Due to drug resistance, a majority of patients with non-small cell lung cancer (NSCLC) experience disease progression following immunotherapy. Therefore, there is an urgent need to develop novel biomarkers to predict the prognosis of NSCLC patients. Clinical data from 544 patients with advanced NSCLC who underwent immune checkpoint blockers (ICBs) at our clinical center were collected in this study. The results indicated that low Albumin-Globulin Ratio (AGR) and Lymphocyte-Monocyte Ratio (LMR) and high Systemic Immune-Inflammation Index (SIRI) were significantly correlated with both poor overall survival (OS) and progression-free survival (PFS) in NSCLC patients (P <
0.01). These three indicators collectively formed the most effective combined model for predicting the prognosis of NSCLC. Importantly, risk stratification based on AGR, LMR and SIRI was better than that based on the TNM stage, and served as an independent predictor of OS and PFS. Notably, the nomogram model developed by risk stratification, sex, age, smoking history, and pathological type demonstrated a good ability to predict the 1 to 5-year OS rates for NSCLC patients. In summary, AGR, LMR, and SIRI represented the optimal combined models for forecasting the prognosis of patients with advanced NSCLC who underwent ICBs, offering promising potential as biomarkers to direct personalized clinical interventions.