Background: Identification of molecular alterations has become a key component in the management of advanced non-small cell lung cancer (aNSCLC) patients, because these biomarkers can provide essential information to guide treatment selection. The National Comprehensive Cancer Network (NCCN) guideline recommends molecular diagnostic testing as part of the pathologic evaluation for non-squamous aNSCLC. However, in routine oncology practice, not all eligible patients are tested. Few studies have examined the impact of adherence to testing on patient outcomes in a real-world setting.
Methods: Patients diagnosed with non-squamous aNSCLC (stage IIIB and above) between 2011 and 2019 from the de-identified nationwide Flatiron Health electronic health record-derived database were included in this analysis. Testing status was defined as follows: tested patients had any of the NCCN recommended molecular test (EGFR, ALK, ROS1, KRAS, BRAF, or PDL1) any time between 14 days prior to and 90 days after aNSCLC diagnosis date; patients with no evidence of testing did not have documentation of receipt of any above-mentioned tests during this time period. Median survival was calculated using Kaplan-Meier analysis. Unadjusted and adjusted Cox proportional hazards regression models were used to evaluate the association between testing status and overall survival.
Results: A total of 29,658 eligible patients were included (75.3% tested; 24.7% no evidence of testing). Mean age at diagnosis was 69 years (SD: 15 years). The majority of patients (90%) were from the community clinics, and 31% of patients had commercial insurance plans. Compared to tested patients, patients with no evidence of testing were less likely to be enrolled in a commercial health plan, and more likely to have a history of smoking or be diagnosed before 2013 (p < 0.05). Tested patients had a lower risk of mortality in the unadjusted analysis (hazard ratio 0.84, 95% CI 0.82-0.87) and lived longer than patients with no evidence of testing (median survival 17.1 months [95% CI 16.8 -17.5] vs. 13.8 months [95% CI 13.2-14.4]), and the association remained significant after adjusting for potential confounders including age at diagnosis, sex, history of smoking, initial stage at diagnosis, race, and year of advanced diagnosis (hazard ratio 0.77, 95% CI 0.74-0.79).
Conclusions: Among non-squamous aNSCLC patients, following the NCCN biomarker testing guidelines may be associated with more favorable survival outcomes.