Heterogeneous efficacy and predictors of response to immunotherapy in driver-mutant advanced NSCLC: a focus on EGFR and KRAS subtypes.

BACKGROUND: Programmed death 1/programmed death-ligand 1 (PD-1/PD-L1) inhibitors improve survival in driver-negative advanced non-small cell lung cancer (NSCLC); however, their efficacy in oncogene-driven NSCLC has not yet been established. This study evaluated the clinical and molecular predictors of immunotherapy outcomes in this population. METHODS: We conducted a retrospective, two-center cohort study of 129 patients with advanced NSCLC harboring driver mutations who received PD-1/PD-L1 inhibitors (either as monotherapy or in combination with other antineoplastic drugs). Treatment efficacy and predictors were analyzed in the overall cohort and key mutational subgroups, including epidermal growth factor receptor (EGFR)- and Kirsten rat sarcoma (KRAS)-mutant NSCLC. We examined the efficacy of immunotherapy and assessed the correlation between treatment responses in 77 patients who received both immunotherapy and targeted therapy. RESULTS: The efficacy of immunotherapy varied significantly by genotype. Patients with KRAS mutations, a subgroup characterized by a higher prevalence of smoking history and elevated PD-L1 expression, demonstrated superior outcomes compared to those with EGFR, human epidermal growth factor receptor 2 (HER2), or other rare mutations. Multivariate analysis identified smoking history, high PD-L1 expression, and receipt of first-line immunotherapy as independent predictors of progression-free survival (PFS) in the overall cohort. Among the patients with EGFR mutations, receipt of first- or second-generation (1G/2G) tyrosine kinase inhibitors (TKIs) alone prior to immunotherapy, Threonine 790 Methionine (T790M) mutation-negative status, and the Leucine 858 Arginine (L858R) subtype were associated with improved outcomes. However, multivariate analysis showed that a history of treatment with 1G/2G TKIs only was the sole independent predictor of PFS. In KRAS-mutant patients, smoking history and high PD-L1 expression were positive predictors of PFS in the multivariate analysis. Smokers with a tumor proportion score (TPS) ≥1% achieved a median PFS of 19.0 months, compared with 8.5 months for non-smokers with a TPS ≥1% and 1.5 months for non-smokers with a TPS <1% (P<0.001). Targeted therapy was associated with a significantly longer median PFS than immunotherapy in both the overall cohort and the EGFR- and rare-mutation subgroups. No correlation was observed between responses to targeted therapy and subsequent immunotherapy. CONCLUSIONS: KRAS-mutant NSCLC is characterized by a high proportion of patients with a smoking history and elevated PD-L1 expression, which may contribute to its superior response to immunotherapy compared with other mutation subtypes. In the overall driver-positive population, smoking history, high PD-L1 expression, and receipt of first-line treatment were associated with immunotherapy benefits. Among the patients with EGFR-mutant NSCLC, prior treatment limited to 1G/2G TKIs was associated with better subsequent immunotherapy efficacy. Among the patients with KRAS-mutant NSCLC, smoking history and high PD-L1 expression were associated with improved immunotherapy outcomes. Overall, targeted therapy demonstrated superior efficacy than immunotherapy in this cohort. The efficacy of prior targeted therapy did not predict subsequent response to immunotherapy.