EFFICACY OF FT218, A ONCE-NIGHTLY SODIUM OXYBATE FORMULATION, IN PATIENTS WITH NARCOLEPSY: POST-HOC SENSITIVITY ANALYSES FROM THE REST-ON TRIAL

Document Type

Conference Proceeding

Publication Date

5-25-2022

Publication Title

Sleep

Keywords

cataplexy, narcolepsy, sodium oxybate, wakefulness, analysis of covariance, sensitivity analysis, surrogate endpoints, multiple wake test, imputation, multiple imputation, missing data, missing not at random, sleep latency

Abstract

Introduction: In REST-ON, once-nightly sodium oxybate (ON-SXB; FT218) achieved significant improvement (P<0.001) vs placebo for all coprimary endpoints: Maintenance of Wakefulness test (MWT) mean sleep latency, Clinical Global Impression of Improvement (CGI-I) rating, and weekly number of cataplexy attacks (NCA). Methods: Individuals aged ≥16 years were randomized 1:1 to receive ON-SXB (1 week, 4.5 g; 2 weeks, 6 g; 5 weeks, 7.5 g; 5 weeks, 9 g) or placebo. Post-hoc sensitivity analyses were conducted with methods to handle missing data: completer population; placebobased multiple imputation (MI) with missing not at random assumption; analysis of covariance (ANCOVA); and tipping pointbased MI of worsening values until P>0.05. Results: Completers (ON-SXB, n=69; placebo, n=79) showed significant improvement (P<0.001) with 6, 7.5, and 9 g ON-SXB vs placebo on all coprimary endpoints; with 9-g dose, mean (95% CI) differences vs placebo were 6.0 min (3.38.7) on MWT and 6.6 (9.6 to 3.6) in NCA; 72.3% and 31.6%, respectively (odds ratio [OR], 5.7 [95% CI: 2.811.6]), were CGI-I responders. All ON-SXB doses achieved significant improvement (P<0.001) vs placebo on all coprimary endpoints with placebo-based MI and ANCOVA. With placebo-based MI, mean (95% CI) differences vs placebo (9-g dose) were 5.4 min (2.88.0) on MWT and 6.4 (11.3 to 3.7) in NCA; 63.0% and 28.5%, respectively (OR, 4.3 [95% CI: 2.38.0]), were CGI-I responders. With ANCOVA, mean (95% CI) differences vs placebo (9-g dose) were 6.0 min (3.68.5) on MWT and 6.4 (9.0 to 3.8) in NCA; CGI-I rating difference was 1.0 (1.3 to 0.7). With MWT tipping point MI, between-treatment differences lost significance with worsening of 7.0, 5.2, and 4.3 min from baseline for 6, 7.5, and 9 g, respectively (implausible for 7.5- and 9-g doses). When withdrawals from ON-SXB were imputed as not improved, CGI-I remained significant (all 3 doses, P<0.001). Mean NCA remained significant for all 3 doses vs placebo with worsening trajectories imputed; positive results were not tipped over with plausible values. Conclusion: These results support the robustness of the primary efficacy data for ON-SXB for narcolepsy treatment.

Volume

45

Issue

SUPPL 1

First Page

A180

Find It @ Sladen

Share

COinS