A Dynamic Prognostic and Adaptive Treatment Framework for Advanced Biliary Tract Cancer
Rattachement africain : cn. Niveau de preuve : code pays fourni par la source.
Le résumé fourni par la source
BACKGROUND & AIMS: First-line immuno-chemotherapy is standard for advanced biliary tract cancer (BTC), but outcomes vary substantially, necessitating longitudinal monitoring and treatment adaptation. We aimed to develop models that dynamically update survival predictions using evolving clinical data to support real-time prognostic stratification. METHODS: We analysed patients with advanced BTC receiving first-line immuno-chemotherapy across eight centres. Cohorts comprised development, internal validation, two retrospective external validation, and one prospective external validation sets. The Bayesian joint model iDREM-BTC integrated baseline clinical and imaging variables with serial C-reactive protein, carbohydrate antigen 19-9, and total bilirubin measurements. iDREM(Pro)-BTC additionally incorporated baseline immunohistochemical and genomic data. Performance was assessed using dynamic area under the curve (AUC), calibration, and comparisons with baseline Cox models; interpretability was examined by ablation analysis (ClinicalTrials.gov: NCT06849193). RESULTS: Among 2314 patients (n=841, 360, 327, 284, and 502, respectively), machine learning identified age, ECOG performance status, tumour burden, tumour stage, and the three longitudinal biomarkers as mortality predictors. iDREM-BTC achieved overall dynamic AUCs of 0·730 (95% CI 0·689-0·794), 0·718 (0·670-0·778), 0·755 (0·707-0·808), 0·705 (0·639-0·773), and 0·745 (0·691-0·802), respectively. Discrimination improved over follow-up in all cohorts, with AUCs increasing from 0·633-0·705 at baseline to 0·778-0·810 at 6 months. iDREM(Pro)-BTC showed higher discrimination in development (n=628; AUC 0·807 [0·781-0·839]) and retained performance in external validation (n=281; 0·718 [0·669-0·787]). Exploratory matched analyses showed overall-survival separation among iDREM-BTC-defined high-risk patients; findings for iDREM(Pro)-BTC were directionally similar but not statistically significant. CONCLUSION: iDREM-BTC and iDREM(Pro)-BTC provide dynamically updated survival estimates during first-line immuno-chemotherapy and support individualized prognostic stratification, potentially informing treatment adjustment across diverse patient populations and immuno-chemotherapy regimens in clinical practice. IMPACT AND IMPLICATIONS: We developed the Individualized Dynamic Risk Estimation Model for biliary tract cancer (iDREM-BTC), a novel prognostic prediction and treatment recommendation system trained on data from 841 patients. The model demonstrated robust performance across multiple validation cohorts including 1473 patients. By integrating baseline Cox models with three mixed models incorporating longitudinal biomarkers (C-reactive protein level, carbohydrate antigen 19-9 level, and total bilirubin grade), iDREM-BTC enables real-time, accurate prognostic predictions, risk stratification, and treatment adjustments. The enhanced version, iDREM(Pro)-BTC, further incorporates immunohistochemistry and genomic markers, improving predictive latency while maintaining dynamic modelling advantages. iDREM-BTC and iDREM(Pro)-BTC can serve as valuable bedside resources for clinicians in the routine monitoring and treatment of patients with BTC. These models have also been integrated into an online platform as a research deployment. THE CLINICAL TRIAL NUMBER: NCT06849193.
Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.
Le contrôle bibliographique ouvert
DOI retrouvé dans Crossref DOI retrouvé ; titre concordant.
- Titre Crossref
- A Dynamic Prognostic and Adaptive Treatment Framework for Advanced Biliary Tract Cancer
- Date Crossref
- 01/08/2026
- Éditeur
- Elsevier BV
- Type
- journal-article
Ce recoupement confirme des métadonnées liées au DOI. Il ne confirme ni la méthode ni les conclusions de l’étude, et il ne compte pas comme une seconde source scientifique indépendante.
Où se fait cette recherche
-
Zhongda Hospital Southeast University pays non établi dans la noticeÉtablissement de santé
-
Chinese Academy of Medical Sciences & Peking Union Medical College pays non établi dans la noticeUniversité ou école supérieure
-
Shanghai Changzheng Hospital pays non établi dans la noticeÉtablissement de santé
-
First Affiliated Hospital of Zhengzhou University Department of Interventional Radiology pays non établi dans la noticeÉtablissement de santé
-
Chinese PLA General Hospital pays non établi dans la noticeÉtablissement de santé
-
The 309th Hospital of Chinese People's Liberation Army pays non établi dans la noticeÉtablissement de santé
-
Jiangsu Cancer Hospital Department of Medical Oncology pays non établi dans la noticeÉtablissement de santé
-
Nanjing Medical University pays non établi dans la noticeUniversité ou école supérieure
-
University of Science and Technology of China Department of Radiology pays non établi dans la noticeUniversité ou école supérieure
-
Soochow University Department of Oncology pays non établi dans la noticeUniversité ou école supérieure
-
Jiangsu University Department of Radiology pays non établi dans la noticeUniversité ou école supérieure
-
Jintan People's Hospital pays non établi dans la noticeÉtablissement de santé
Zhongda Hospital Southeast University, Chinese Academy of Medical Sciences & Peking Union Medical College et Shanghai Changzheng Hospital, avec 9 autres affiliations.
Une affiliation ne permet pas de déduire la nationalité d’un auteur.