P1.05 Upfront Genomic Testing for Patients With Non-Small Cell Lung Cancer Receiving First-Line Platinum-Based Regimen: Preliminary Result of the Msn Study

ABSTRACT Background Recent advances in lung cancer have identified potential driver mutations that may be targeted. To identify new predictors of response as well as novel targets for therapy, we have initiated a comprehensive large-scale re-sequencing analysis of genes potentially mutated in NSCLC. Methods Genomic DNA was extracted prospectively from untreated advanced NSCLCs. All tumors were obtained on IRB-approved protocols and after patientsu0027 consent (MSN trial ‘Melanoma - Small-cell lung cancer - Non-small cell lung cancer ‘). Pathology specimens were macrodissected in order to obtain more than 30% of tumor cells and, after DNA extraction, 96 selected exons from 30 genes (AKT1-3, ALK, BRAF, EGFR, ERBB2,4, FGFR2-4, GNAS, HRAS, KIT, KRAS, MAP2K1-2, MET, NOTCH1, PDGFRA, PDPK1, PI3KCA, PTEN, RET, STK11, TP53, and VHL mutations) were analyzed by Sanger sequencing. ALK rearrangements and FGFR1 amplification were detected by fluorescence in-situ hybridization. All results were discussed monthly in a molecular thoracic multidisciplinary staff. Second or third lines of treatment were adapted to mutation profiling. Results Thus far (between 09/01/2010 and 06/01/2012), 280 patients have been included and the whole genomic analysis was done in 100 tumors. The median age was 60 years (range 33-79), 36 (36%) were female, 70 (70%) had adenocarcinoma, 85 (85%) were former/current smokers. Five patients had incomplete genomic analysis due mostly to insufficient tumor cells in the specimen or poor quality DNA. Median tumor cells ratio was 50%. Mutations of an actionable driver mutations were identified in 50% of patients (65% in adenocarcinoma and 19% in squamous cell) (KRAS: 24%; EGFR: 11%; STK 11: 7%; BRAF: 2%; PDK1: 2%; MET: 1%; ERBB2: 1%; PIK3CA: 1%; ALK translocation: 2%, FGFR1 amplification: 1%) of whom 11 had concurrent mutations. EGFR and KRAS mutations were mutually exclusive. No mutations were identified among 5 never smokers analyzed. The median time to complete testing for this initial phase was 30 days. Forty percent of patients with genomic alterations were treated with molecularly targeted therapy based on their genetic alteration. Conclusions Mutational profiling of NSCLC is feasible, can distinguish relevant molecular subsets of lung cancer, and may present an impact on treatment at our cancer institute.