Radiation Oncology (Aug 2024)
A nomogram based on circulating CD8+ T cell and platelet-to-lymphocyte ratio to predict overall survival of patients with locally advanced nasopharyngeal carcinoma
Abstract
Abstract Purpose To explore the influence of circulating lymphocyte subsets, serum markers, clinical factors, and their impact on overall survival (OS) in locally advanced nasopharyngeal carcinoma (LA-NPC). Additionally, to construct a nomogram predicting OS for LA-NPC patients using independent prognostic factors. Methods A total of 530 patients with LA-NPC were included in this study. In the training cohort, Cox regression analysis was utilized to identify independent prognostic factors, which were then integrated into the nomogram. The concordance index (C-index) was calculated for both training and validation cohorts. Schoenfeld residual analysis, calibration curves, and decision curve analysis (DCA) were employed to evaluate the nomogram. Kaplan-Meier methods was performed based on risk stratification using the nomogram. Results A total of 530 LA-NPC patients were included. Multivariate Cox regression analysis revealed that the circulating CD8+T cell, platelet-to-lymphocyte ratio (PLR), lactate dehydrogenase (LDH), albumin (ALB), gender, and clinical stage were independent prognostic factors for LA-NPC (p < 0.05). Schoenfeld residual analysis indicated overall satisfaction of the proportional hazards assumption for the Cox regression model. The C-index of the nomogram was 0.724 (95% CI: 0.669–0.779) for the training cohort and 0.718 (95% CI: 0.636-0.800) for the validation cohort. Calibration curves demonstrated good correlation between the model and actual survival outcomes. DCA confirmed the clinical utility enhancement of the nomogram over the TNM staging system. Significant differences were observed in OS among different risk stratifications. Conclusion Circulating CD8+ T cell, PLR, LDH, ALB, gender and clinical stage are independent prognostic factors for LA-NPC. The nomogram and risk stratification constructed in this study effectively predict OS in LA-NPC.
Keywords