P3-077: EGFR mutations and response to tyrosine kinase inhibitors therapy in NSCLC patients previously treated with chemotherapy

The aim of this work was to study the association of epidermal growth factor receptor (EGFR) mutations and response to conventional cytotoxic chemotherapy agents and tyrosine kinase inhibitors (TKIs) in non-small-cell lung cancer patients of Caucasian origin since conclusive data have not yet been published. Patients with stage I to IV NSCLC that were treated with gefitinib or erlotinib after failure of 1st and 2st-line platinum-based chemotherapy were subjected to this analysis. Genomic DNA was isolated from paraffin-embedded tumor specimens, amplified for EGFR (exons 18, 19, 20 and 21) by touchdown hemi-nested polymerase chain reaction and sequenced in both sense and antisense directions. RECIST criteria were used to assess response to chemotherapy and TKIs. This mono-istitutional retrospective analysis was conducted on 103 pts from July 2002 to November 2006. Median age was 60 yrs (range 25-81.5), M/F: 59/44, ECOG-PS 0/1/2/3:68/30/4/1; stage: I/II/III/IV 4/5/35/59; adeno/bronchioloalveolar/squamous/large-cells/other:66/9/20/3/5;never/former/current smokers: 26/20/36. EGFR mutations were detected in 19 pts (18.5%); 13 pts (13.8%) had deletional mutations in exon 19, 3 pts (2.9%) had point mutations in exon 20 (one patient had two mutations) and 3 pts (2.9%) had missense mutations in exon 21. No associations were detected between EGFR mutations and adeno/bac histotype (p=0.51), smoke (p=0.23) and gender (p=0.14). Disease control rate to TKIs therapy was 54.4%, including 1 complete (pt. with EGFR mutations), 21 partial response and 34 stable disease and objective response rate was 42%. Median time to progression was 4.4 months and median survival was 23.7 months. Patients with EGFR mutations had a significant impact on response rate (PR/CR, 42% p=0.015) to TKIs. Instead EGFR mutations did not influence the response to chemotherapy. Cox’s multivariate analysis including gender, smoke, stage and histotype showed that EGFR mutations did not reach statistical significance for progression free survival (PFS) (p=0.52) and overall survival (OS) (p=0.50); only adeno/bac histotype was a prognostic factor for longer PFS (p=0.01) and OS (p=0.04). In our experience EGFR mutations seem to influence response rate in pts treated with TKIs after platinum-based chemotherapy. These data will be more thoroughly defined by fluorescence in situ hybridization analysis of EGFR gene copy number, measurement of EGFR and phosphorylated Akt (p-Akt) protein expression by immunohistochemistry and KRAS mutations.