PD‐L1 expression is associated with inferior outcomes independent of EGFR subtype in advanced non–small cell lung cancer with uncommon EGFR mutations treated with first‐line second‐generation EGFR‐tyrosine kinase inhibitors

BACKGROUND: Second-generation epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKIs) are key first-line therapies for advanced non-small cell lung cancer (NSCLC) with uncommon EGFR mutations, with risk stratification mainly based on EGFR mutation subtypes. Emerging evidence suggests programmed death-ligand 1 (PD-L1) expression may affect EGFR-TKI efficacy. This study compared associations of PD-L1 expression and EGFR mutation subtypes with outcomes in patients receiving first-line second-generation EGFR-TKI monotherapy. METHODS: This dual-center retrospective real-world study included 215 patients with advanced NSCLC harboring uncommon EGFR mutations. Efficacy was evaluated across mutation subtypes (L861Q, G719X, S768I, other single-site, compound mutations). Stratified analyses by PD-L1 expression were performed, with prognostic value explored in primary lung lesions and metastatic lymph nodes. Primary end points include progression-free survival (PFS), objective response rate (ORR), and disease control rate (DCR). RESULTS: The median PFS for all patients was 14.07 months (ORR, 49.3%; DCR, 89.8%). No statistically significant differences in ORR, DCR, or median PFS were observed among uncommon EGFR mutation subtypes. In contrast, high PD-L1 expression was significantly associated with shorter PFS. Furthermore, PD-L1 expression (tumor proportion score [TPS] ≥1%) in primary lung lesions was associated with shorter PFS, whereas no similar association was observed in metastatic lymph nodes. CONCLUSION: In advanced NSCLC with uncommon EGFR mutations, first-line second-generation EGFR-TKI monotherapy shows no statistically significant efficacy differences across mutation subtypes. PD-L1 expression is inversely associated with outcomes, with prognostic value differing between primary tumors and metastatic lymph nodes, highlighting the importance of tissue sampling site in risk stratification.