Pathology molecular reflex testing improves turn-around-time and overall molecular testing rates in NSCLC.

Abstract

e13507 Background: Dozens of genetic aberrations have become druggable targets or act as biomarkers of diseases for which pts may qualify for precision medicine treatment options. Biomarker informed treatments can lead to vast improvements in patient outcomes over traditional approaches like chemo. However, comprehensive molecular profiling (CMP) to identify eligible pts is not always accomplished, particularly in the community setting where molecular testing is often outsourced to third party laboratories. Here we report our experience with reflex testing protocols, where a pathology department has responsibility for initiating biomarker testing without the need for direct oncologist involvement. In this study we investigate and report how pathologist initiated reflex protocols influences access to molecular biomarker testing and assessed changes in testing rates, result tracking, and timeliness of results. Methods: We reviewed a reflex testing protocol at Hoag Hospital (Newport Beach, California), for NSCLC pts. A process map was generated to identify pts who qualified for reflex testing. After developing the process map, we used VIPER software that gathers data from electronic medical records and pathology systems to perform our analysis. We evaluated protocol adherence, failures, QNS rates, and TAT from test ordering to sample receipt and processing and ultimately all the way to results being received and compared patients that followed the reflex protocol as compared to those that did not to determine any differences. Data analysis was performed to understand how effective the reflex system was in ensuring the appropriate patients received CMP and any benefits in efficiency that resulted. Results: We evaluated 300 NSCLC pts over a 12-month timeframe for their reflex testing statuses. 78.5% of those pts tested received Next Generation Sequencing (NGS) tests, vs. 21.5% receiving non-NGS based tests. We also found over 75% of the pts that were qualified for reflex testing had complete tests, while 25% were not able to be completed. Of those who did not meet reflex criteria, only 28% went on to receive CMP. The primary reason for a patient who qualified for reflex testing but did not have testing initiated was due to insurance or reimbursement concerns (42%). The average time from a reflex test being ordered to results available was 25.9 days. Without reflex the TAT averaged 40.5 days (15.4 days difference). For reflexed cases, 78% of CMP results were available prior to treatment initiation. Conclusions: We demonstrate that implementation of a CMP reflex testing program is feasible and can ensure that a higher percentage of NSCLC pts receive molecular testing and that these results can be provided nearly two times sooner than without such programs. Our next steps are to automate this process to allow practices that do not have the personnel to manage such programs can still benefit via technology assistance.