Genomic testing among patients (pts) with newly diagnosed advanced non-small cell lung cancer (aNSCLC) in the United States: A contemporary clinical practice patterns study.

Abstract

9592 Background: We describe contemporary clinical patterns of guideline-mandated genomic testing in newly diagnosed US pts with aNSCLC. Methods: From the Flatiron Health electronic health record-derived de-identified database, we included pts with newly diagnosed advanced non-squamous cell carcinoma of the lung between 1.1.2018–6.30.2019 who had received first-line (1L) therapy. We defined inadequate testing as no successful test for at least one of four examined genes: ALK, BRAF, EGFR, and ROS1. We grouped pts according to testing received into users of next-generation sequencing (NGS) testing, including a subgroup using comprehensive genomic profiling (CGP, exemplified by Foundation Medicine, Inc.), users of non-NGS testing, and no testing. We describe the following aspects of genomic testing before the start of 1L therapy: occurrence of testing, patterns of use of testing technologies, occurrence of inadequate testing, test failures, percentage of pts with potentially missed targeted therapy with US Food and Drug Administration approval (no positive test and <4 successful tests), and recent trends in genomic testing. Results: Among 2971 included pts, 690 (23.2%) had no genomic testing before 1L treatment. Among pts who had a test (n=2281), 59.4% (n=1355) received NGS (CGP: 18.8%, n=429), while 40.6% (n=926) received non-NGS tests only. In the CGP user group, 79.7% of pts received no other type of test, compared with 29.8% of pts in the other NGS group. Inadequate testing was recorded in 13.4% of NGS-tested pts (CGP: 4.9%), compared with 52.5% of pts tested by non-NGS only. Test failures contributed to unsuccessful testing in 4.2% of pts tested by NGS (CGP: 1.2%) and in 6.8% of pts who received non-NGS tests. In the NGS group, 10.1% (CGP: 3.0%) of patients potentially missed a targeted therapy option, compared with 40.3% in the non-NGS group. EGFR and ALK testing were performed in ≥95% of pts, regardless of the testing group; however, only 83.6% and 55.7% of pts received tests for ROS1 and BRAF, respectively, in the non-NGS group. In the latter group, for the first 6 months of 2019, 88.4% and 58.2% of pts were tested for ROS1 and BRAF mutations, respectively. Conclusions: Not performing any, or performing only inadequate genomic testing in pts with newly diagnosed aNSCLC remains a concern in clinical practice. The use of NGS, particularly CGP, may help to avoid suboptimal testing, minimize test failures, and improve uptake of testing for newly introduced biomarkers, enabling individualized, targeted therapy.