Value of Precision Medicine in Advanced Non-Small Cell Lung Cancer: Real-World Outcomes Associated with the Use of Companion Diagnostics

Abstract

e20015Background: Five-year survival rates as low as 2.8% have been reported in patients with non-small cell lung cancer (NSCLC), highlighting the need for individualized diagnosis and treatment. Companion diagnostic testing (CDx) identifies patients with molecular targets likely to respond better to particular therapies; however, not all cancer patients receive CDx in the real-world setting. This study evaluated the clinical value of CDx in the real world with respect to overall survival among patients with non-squamous advanced (Stage IIIB/IV) NSCLC (aNSCLC). Methods: Patients were from the Flatiron Health electronic health-derived database, treated with systemic therapy, and diagnosed with aNSCLC between January 1, 2011 and May 31, 2018; those who received CDx with their first line of treatment were compared with those who did not. Logistic regression using components of the modified Lung Cancer Prognostic Index (LCPI; age, sex, stage, actionable mutation(s), smoking, respiratory comorbidity; Alexander et al. Br J Cancer. 2017) and other factors were used to predict characteristics associated with receiving CDx. Overall survival was evaluated using Kaplan-Meier analysis. Unadjusted and adjusted Cox proportional hazards regression models were used to evaluate the association between CDx and overall survival. Results: A total of 17,143 patients with aNSCLC (CDx, n = 14,389; no CDx, n = 2754) and a mean (SD) age at diagnosis of 67.2 (10.0) years (CDx, 67.1 [10.1]; no CDx, 67.5 [9.2]) were included. There were more nonsmokers in the CDx group (17.4%) than the no CDx group (5.5%). Patients who were female, diagnosed after 2014, receiving multiple lines of therapy or had advanced stage at diagnosis were more likely to receive CDx. Patients receiving CDx had decreased mortality risk (unadjusted HR [95% CI] = 0.54 [0.52-0.57]) and lived longer than those not receiving CDx (median survival = 14 vs 7 months). The significant reduction in mortality associated with CDx remained after adjusting for factors included in the modified LCPI (adjusted HR [95% CI] = 0.78 [0.75-0.82]) as well as a model without actionable mutations (adjusted HR [95% CI] = 0.70 [0.66-0.73]). Conclusions: Among patients with non-squamous aNSCLC, use of CDx was associated with reduced risk of mortality compared with no CDx.