Comprehensive genomic profile by Foundation Medicine test in guiding routine decisions for second-line treatment in advanced non-small cell breast cancer (NSCLC): Preliminary results of lung-ONE study

e21555 Background: Lung-ONE study aims to determine the clinical utility in making decisions on planned 2nd line therapy in advanced NSCLC based on a comprehensive genomic profile (CGP) test as FoundationOne CDx or FoundationOne Liquid and describe alterations found on these patients. Methods: Lung-ONE included advanced/metastatic NSCLC patients being treated on first-line, with molecular diagnostic wild-type (or unknown) for at least ALK, EGFR, and ROS-1 genes. Patients treated with targeted therapies or immunotherapy were excluded. This interim analysis describes the findings of CGP from patients of Lung-ONE with driver genomic alterations (GAs) and the investigators decision to change the planned second line treatment or not. Patients continue being followed-up for up to 2 years. Results: As of November 15 2019, 152 patients were included. At the time of sampling, 90.7% had received ≥2 different chemotherapy regimens and other therapies. Main testing for ALK and ROS-1 was IHC in 91.4% and 73.8%, respectively; and for EGFR, it was qPCR in 99.3%. Planned 2nd-line based on routine molecular profile was chemotherapy and, mainly checkpoint inhibitors (61.7%). CGP was assayed on 102 (67.1%) tissue specimens and 50 (32.9%) blood samples. Of the 1,581 GAs identified in 141 patients, 541(34.2%) were potential drivers in 131. The most frequently mutated genes were TP53 in 72(55%) patients and KRAS in 52(40%). TMB and MSI signatures were calculated on 82 patients. 21% had < = 16 mut/Mb. Other GAs of interest are presented in table. Therapeutic orientations were recommended for 116/152(76.8%) patients. In 39/151(25.8%) patients the GAs had approved drugs in NSCLC or other tumor types. CGP outcomes, would be used by 27% of oncologists to guide targeted 2nd-line. Conclusions: CGP detected mutations that are informative for choosing personalized treatment in 25%. CGP demonstrates the ability of detect druggable genomic alterations missed at diagnosis opening new treatment options. [Table: see text]