Abstract 2243: Comprehensive genomic profiling - does timing matter

Abstract

Abstract Background: Comprehensive Genomic Profiling (CGP) is performed routinely in patients (pts) with metastatic solid tumors (MSTs) to guide treatment. However, there is variability in the interval between metastatic diagnosis (MDx) and performance of CGP based on histology and/or provider preference. We analyzed the CGP utilization patterns and its impact on outcomes at our academic medical center. Methods: All MST pts with CGP data (January 2012 - April 2020), their date of MDx, and the date of CGP were identified by electronic medical record review. Pts who had CGP after MDx were divided into tertiles (T1-earliest, T3-latest) within each cancer type, while some pts had CGP performed prior to MDx (pre-mets). Overall survival (OS) was estimated from the time of MDx with left truncation at the CGP time. Cox regression model was used to estimate the impact of ‘timing of CGP' on survival for individual cancer (CA) types. Results: Among 1,358 pts identified, 710 (52%) were female, 1,109 (82%) Caucasian, 186 (14%) Afro-Americans, and 1,320 (97%) non-Hispanic. Lung (254; 19%), colorectal (203; 15%), gynecologic (121; 8.9%), pancreatic (106; 7.8%) and connective tissue/soft tissue CAs (95; 7%) were the most frequently identified. Sex, race, and ethnicity had no impact on the time interval between MDx and CGP with 2 exceptions - Hispanics with lung CA and females with pancreas CA had delayed CGP compared to non-Hispanics and males respectively (p =0.019, p =0.025). The median time to CGP after MDx was shortest for urothelial CA (2.4 months) and longest for prostate CA (22 months). Pts with lung [T2 Hazard ratio (HR) of 2.25, p<0.001; T3 HR of 2.67, p<0.001], gastro-esophageal (T2 HR of 1.88, p=0.079; T3 HR of 2.49, p= 0.024), and gynecologic CAs (T2 HR of 2.58, p=0.012; T3 HR of 5.19, p=0.003) had better survival if they had CGP performed during the first tertile after MDx. Conclusion: Early CGP after MDx may have a greater impact on OS in CA types with more actionable therapeutic targets. The inherent bias associated with limiting the analyses only to pts who had CGP performed hampers our ability to contrast these results with those who did not have CGP performed. The establishment of national quality metrics for CGP utilization by CA type is imperative. Citation Format: Bicky Thapa, Gulrayz Ahmed, Matthew Lasowski, James P. Thomas, Honey V. Reddi, Michael T. Zimmerman, Raul Urrutia, Aniko Szabo, Ben George. Comprehensive genomic profiling - does timing matter [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 2243.