OBJECTIVES: No head-to-head data are available for overall survival (OS) for brigatinib vs. alectinib in patients with ALK+ NSCLC previously untreated with an ALK inhibitor; data for both drugs were reported relative to a common comparator arm, crizotinib. However, estimates of indirect relative OS efficacy are biased by the treatment crossover specified in the ALTA-1L trial protocol (brigatinib vs. crizotinib) which was not a feature of the ALEX trial design (alectinib vs. crizotinib). The objective of this study was to estimate the relative OS for brigatinib vs. alectinib with adjustment for treatment crossover using the final results from the ALTA-1L and ALEX trials.

METHODS: The rank preserving structural failure time model (RPSFTM; with and without re-censoring) was applied to attempt to adjust for the bias introduced from 80 patients switching from the crizotinib arm to the brigatinib arm in ALTA-1L – including both official switchers and those identified through subsequent therapies. Simple Bucher methods, anchored matched adjusted indirect comparisons (MAICs) and unanchored MAICs were explored using these adjusted data to estimate the relative effect of brigatinib vs. alectinib for OS outcomes.

RESULTS: The hazard ratio for brigatinib vs. crizotinib using the unadjusted OS data from ALTA-1L was 0.815 (95% CI: 0.539 – 1.232); the hazard ratio improved to 0.673-0.717 when applying RPSFTM methods. The hazard ratio for alectinib vs. crizotinib from ALEX was 0.67 (95% CI: 0.46 – 0.98). Indirect treatment comparison methods estimated hazard ratios between 0.902 – 1.088 for brigatinib vs. alectinib following adjustment for treatment crossover in the ALTA-1L trial.

CONCLUSIONS: This study presents estimates of relative efficacy for brigatinib vs. alectinib following adjustment for treatment crossover in the ALTA-1L clinical trial. All methods indicate that OS is comparable between the two ALK inhibitors. A key limitation relating to this study is that other subsequent ALK inhibitor use was unaccounted for.