Abstract
BACKGROUND: Although optimal sequencing of systemic therapy in cancer care is critical to achieving maximal clinical benefit, there is a lack of analysis of treatment sequencing in advanced non-small cell lung cancer (aNSCLC) in real-world settings. METHODS: A retrospective cohort study of 13,340 lung cancer patients within the Mount Sinai Health System (MSHS) was performed. Systemic therapy data of aNSCLC in 2,106 patients was the starting point in our analysis to investigate how treatment sequencing has evolved, the impact of sequencing patterns on clinical outcomes, and the effectiveness of 2(nd) line chemotherapy after patients progressed on immune checkpoint inhibitor (ICI)-based therapy as the 1(st) line of therapy (LOT). RESULTS: There is a significant shift to more ICI-based therapy and multiple lines of targeted therapy after 2015. We compared clinical outcomes of two patient populations with different treatment sequencing patterns, with the 1(st) group receiving chemotherapy as the 1(st) LOT followed by ICI-based treatment, and the 2(nd) group treated in the opposite order receiving a 1(st) line ICI-containing regimen followed by a 2(nd) line chemotherapy. No statistically significant difference in overall survival (OS) was observed between the two groups [group 2 vs. group 1, adjusted hazard ratio (aHR) =1.36, P=0.39]. We assessed the efficacy of the 2(nd) line chemotherapy in three patient populations given either 1(st) line ICI single agent, 1(st) line ICI-chemotherapy combination, or 1(st) line chemotherapy alone, there was no statistically significant difference in time-to-next treatment (TTNT) and in OS among the three patient groups. CONCLUSIONS: Analysis of real-world data has shown two treatment sequencing patterns in aNSCLC, ICI followed by chemotherapy or chemotherapy followed by ICI, achieved similar clinical benefit. The chemotherapies routinely used following platinum doublet 1(st) LOT, is effective as the 2(nd) line option after ICI-chemotherapy combination in the 1(st) line setting.