Table 2.
Results from hierarchical testing of coprimary and secondary endpoints
| Populationa | Endpoint analysisb | PBO, n | PEPI, n | LS mean diff. (95% CI) PEPI versus PBO | One-sided P value | Favors PEPIc | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Coprimary efficacy endpoints | ||||||||||||
| Cohort B1 mITT | Two-item cognitive family | OTS (s)d | 88 | 90 | −1.98 (−4.00, 0.05) | 0.028 | (+) | |||||
| PTAP (1 s–1)e | 87 | 89 | 1.43 (−0.37, 3.23) | 0.060 | (+) | |||||||
| Cohort B1 CGIC | CGICf | 83 | 84 | 0.06 (−0.24, 0.37) | 0.35 | (+) | ||||||
| Secondary efficacy endpoints | ||||||||||||
| Cohort B1 mITT | Q-motor tap speed IOI duration (ms)g | 87 | 89 | −1.44 (−24.26, 21.38) | 0.46 | (+) | ||||||
| UHDRS–TFCh | 88 | 88 | −0.29 (−0.81, 0.23) | 0.87 | (−) | |||||||
aThe cohort B1 mITT population comprised 178 participants (PBO, 88; PEPI, 90) while the CGIC population comprised 169 participants (PBO, 83; PEPI, 86). The number of subjects included in an analysis may differ from the full analysis population (for example, mITT or CGIC) size if any subjects do not have both baseline (if applicable) and post-baseline observation.
bAnalysis results are from an MMRM of change from baseline values unless otherwise specified, with estimation of the difference in means between groups at month 17.
cThe sign next to the P value indicates whether the direction of the estimated difference indicates better (+) or worse (–) performance for PEPI relative to PBO.
dTime to a correct response (averaged over all trials per visit); lower values indicate better performance.
eTapping consistency measured as the reciprocal of the average standard deviation of inter-tap interval (ITI) durations (over all trials per visit); higher scores indicate better performance.
fA seven-point Likert scale, ranging from very much worse (−3) to very much improved (+3); values are set to −3 following deaths adjudged by a blinded data review committee to be related to HD.
gIOI duration in the more-affected hand at baseline (averaged over all trials per visit); lower values indicate better performance. The more affected hand at baseline is defined for each outcome as the side with the worse score at baseline; if both hands have the same baseline score, the more affected hand will be the self-reported dominant hand; if the dominant hand is mixed, the right hand will be reported. If a subject has a baseline score for only one hand, that hand will be the more affected hand.
hScore ranging from 0 to 13; higher scores indicate better functioning.