Strength of evidence underlying the CMS-FDA parallel review of comprehensive genomic profiling tests in the cancer setting

Abstract

Background Although use of comprehensive genomic profiling (CGP) was approved by a novel CMS/FDA parallel review process, the quality of the supporting evidence is unclear. We evaluated the rigor of the peer-reviewed literature cited in the National Coverage Determination Memorandum for the FoundationOne CDx (F1CDx). Methods We identified studies cited in the memorandum. Two independent researchers evaluated each study and applied a modified version of the Fryback and Thornbury hierarchy[1], an established framework for evaluating the efficacy of diagnostic tests. Studies focused on clinical outcomes were then categorized by study design, guided by recommendations from the Center for Medical Technology Policy. Results The sample included 113 scientific studies. The majority (n = 60, 53.1%) used CGP outside the course of clinical care, and there was significant heterogeneity in the cancer types assessed and sequencing depth. We found 8 (7.1%) studies that assessed whether clinical care had changed due to CGP testing, and 38 (33.6%) assessed clinical outcomes. After excluding studies that tested for five or fewer genomic alterations, 25 remained in the clinical outcomes sample: Of these, only one included a comparator group that did not receive CGP testing. Only four studies used F1CDx as the primary genomic test, none of which compared the outcomes of patients who did vs did not receive the F1CDx test. Conclusions The findings indicate gaps in the supporting evidence for broad CGP use in patients with solid tumors. More rigorous studies that assess clinical utility would better inform the approval process for novel diagnostic tests.