Abstract
Background: Immune checkpoint inhibitors (ICIs) have transformed treatment for non-small cell lung cancer (NSCLC), but their efficacy varies across oncogenic driver alterations. The benefit of adding ICIs to platinum-based chemotherapy in rare molecular subgroups remains uncertain.Patients and Methods: In this retrospective multicenter study, we analyzed 902 patients with NSCLC harboring ALK, ROS1, or RET fusions; MET exon 14 skipping; BRAF V600E; KRAS; or HER2 mutations across 45 institutions in Japan. Objective response rate (ORR), progression-free survival (PFS), overall survival, and immune-related adverse events were evaluated. Survival outcomes were analyzed using Kaplan-Meier estimates and Cox proportional hazards models.Results: MET exon 14 skipping tumors had the highest prevalence of PD-L1 tumor proportion score >=50% (62.3%), whereas HER2-mutated tumors had the lowest (13.9%). With ICI monotherapy, MET tumors had a higher ORR than ALK-rearranged tumors. With ICI plus platinum chemotherapy, KRAS G12C tumors had a higher ORR than ALK-rearranged tumors. ICI plus platinum-doublet chemotherapy significantly prolonged PFS versus chemotherapy alone in BRAF V600E (12.7 vs 6.6 months; HR, 0.34; P=.02) and KRAS (8.8 vs 5.4 months; HR, 0.60; P=.001) cohorts. No new safety signals were observed.Conclusion: ICI efficacy differed by driver genotype. BRAF V600E- and KRAS-mutated tumors derived greater benefit from immunochemotherapy, whereas fusion-driven tumors showed limited responsiveness.