LAUSR

Modern cancer care has seen an increase in the use of targeted therapies and the number of FDA-approved biomarkers. To take advantage of these options, many have turned to commercially available comprehensive genome profiling (CGP) services, like Foundation Medicine9s FoundationOne assay. FoundationOne (FO) offers a list of clinically relevant genomic alterations and biomarkers, as well as a list of clinical trials, on-label and off-label therapies that may benefit the patient, that is, therapeutic associations. We performed a retrospective cohort study to assess the clinical benefit of therapy associations from FO reports. We collected demographics, therapy records, and clinical outcomes for 1004 patients who had a history of advanced cancer in the Cleveland Clinic system and had received a FO report between 2012 and 2017. FO therapy associations were considered as “applied” if a patient received an order for said therapy for the first time after their report date. We classified patients as follows: No Associated Therapy (NAT) if the patient did not receive any recommendations, Therapy Applied (TA) if at least one associated therapy was ordered after report date, and Therapy Not Applied (TNA) if none of the associated therapies were applied. We evaluated differences in demographic and clinical features among the 3 groups using Chi-squared and Kruskal-Wallis tests where appropriate. We performed survival analysis using univariate and multivariate COX Proportional Hazards regression models, and Log-Rank tests on Kaplan Meier Curves with overall survival (OS) as our metric. Our cohort demographics were 55% male, 85% white, 92% non-Hispanic, and a median age at report of 60 (IQR: 51-69). The most common diagnoses were Lung Adenocarcinoma (14%), Glioblastoma (8%), Colon Adenocarcinoma (8%), and Breast Cancer NOS (4%). Most of our patients belonged to the TNA class (64%), 21% were NAT, and 15% were TA. There was no statistical significance in the demographic distribution among the therapy classes. In the pan-cancer analysis, we found no statistically significant difference in OS among the therapy groups or any of the adjusted covariates except for metastasis status. Similarly, when evaluating the top cancer diagnoses individually, we found no significant differences in OS. The data in our study indicates that the application of FO therapy associations is not correlated with a statistically significant difference in OS for advanced cancer patients. This suggests that larger studies should be performed to better understand how CGP services provide clinical benefits to patients, and how we can maximize these benefits in the real-world community setting. Citation Format: Jean Rene Clemenceau, Sung Hak Lee, Alex Milinovich, Jian Jin, Nathan Pennell, Davendra Sohal, Tae Hyun Hwang. Analysis of the clinical benefit of comprehensive genome profiling derived therapeutic associations in advanced cancer [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 3638.