Abstract
AIM: To develop and validate prognostic nomograms incorporating peripheral blood lymphocyte counts and their dynamic changes in patients with nasopharyngeal carcinoma (NPC) undergoing radiotherapy. METHODS: In this retrospective cohort study, consecutive patients with NPC who received radiotherapy at Chongqing University Cancer Hospital were included as the internal cohort (randomly divided 70 %/30 % for training and validation), while patients treated at Guangxi Medical University Cancer Hospital served as the external validation cohort. Prognostic nomograms for progression-free survival (PFS) and overall survival (OS) were constructed using multivariable Cox regression analyses. RESULTS: The internal and external cohorts comprised 746 and 138 patients, respectively. Age, gross tumor volume dose, neoadjuvant chemotherapy, clinical stage, plasma EBV-DNA level, baseline total T-cell count, and its post-treatment change (ΔT cells) were identified as independent prognostic factors. The nomograms demonstrated strong predictive performance, with concordance indices of 0.701, 0.716, and 0.714 for PFS, and 0.759, 0.735, and 0.734 for OS in the training, internal validation, and external datasets, respectively. The areas under the receiver operating characteristic curves for 3-year and 5-year PFS and OS exceeded 0.7 across all cohorts. Calibration plots indicated good agreement between predicted and observed outcomes, and decision curve analysis confirmed greater net clinical benefit compared with TNM staging and EBV-DNA-based models. CONCLUSION: The proposed T-cell-based nomograms reliably predict PFS and OS in patients with NPC undergoing radiotherapy and were validated in an external cohort. These models provide improved prognostic discrimination beyond conventional staging systems and may assist in individualized risk stratification and management planning.