Clinical utility of companion diagnostic biomarker results below the limit of detection in comprehensive genomic profiling of patients with advanced non-small cell lung cancer

Background: When the limit of blank (LoB) of comprehensive genomic profiling (CGP) for a given biomarker is acceptably demonstrated (ie, alpha <= 0.05 or LoB equal to zero), biomarkers detected below the assay limit of detection (LoD) can be reported with a high degree of confidence. However, it is unknown whether variants detected below LoD have clinical utility. Patients and methods: This study used a de-identified nationwide (US-based) non-small cell lung cancer clinico-genomic database (CGDB) containing linked FDA-approved CGP testing from Foundation Medicine, Inc (FMI). We selected patients who received an FMI CGP report with an actionable biomarker detected below LoD. We assessed clinical utility among those patients who received an appropriately matched targeted therapy, defined as a real-world overall response rate exceeding a prespecified threshold of 30% based on historical chemotherapy response rates. Results: Among 129 patients who had a biomarker detected and reported below LoD, received the appropriate matched targeted therapy, and were assessed for response, partial or complete response was observed in 36/54 (67%, one-tailed 95% CI: >55%, P < .001) patients tested with a tissue-based CGP test and 54/75 (72%, one-tailed 95% CI: >62%, P < .001) patients tested with a liquid-based CGP test, both of which exceeded the prespecified threshold for clinical utility. Conclusions: Most patients who receive a targeted therapy matched to a companion diagnostic biomarker detected and reported below LoD demonstrate clinical benefit from that therapy. This clinical observation suggests actionable variants should continue to be reported when detected with FMI CGP tests.