Genomic profiling of lung adenocarcinoma patients reveals potential therapeutic targets when standard molecular testing is negative.

e20575 Background: Identification of oncogenic drivers is critical in selecting appropriate targeted therapy. Using next-generation sequencing (NGS), we performed comprehensive genomic profiling (CGP) of lung adenocarcinoma tumors. Methods: Formalin-fixed paraffin-embedded tumors from 51 lung adenocarcinoma patients whose tumors previously tested negative for EGFR/KRAS/ALK by conventional methods in an ongoing trial (NCT01964157) were collected, and CGP was performed via hybridization capture of 4,557 exons from 287 cancer-related genes and 47 introns from 19 genes frequently rearranged in cancer and next generation sequencing. Results: Genomic profiles of all 51 cases were obtained, with a median coverage of 564x and a total of 190 individual genomic alterations (GA). GA per specimen was a mean of 3.7 (range 0-10).Cancer genomes are characterized by 50% (80/190) non-synonymous base substitutions, 15% (29/190) insertions or deletion, and 3% (5/190) splice site mutation. TP53 was the most commonly mutated gene (30%, n = 24/80) among non-synonymous base substitutions, followed by KRAS (10%, n = 8/80) and EGFR (9%, 7/80). Insertions or deletions commonly occurred TP53 (17%, 5/29) and ERBB2 (14%, 4/29), and splice site mutations occurred in TP53, INPP4B, ATR, and MAP2K4. Gene amplification comprised 20% (39/190) of genomic alterations, and MDM2 amplification was the most frequent (13%, 5/39). Homozygous loss comprised 5% (10/190) of all GA which were CDKN2A/B. GA found with NCCN guidelines for NSCLC were identified in 16 patients (31%) (BRAF mutation [n = 1], EGFR mutation [n = 8], ERBB2 mutation [n = 4], MET amplification [n = 1], KIF5B-RET rearrangement [n = 2], CCDC6-RET rearrangement [n = 1], CD74-ROS1 rearrangement [n = 1], EZR-ROS1 rearrangement [n = 5], and SLC34A2-ROS1rearrangement [n = 1]). Other genes that conferred the possibility of benefit from targeted therapy were found in another 14 patients (27%). Conclusions: Fifty eight percent of patients wild type by standard testing for KRAS/EGFR/ALK have GA identifiable by CGP that suggest benefit from target therapy. CGP used when standard molecular testing for NSCLC is negative is able to reveal additional avenues of benefit from targeted therapy.