Association of Next Generation Sequencing and Routine Testing with Survival of Advanced Non-Small Cell Lung Cancer Patients in Clinical Practice

Background: The rapid development of detection technologies has resulted in an increased use of next generation sequencing (NGS) in advanced non-small cell lung cancer (NSCLC) patients. Here, we investigated the clinical benefits of different detection technologies for patients with advanced lung cancer in a real world setting. Methods: : This retrospective study enrolled advanced NSCLC patients diagnosed between January 1, 2013 and June 30, 2019 in Zhejiang Cancer Hospital who received genetic testing. Patients received either NGS or routine testing (EGFR and/or ALK and/or ROS1 only). Although different panels were used for NGS, they all included nine genes mentioned as molecular biomarkers in the NCCN guidelines (EGFR, ALK, ROS1, KRAS, HER2, BRAF, MET, RET, and NTRK). Clinical outcomes were compared between advanced NSCLC patients who received NGS and those undergoing routine testing. Overall survival (OS) and survival rates were analyzed. Propensity score analysis was used to prevent survival bias associated with patient characteristics. Results: : Of 6,451 patients with advanced NSCLC included in the analysis, 5,666 (87.8%) patients underwent routine testing and 785 (12.2%) were tested by NGS. The median OS was 22.6 months and the 5-year survival rate was 22.89%. Comparison of OS showed an absolute difference between the two cohorts according to unadjusted survival curves (24.8 months vs. 22.1 months, P=0.0097), whereas no difference in OS was observed using propensity score analysis (P=0.533). The 1-year survival rate was significantly higher in patients tested by NGS than in those undergoing routine testing (82.0% vs. 70.7%, P=0.0004). In patients with uncommon mutations and in those receiving targeted therapies, propensity score analysis showed that the 1- and 2-year mortalities were significantly lower in the NGS group than in the routine testing group without mutations. This could be due to the higher rate of patients receiving targeted therapy (68.0% vs. 53.6%; P=0.042) and off-label drugs (28.10% vs. 11.40%; P=0.002) in the NGS testing cohort. Conclusions: : Although propensity score analysis showed NGS had no significant benefit with RT, NGS detected more uncommon mutations and improved off-label drugs treatments. The findings highlight the importance of NGS detection, lead to a clinical benefit in uncommon mutations.