Molecular reflex testing in non-small cell lung cancer: An optimal approach?

Abstract

3127 Background: Molecular testing of non-squamous non-small cell lung cancer (NSCLC) tumors can guide appropriate treatment decisions and improve patient outcomes, but guideline complexity and frequent revision may negatively affect adherence. To assist oncologists in making timelier informed treatment decisions, we implemented a pathologist-directed molecular reflex pathway for non-squamous NSCLC at our clinical laboratory. Testing patterns and adherence to NCCN testing guidelines before and after implementation were reviewed. Methods: This retrospective cohort analysis included patients with diagnosed NSCLC who had molecular testing performed without a molecular reflex testing pathway in place (April 2016-March 2018, cohort A) and after the pathway was implemented (April 2018-March 2020, cohort B) at our clinical laboratory. TATs were calculated using dates of biopsy specimen submission and of molecular testing completion. Molecular testing methods (e.g., polymerase chain reaction, next-generation sequencing [NGS]), genetic alterations tested for and identified, PD-L1 by immunohistochemistry, and specimen quantity not sufficient (QNS) for testing were obtained from patients’ clinical laboratory reports. NSCLC diagnosis and cancer stage were obtained from electronic medical records, and adherence to NCCN guidelines (i.e., under-tested, over-tested genetic alterations) was evaluated according to specimen submission. Results: The mean TAT was 35.1 days for cohort A (n = 123) and 15.8 days for cohort B (n = 168), though the median in both was 14 with a range of 6 – 674 days in cohort A and 9 – 44 in cohort B (Table).Targetable genetic alterations were identified in 12.4% of patients in cohort A and 61.3% in cohort B; 46.3% in cohort A and 60.1% in cohort B had ≥1% positive PD-L1 staining. QNS results declined from 16.7% (n = 20) in cohort A to 12.5% (n = 21) in cohort B. Non-NCCN guideline testing was observed in 86.7% of patients in cohort A and 25.6% in cohort B, with those in cohort A being primarily under-tested (70%) and those in cohort B being over-tested. Only 13.3% of cohort A had alignment with NCCN guidelines for appropriate time of testing. Conclusions: Implementation of a pathology-driven molecular reflex pathway for non-squamous NSCLC was associated with increased identification of potentially targetable genetic alterations and improved adherence to NCCN NSCLC testing guidelines, along with reduction in QNS results. Though the TAT mean decreased after pathway implementation, the median did not change, offsetting the often increased lab TAT for NGS methodology vs. PCR and FISH alone. [Table: see text]