The present study analyzed the impact of age on the causes of death (CODs) in patients with nasopharyngeal carcinoma (NPC) undergoing chemoradiotherapy (CRT) using machine learning approaches. A total of 2841 patients (1037 classified as older, ≥ 60 years and 1804 as younger, < 60 years) were enrolled. Variations in the CODs between the two age groups were analyzed before and after applying inverse probability of treatment weighting (IPTW). Additionally, seven different machine learning models were employed as predictive tools to identify key variables and assess the therapeutic outcomes in NPC patients receiving CRT. The younger group exhibited a significantly longer overall survival (OS) than the older group, both before the IPTW adjustment (140 vs. 50 months, P < 0.001) and after the adjustment (137 vs. 53 months, P < 0.001). After IPTW, the older group was associated with worse 5-, 10-, and 15-year cumulative incidences in terms of NPC-related deaths (30, 34, and 38% vs. 21, 27, and 30%; P < 0.001), cardiovascular disease (CVD; 4.1, 7.2, and 8.8% vs. 0.5, 1.8, and 3.0%; P < 0.001), and other causes (8.3, 17, and 24% vs. 4.1, 8.7, and 12%; P < 0.001). However, cumulative incidences of secondary malignant neoplasms were comparable between the two groups (P = 0.100). The random forest (RF) model demonstrated the highest concordance index of 0.701 among all models. Time-dependent variable importance plots indicated that age was the most influential factor affecting 3-, 5-, and 10-year survival, followed by metastasis and tumor stage. Younger patients had significantly longer OS than their older counterparts. Older patients had a higher likelihood of dying from non-NPC-related causes, particularly CVDs. The RF model showed the best predictive accuracy, identifying age as the most critical factor influencing OS in NPC patients undergoing CRT.
Keywords: Age; Cause of death; Chemoradiotherapy; Machine learning models; Nasopharyngeal carcinoma.
© 2025. The Author(s).