The administration of immune checkpoint inhibitors (ICIs) in advanced non-small cell lung cancer (NSCLC) with oncogenic driver alterations other than epidermal growth factor receptor (EGFR) aroused a heated discussion. We thus aimed to evaluate ICI treatment in these patients in real-world routine clinical practice.A multicenter, retrospective study was conducted for NSCLC patients with at least one gene alteration (KRAS, HER2, BRAF, MET, RET, ALK, ROS1) receiving ICI monotherapy or combination treatment. The data regarding clinicopathologic characteristics, clinical efficacy, and safety were investigated.A total of 216 patients were included, the median age was 60 years, 72.7% of patients were male, and 46.8% had a smoking history. The molecular alterations involved KRAS (n=95), HER2 (n=42), BRAF (n=22), MET (n=21), RET (n=14), ALK (n=14), and ROS1 (n=8); 56.5% of patients received immunotherapy in the first-line, and the rest 43.5% were treated as a second-line and above. For the entire cohort who received immunotherapy-based regimens in the first-line, the median progression-free survival (PFS) was 7.5 months and the median overall survival (OS) was 24.8 months. For the entire cohort who received immunotherapy-based regimens in the second-line and above, the median PFS was 4.7 months and median OS was 17.1 months. KRAS mutated NSCLC treated with immunotherapy-based regimens in the first-line setting had a median PFS and OS were 7.8 and 26.1 months, respectively. Moreover, the median PFS and OS of immunotherapy-based regimens for KRAS-mutant NSCLC that progressed after chemotherapy were 5.9 and 17.1 months. Programmed death ligand 1 (PD-L1) expression level was not consistently associated with response to immunotherapy across different gene alteration subsets. In the KRAS group, PD-L1 positivity [tumor proportion score (TPS) ≥1%] was associated with better PFS and OS according to the multivariate Cox analysis. No statistically significant association was found for smoking status, age, or gender with clinical efficacy in any gene group analyses.KRAS-mutant NSCLC could obtain clinical benefits from ICIs either for treatment-naive patients or those who have experienced progression after chemotherapy, and PD-L1 positive expression (TPS >1%) may be a potential positive predictor. For NSCLC with ALK, RET and ROS1 rearrangement, MET exon 14 skipping mutation, or BRAF V600E mutation, effectiveness of single or combined ICI therapy remains limited, therefore, targeted therapies should be considered prior to immunotherapy regimens. Future studies should address the investigation of better predictive biomarkers for immunotherapy response in oncogene-driven NSCLC.2024 Translational Lung Cancer Research. All rights reserved.