OBJECTIVES : This study aimed to determine how real-world ALK inhibitor (ALKi) treatment duration correlates with progression-free survival (PFS) and overall survival (OS).

METHODS : Retrospective analysis of the de-identified Flatiron Health EHR-derived database was used to measure ALKi time to treatment discontinuation (TTD) and outcomes of patients with ALK+ advanced NSCLC. TTD was plotted vs. PFS and OS and the correlation was estimated via Pearson’s correlation. Kaplan-Meier estimates were used to describe PFS and OS by treatment duration categories. We used a time-varying covariate in a Cox proportional hazards model to adjust for potential immortal time biases and estimate odds ratios and 95% confidence intervals to describe the association of treatment duration with PFS and OS.

RESULTS : 482 patients were included in this analysis. Longer ALKi treatment duration (<3, 4-6, 7-12, >12 months) corresponded with higher median PFS (2.5, 5.3, 8.6, 20.3) and higher median OS (7.4, 14.8, 21.6, median not reached) respectively. Each additional month of time on treatment up to 12 months was associated with a lower risk of progression or death within 12 months (HR [95% CI]: 0.87 [0.82 – 0.93]) and lower likelihood of death within 12 months (OR [95% CI]: 0.68 [0.62 – 0.75]). A strong correlation was observed between TTD and PFS (R= 0.72, p<0.001), while a moderate correlation was observed between TTD and OS (R = 0.63, p<0.001). 41.7% of patients had a progression or death event within 2 weeks of discontinuing treatment while 53.7% had a progression or death event within 4 weeks of discontinuing treatment.

CONCLUSIONS : ALKi treatment duration correlated highly with PFS and moderately with OS. This study suggests an indirect link between factors associated with treatment persistence and clinical outcomes. Enabling factors of treatment persistence, such as co-pay assistance, may potentially lead to improved clinical outcomes in patients with NSCLC.