Abstract
BACKGROUND: Immune checkpoint inhibitors (ICIs) have improved survival for non-small cell lung cancer (NSCLC) patients, but immune-related adverse events (irAEs), like immune-mediated thyroid dysfunction (IMTD), have been reported. IMTD causes irreversible thyroid damage, affecting NSCLC patients' quality of life. This study aims to explore IMTD risk factors and develop a Nomogram to predict IMTD risk at 6, 12, and 24 months. METHODS: Data from 1,917 NSCLC patients from Chongqing University Cancer Hospital treated with ICIs were randomly split into training (70%) and validation (30%) cohorts. After variable selection, a Nomogram with 11 common clinical variables was built from the training cohort. The validation cohort was used to assess the model comprehensively using the Time C-index, Time AUC, Delong test, calibration curves, and decision curve analysis (DCA) to ensure its clinical effectiveness. RESULTS: IMTD occurred in 343 (17.89%) patients. Among the 11 model factors, Age (OR = 1.02, 95% CI: 1.01 - 1.04), Female (OR = 1.78, 95% CI: 1.31 - 2.42), Mono (OR = 3.52, 95% CI: 1.72 - 7.17), and TCHO (OR = 1.13, 95% CI: 1.03 - 1.24) were significant IMTD risk factors. WBC and FT4 were protective factors (OR = 0.9, 95% CI: 0.83 - 0.98 and OR = 0.94, 95% CI: 0.90 - 0.97). The Nomogram showed good predictive accuracy and generalizability in both cohorts, with C - indices of 0.77 (95% CI: 0.74 - 0.80) and 0.72 (95% CI: 0.67 - 0.78), and AUC values above 0.7. Kaplan - Meier curves confirmed its effective IMTD risk stratification. CONCLUSION: The developed Nomogram has good predictive performance and can identify high-risk IMTD patients. The web calculators are user-friendly, providing a basis for early clinical intervention to reduce IMTD incidence.