High correspondence between EGFR mutations in tissue and in circulating DNA form non-small-cell lung cancer (NSCLC) patients (p) with poor performance status (PS)

7505 Background: We evaluated the correspondence between EGFR mutations in NSCLC tissue and matched serum DNA and the value of EGFR mutations as a serological marker. Methods: 121 Spanish stage IV NSCLC p received customized first- or second-line erlotinib monotherapy based on the presence of EGFR mutations in the tumor tissue. Serum genomic DNA was obtained from all p prior to erlotinib administration. EGFR exon 19 deletions were studied by length analysis of fluorescently labeled PCR products and the exon 21 L858R mutation by a PCR Taqman assay. Results: The EGFR mutation status in the serum was consistent with that in the tumor tissue of 82/121 p (68%) and of 15/16 p (93.8%) with PS 2 had mutations. Overall, 64.3% of p had an exon 19 deletion and 35.7% had L858R. 78% of mutations were found in females (P=0.01) and 77.6% in never-smokers (P=0.07). Response rate was 88% in p with mutations only in the tumor and 87% in p with mutations in tumor and serum. Complete responses were observed in 20% of p with mutations in tumor and serum vs 4% of p with mutations only in tumor (P=0.09). With a median follow-up of 6.8 months (m) (range, 1.2–17.6), time to progression (TTP) and median survival have not been reached. A trend to better outcome was seen in p without serum EGFR mutations. TTP was longer for p with EGFR exon 19 deletions (not reached) than for p with L858R (7.7 m) (P=0.02). TTP for p with PS 2 with exon 19 deletions was not reached, while it was 2.7 m for p with L858R (P=0.17). Conclusions: EGFR mutations in serum could be a non-invasive source of information on the genotype of the original tumor cells and could be a useful tool to predict p response to erlotinib, especially in p with poor PS. No significant financial relationships to disclose.