初诊边缘区淋巴瘤早期进展预后模型的开发与验证:一项多中心回顾性研究OA

Objective Marginal zone lymphoma(MZL)exhibits significant heterogeneity in its prognostic course.This study aims to identify clinical prognostic factors associated with patient outcomes and construct a predictive model.Methods This study integrates clinical data from 125 newly diagnosed MZL patients across multiple centers,uses the LASSO-Cox method to select variables,and validates the findings in an independent external validation cohort.Results Five independent prognostic factors were identified through screening:gender,B symptoms,thrombocytopenia,hypoalbuminemia,and age.Based on these factors,the ASPAB-PI prognostic model for MZL was established.Internal validation yielded a C-index of 0.830(95％CI 0.76-0.90),and external validation yielded a C-index of 0.811(95％CI 0.65-0.98),demonstrating superior predictive capability compared to the existing MZL-1PI model.Risk stratification based on optimal cut-off points resulted in significantly separated survival curves for the low,intermediate,and high-risk groups in both the training and validation sets.The 24-month PFS rates for these groups were 91.92％,75.89％and 10.71％,respectively.Conclusion The five-factor MZL prediction model developed in this study has been validated as robust and feasible,effectively enabling individualized risk assessment.