We assessed the integrated evidence of efficacy by performing global statistical tests (GSTs) for the primary and secondary endpoints of MOXIe Part 2 (NCT02255435), a 48-week study of the safety and efficacy of omaveloxolone (omav) in patients with Friedreich's ataxia (FA).

MOXIe Part 2 showed that omav significantly improved the primary endpoint, modified FA Rating Scale (mFARS) scores, by -2.40 points relative to placebo (p=0.014; n=82). Key secondary endpoints, Patient and Clinician Global Impression of Change (PGIC and CGIC, respectively) demonstrated non-significant but numerically improved results in patients randomized to omav relative to placebo (-0.1 and -0.07, respectively). Omav also improved the secondary endpoint score of FA-Activities of Daily Living (FA-ADL) and achieved nominal statistical significance relative to placebo (?1.3; p=0.042). A GST is an objective way of combining evidence mathematically across multiple endpoints, accounting for redundancy.

Two different post-hoc GSTs were performed for MOXIe Part 2. For GST1, the SAP-specified hierarchy was used as the basis for combining the primary and key secondary endpoints (mFARS and PGIC/ CGIC). For GST2, the two most clinically important endpoints were combined (mFARS and FA-ADL).

A nonparametric, rank-sum test was used for the GSTs. GST1, which combined mFARS, PGIC, and CGIC, resulted in a p-value of 0.0124. GST2, which combined the two most clinically important endpoints, mFARS and FA-ADL, resulted in a p-value of 0.0041. Analysis of mFARS alone using the rank-sum test resulted in a p-value of 0.0280.

The GSTs integrated evidence from the primary and secondary endpoints and demonstrated stronger evidence of efficacy compared to the primary endpoint alone. This strengthening of evidence occurs when endpoints are directionally consistent with sufficient independence from the primary endpoint, suggesting that secondary endpoints confirm that omav improved the key clinical aspects of FA that were studied.