1231P - Efficacy of first-generation EGFR-TKIs on patients with NSCLC harboring EGFR uncommon mutations: a pooled analysis

Abstract

NSCLC Patients with common sensitive EGFR mutations (deletion in exon 19 or L858R mutation in exon 21) benefit remarkably from first-generation EGFR-TKIs (gefitinib and erlotinib). In this meta-analysis, we aim to investigate their treatment efficacy in patients with uncommon EGFR mutations (S768I, L861Q, G719X, R705K, etc.) by comparing with that in patients with common mutations. We searched PubMed for eligible studies from the date of inception to 31th December, 2015. Overall objective response rate (ORR) and 6-month progression free survival (PFS) rates were estimated by fix-effect model and relative effects were presented using odds ratio (OR). Mutations within the same exons were grouped in the subgroup analyses. Of 6404 patients from 13 included studies, 466 (7.3%) patients were diagnosed as EGFR uncommon mutations and chose gefitinib and erlotinib as any line of treatment. In single-arm synthesis, the overall ORR in uncommon and common mutations was 34.0% (95% CI 22.5 to 45.5) and 71% (66.7 to 75.3), respectively. Direct comparison indicated significantly less response in uncommon-mutation patients (OR = 0.30, 95% CI 0.23 to 0.41, P < 0.001). Patients with uncommon mutations, compared with common ones, were associated with an inferior 6-month PFS rate (OR = 0.44, 95% CI 0.32 to 0.59; P < 0.001). In exploratory analyses, ORR was 30.7% (16.3 to 45.0), 33.3% (N.A.), 32.1% (13.2 to 51.0) and 38.5% (19.5 to 57.6) in patients with rare mutations in EGFR exon 18, 19, 20 and 21 exons, respectively. In patients with complex mutations (two or more uncommon mutant sites), the ORR was 64.2 % (49.9 to 78.5). Compared with those in sensitive mutations, first-generation EGFR-TKIs presented less clinical benefits in uncommon mutations; but the responses are still considerable, historically compared with that of chemotherapy, especially in complex mutations. First-generation EGFR-TKIs remained an option for uncommon mutations but decision-making should be cautious. Exact efficacy in each specific mutation site merits future studies with larger sample size.