15330338251336032","DOI":"10.1177\/15330338251336032","article-title":"Extracellular volume derived from equilibrium CT for the prediction of survival outcomes in patients with pancreatic ductal adenocarcinoma","volume":"24","author":"Xiong","year":"2025","journal-title":"Technol Cancer Res Treat"},{"key":"ref7","doi-asserted-by":"publisher","first-page":"868004","DOI":"10.3389\/fonc.2022.868004","article-title":"Prevalence and Clinicopathologic features of canine metastatic melanoma involving the central nervous system: a retrospective analysis and comparative review","volume":"12","author":"Razmara","year":"2022","journal-title":"Front Oncol"},{"key":"ref8","doi-asserted-by":"publisher","first-page":"792","DOI":"10.1186\/s12885-020-07267-5","article-title":"Effectiveness of the AJCC 8th edition staging system for selecting patients with T1-2N1 breast cancer for post-mastectomy radiotherapy: a joint analysis of 1986 patients from two institutions","volume":"20","author":"Wang","year":"2020","journal-title":"BMC Cancer"},{"key":"ref9","doi-asserted-by":"publisher","first-page":"996","DOI":"10.1007\/s12094-015-1472-y","article-title":"Ki-67 is a prognostic marker for hormone receptor positive tumors","volume":"18","author":"Perez-Lopez","year":"2016","journal-title":"Clin Transl Oncol"},{"key":"ref10","doi-asserted-by":"publisher","first-page":"108","DOI":"10.1186\/s12935-019-0818-x","article-title":"MiR-591 functions as tumor suppressor in breast cancer by targeting TCF4 and inhibits hippo-YAP\/TAZ signaling pathway","volume":"19","author":"Huang","year":"2019","journal-title":"Cancer Cell Int"},{"key":"ref11","doi-asserted-by":"publisher","first-page":"6258","DOI":"10.3748\/wjg.v25.i41.6258","article-title":"Blood parameters score predicts long-term outcomes in stage II-III gastric cancer patients","volume":"25","author":"Lin","year":"2019","journal-title":"World J Gastroenterol"},{"key":"ref12","doi-asserted-by":"publisher","first-page":"6603","DOI":"10.2147\/CMAR.S204346","article-title":"The impact of the patient's condition, diagnostic procedures and treatment on the survival of carcinoma of unknown primary site patients","volume":"11","author":"Dorobisz","year":"2019","journal-title":"Cancer Manag Res"},{"key":"ref13","doi-asserted-by":"publisher","first-page":"39082","DOI":"10.1038\/s41598-025-26903-3","article-title":"Predictors and nomogram for amputation risk in pit viper snakebite envenoming at hospital admission","volume":"15","author":"Fu","year":"2025","journal-title":"Sci Rep"},{"key":"ref14","doi-asserted-by":"publisher","first-page":"107831","DOI":"10.1016\/j.asoc.2021.107831","article-title":"SANE: a sequence combined attentive network embedding model for COVID-19 drug repositioning","volume":"111","author":"Su","year":"2021","journal-title":"Appl Soft Comput"},{"key":"ref15","doi-asserted-by":"publisher","first-page":"111058","DOI":"10.1016\/j.isci.2024.111058","article-title":"Early prediction of acute pancreatitis with acute kidney injury using abdominal contrast-enhanced CT features","volume":"27","author":"Yuan","year":"2024","journal-title":"iScience"},{"key":"ref16","doi-asserted-by":"publisher","first-page":"890","DOI":"10.1136\/thorax-2023-221253","article-title":"Lung cancer risk assessment by prediction model: a global perspective","volume":"80","author":"Zhang","year":"2025","journal-title":"Thorax"},{"key":"ref17","doi-asserted-by":"publisher","first-page":"148","DOI":"10.3390\/mps8060148","article-title":"A study protocol on risk prediction modelling of mortality and in-hospital major bleeding following percutaneous coronary intervention in an Australian population: machine learning approach","volume":"8","author":"Chowdhury","year":"2025","journal-title":"Methods Protoc"},{"key":"ref18","doi-asserted-by":"publisher","first-page":"e70124","DOI":"10.1002\/clc.70124","article-title":"Predicting in-hospital mortality in patients with acute myocardial infarction: a comparison of machine learning approaches","volume":"48","author":"Soleimani","year":"2025","journal-title":"Clin Cardiol"},{"key":"ref19","doi-asserted-by":"publisher","first-page":"19561","DOI":"10.1038\/s41598-023-46930-2","article-title":"Calculation