基于 CA 分期的新型预后模型用于增强自然杀伤/T 细胞淋巴瘤患者的分层和生存预测

This study aimed to identify key prognostic factors for Natural killer/T-cell lymphoma (NKTCL) in the context of L-asparaginase/pegaspargase-based therapy and to develop a simplified yet accurate prognostic model for risk stratification. Data from 854 NKTCL patients at the Sun Yat-sen University Cancer Center were divided into a training cohort (n = 598) and an internal validation cohort (n = 256). A further 222 patients from Sichuan Cancer Hospital & Institute were used to create an external validation cohort. Least absolute shrinkage and selection operator (LASSO) and Cox regression were used to identify independent risk factors for overall survival (OS).

A nomogram (Nomogram-CA) was constructed and evaluated using the consistency index (C-index), calibration curves, time-dependent ROC (tdROC) and decision curve analysis (DCA). Kaplan-Meier survival curves were generated to show the difference in OS between groups.

Age, CA stage, B symptoms and hemoglobin (Hb) level were all identified as independent risk factors for OS. Nomogram-CA was constructed based on multivariate analysis results. The DCA curves demonstrated that Nomogram-CA provided more net benefit to patients in the training, internal validation and external validation cohorts.

Furthermore, analysis of the Kaplan-Meier survival curve revealed a significantly lower survival rate among patients identified as high-risk by Nomogram-CA when compared to those classified as low-risk (P < 0. 05). Nomogram-CA constructed based on independent prognostic factors has better predictive ability compared to the traditional staging system, which can assist clinical doctors in evaluating patient prognosis.