Background: Immune checkpoint inhibitor (ICI) therapy is a cornerstone in the treatment of advanced non-small cell lung cancer (NSCLC), but immune-related cardiotoxicities remain a concern. Whether immune-related cardiotoxicities serve as independent prognostic predictors for NSCLC remains debated. This study aimed to investigate the prognostic value of immune-related cardiotoxicities and develop a predictive model for survival outcomes in NSCLC patients receiving ICI therapy. Methods: This retrospective multi-center study included NSCLC patients from nine centers who received at least one ICI cycle from 2019 to 2024. Demographic and clinical indicators were analyzed for associations with overall survival (OS). A prognostic model based on Cox regression and a nomogram was developed and validated. Results: Of 187 patients, the 1-, 3-, and 5-year survival rates of all participants was 0.903 [95% confidence dinterval (CI) 0.857-0.951], 0.759 (95% CI: 0.666-0.865) and 0.528 (95% CI: 0.361-0.773) respectively and 80 (42.8%) experienced immune-related cardiotoxicities. Kaplan-Meier survival analyses revealed that immune-related cardiotoxicities and immune-related adverse events (irAEs) were not associated with OS (P=0.10; P=0.20). Multivariate Cox regression analyses on OS showed an independent predictive value of platelet-to-lymphocyte ratio (PLR) [hazard ratio (HR) 1.002, 95% CI: 1.000-1.004, P=0.02], carcinoembryonic antigen (CEA) (HR 1.005, 95% CI: 1.002-1.007, P<0.001) and cytokeratin-19 fragment 21-1 (CYFRA21-1) (HR 1.004, 95% CI: 1.001-1.008, P=0.02) for OS in advanced NSCLC patients. A nomogram was constructed and the area under the curve (AUC) was 0.757, 0.741, and 0.815 at 1-, 3-, and 5-year. A prognostic model incorporating PLR, CEA and CYFRA21-1 resulted in an optimal performance to predict the prognosis. Additionally, smoking was associated with the occurrence of immune-related cardiotoxicities. Conclusions: Mild irAEs or immune-related cardiotoxicities (grade 1-2) did not extend median OS in NSCLC patients. The prognostic model incorporating PLR, CEA, and CYFRA21-1 could contribute to predicting the treatment efficacy and prognosis of ICI treatment.