of exact Shapley values for explaining support vector machine models using the radial basis function kernel","volume":"13","author":"Mastropietro","year":"2023","journal-title":"Sci Rep"},{"key":"ref20","doi-asserted-by":"publisher","first-page":"1264803","DOI":"10.3389\/fbinf.2023.1264803","article-title":"Editorial: transparent machine learning in bio-medicine","volume":"3","author":"Diaz Ochoa","year":"2023","journal-title":"Front Bioinform"},{"key":"ref21","doi-asserted-by":"publisher","first-page":"1659","DOI":"10.1038\/s41598-025-31260-2","article-title":"An XGBoost-SHAP analysis of the driving factors of carbon emissions in China's first-tier cities","volume":"16","author":"Li","year":"2025","journal-title":"Sci Rep"},{"key":"ref22","doi-asserted-by":"publisher","first-page":"e0327120","DOI":"10.1371\/journal.pone.0327120","article-title":"Enhancing diabetes risk prediction through focal active learning and machine learning models","volume":"20","author":"Zhang","year":"2025","journal-title":"PLoS One"},{"key":"ref23","doi-asserted-by":"publisher","first-page":"409","DOI":"10.1007\/s12672-025-02197-9","article-title":"Comprehensive identification of a migrasomes-associated long non-coding RNA signature to predict the prognosis and treatment options in colon adenocarcinoma","volume":"16","author":"Zheng","year":"2025","journal-title":"Discov Oncol"},{"key":"ref24","doi-asserted-by":"publisher","first-page":"19","DOI":"10.1038\/s41698-024-00794-5","article-title":"Integrins identified as potential prognostic markers in osteosarcoma through multi-omics and multi-dataset analysis","volume":"9","author":"Cui","year":"2025","journal-title":"NPJ Precis Oncol"},{"key":"ref25","doi-asserted-by":"publisher","first-page":"e41585","DOI":"10.1097\/MD.0000000000041585","article-title":"Threshold effect of non-high-density lipoprotein to high-density lipoprotein cholesterol ratio and hypertension in U.S. adults: NHANES 2005\u20132016","volume":"104","author":"Sha","year":"2025","journal-title":"Medicine"},{"key":"ref26","doi-asserted-by":"publisher","first-page":"6","DOI":"10.1186\/s12911-025-03302-2","article-title":"Exploring prognostic factors on vascular outcomes among maintenance dialysis patients and establishing a prognosis prediction model using machine learning methods","volume":"26","author":"Wu","year":"2025","journal-title":"BMC Med Inform Decis Mak"},{"key":"ref27","doi-asserted-by":"publisher","first-page":"1254","DOI":"10.3390\/diagnostics12051254","article-title":"Lymph node metastases identified at the post-ablation 131I SPECT\/CT scan is a prognostic factor of intermediate-risk papillary thyroid Cancer","volume":"12","author":"Jia","year":"2022","journal-title":"Diagnostics"},{"key":"ref28","doi-asserted-by":"publisher","first-page":"2587490","DOI":"10.1080\/0886022X.2025.2587490","article-title":"Systematic review and comparison of machine learning and conventional statistical models for predicting cardiovascular events in dialysis patients","volume":"47","author":"Lu","year":"2025","journal-title":"Ren Fail"},{"key":"ref29","doi-asserted-by":"publisher","first-page":"981","DOI":"10.1186\/s12885-025-14331-5","article-title":"Synergistic effects of PD-1 antibody and chemotherapy followed by surgery-centric local treatment in patients with limited-metastatic gastric or gastroesophageal adenocarcinoma (ROSETTE trial): an open-label, single-center, randomized phase 2 trial","volume":"25","author":"Wu","year":"2025","journal-title":"BMC Cancer"}],"container-title":["Frontiers in Medicine"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2026.1821410\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,10]],"date-time":"2026-06-10T04:29:37Z","timestamp":1781065777000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2026.1821410\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,10]]},"references-count":29,"alternative-id":["10.3389\/fmed.2026.1821410"],"URL":"https:\/\/doi.org\/10.3389\/fmed.2026.1821410","relation":{},"ISSN":["2296-858X"],"issn-type":[{"value":"2296-858X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,10]]},"article-number":"1821410"}}