Understanding Contemporary Molecular Biomarker Testing Rates and Trends for Metastatic NSCLC Among Community Oncologists

Timely testing for actionable biomarkers for patients with advanced or metastatic non-small cell lung cancer can ensure that they receive the appropriate treatment at the right time, but the real-world rate of testing has varied. This study assessed molecular testing patterns in a large network of US community-based oncology practices. Although testing rates increased over time and testing turnaround times decreased, results of testing were not available for a considerable number of patients before initiation of first-line treatment. Introduction: Although guidelines recommend testing for actionable biomarkers for patients with advanced or metastatic non-small cell lung cancer (NSCLC), testing rates have varied. This study aimed to assess molecular testing patterns in a large network of US community-based oncology practices. Methods: This retrospective observational study examined adult patients with newly diagnosed stage IV NSCLC with >= 2 visits in The US Oncology Network from July 1, 2016 to September 30, 2019. Testing patterns were examined using electronic health record structured fields and chart review. Structured data were analyzed for the overall study population (cohort A), and structured and unstructured data were analyzed for a select cohor t of 300 patients (cohor t B). Results: In cohor t A (n = 3337), programmed death ligand 1 (37%) was the most frequently tested biomarker documented in structured data, followed by epidermal growth factor receptor (36%), anaplastic lymphoma kinase (35%), ROS1 (20%), and BRAF (16%). According to unstructured data in cohort B (n = 300), epidermal growth factor receptor (80%) was the most frequently tested biomarker, followed by anaplastic lymphoma kinase (79%), programmed death ligand 1 (72%), ROS1 (71%), and BRAF (56%). The proportion of tests ordered prior to first-line (1L) treatment increased from 2016 to 2018 for all biomarkers, as did the proportion of test results available prior to 1L treatment. However, some of the test results became available after 1L or later lines of treatment were in progress. Conclusion: Our study found increased testing rates over time and decreases in testing turnaround times. However, rates of testing for all biomarkers still need to improve, as does completion of testing prior