A Study to Learn About the Effects and Safety of RTA 408 (Omaveloxolone) in People Aged 16 to 40 With Friedreich's Ataxia: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Learn About the Effects and Safety of RTA 408 (Omaveloxolone) in People Aged 16 to 40 With Friedreich's Ataxia (NCT02255435), a randomized phase 2 study. The primary outcome is Part 1: Change From Baseline in Peak Work During Maximal Exercise Testing at Week 12.

2. Clinical Question

3. Trial Design

MOXIe was a phase 2 randomized, double-masked, parallel-group study of omaveloxolone in Friedreich’s ataxia. The registry lists 172 enrolled participants and a completed study. The recorded start was 2015-01-31, and primary completion was 2019-10-31. The sponsor was Biogen.

The study had two main parts and an optional extension. Part 1 tested multiple doses over 12 weeks, including a group that received 2.5 mg followed by 5 mg and groups receiving fixed daily doses. The recorded Part 1 dose groups included 10 mg, 20 mg, 40 mg, 80 mg, 160 mg, and 300 mg, along with placebo. The primary outcome was change in peak work from baseline at Week 12. The summary says the study would use Part 1 dose comparisons to choose a dose for Part 2; the specific dose-selection procedure is not reported in the registry record.

Part 2 compared daily omaveloxolone 150 mg with placebo for 48 weeks. Several outcomes were assessed, including movement and function measures, global impressions, falls, peak work, and daily-living activities. Eligible participants could enter an optional extension, where those who had received placebo could switch to omaveloxolone and eligible participants who had received omaveloxolone could continue it.

Part 1 and Part 2 answer related but distinct questions. The primary endpoint concerns the short-term, multiple-dose Part 1 comparison at Week 12. The Part 2 findings concern a later 48-week comparison at a selected dose. Extension observations are also distinct because treatment assignment and exposure differ from the blinded placebo comparison. Enrollment across the full study should therefore not be confused with the smaller numbers analyzed for individual endpoints or groups.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: A total of 172 participants diagnosed with Friedreich's Ataxia (FA) were enrolled in this study. The study consists of 3 parts: Part 1, Part 2, and the Open-Label Extension (OLE) period. Out of the 68 participants who completed Part 1 and 96 participants who completed Part 2, 57 participants from Part 1 and 92 participants from Part 2 entered the OLE period, and 105 participants completed the OLE period.

Only the first 6 of 12 groups are shown.

Part 1

Reasons for not completing this period:

Part 2

Reasons for not completing this period:

OLE Period

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: ITT population included all enrolled participants in Part 1, whether or not they received the study drug. Here, 'overall number of participants analyzed' signifies the number of participants with data available for outcome measure analysis.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: Part 1: The Intent-to-treat (ITT) population included all enrolled participants in Part 1, whether or not they received the study drug. Part 2: All Randomized Population (ARP) included all participants randomized, categorized by their randomized treatment group (whether or not they received study drug).

Only the first 6 of 11 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Part 1: Change From Baseline in Peak Work During Maximal Exercise Testing at Week 12, measured over Baseline, Week 12.

The primary endpoint was change from baseline in peak work during maximal exercise testing at Week 12, measured in watts per kilogram. A participant exercised on a recumbent stationary bicycle while workload increased incrementally. Peak work was defined as the workload reached at maximal volition, when exhaustion prevented continued exercise. A positive change from baseline was defined in the registry as suggesting improvement.

This endpoint captures performance during a structured physical-effort test. It is different from a rating scale, questionnaire, timed walking task, or count of falls: it quantifies the workload achieved during a cycling test. The trial summary describes physical effort, movement, coordination, and daily-life experience as areas of interest, and includes the cycling measure among the study’s assessments of physical effort.

Change from baseline focuses on how the measured value at follow-up differs from the participant’s baseline value. Comparing that change between randomized groups is intended to distinguish the observed course in the omaveloxolone groups from the course in the placebo group. The registry reports least-squares means and confidence intervals for the primary endpoint, and reports model-based net differences between dose groups and placebo. The endpoint is a measure of test performance; by itself it does not cover every aspect of Friedreich’s ataxia or everyday functioning.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is continuous: peak work is measured on a numeric scale, and the analysis concerns change from baseline. The registry reports least-squares means for the groups and least-squares mean differences for comparisons. These are model-based summaries rather than simple descriptions of one observed participant or one raw score. The reported primary method is mixed models repeated measures, abbreviated MMRM.

What does a confidence interval convey?

A confidence interval expresses the uncertainty around an estimated mean or difference under the assumptions of the analysis. A narrower interval indicates greater precision; a wider interval indicates less precision. For a difference between groups, zero represents no difference on the measured scale. When an interval includes zero, the data do not distinguish the estimate from zero at the confidence level used, although the interval may still include effects in either direction. This is not the same as showing that the groups are equivalent.

The primary endpoint’s reported comparison intervals are 95% confidence intervals. For the individual Part 1 dose comparisons, each interval spans zero. The group means and contrasts should be interpreted together: a point estimate gives a central estimate, while its interval shows how uncertain that estimate is. In this trial, the small numbers analyzed in the Part 1 groups are an important reason not to overread modest differences in point estimates.

What does MMRM contribute?

In general, a repeated-measures model analyzes repeated observations over time while accounting for the fact that measurements from the same participant are related. A mixed model can represent group and time patterns and estimate adjusted means or contrasts. The registry identifies MMRM for the primary comparisons, but details of the model specification, covariance structure, covariates, and missing-data assumptions are not reported in the registry record. Those details can affect how to reproduce or fully assess a model-based result.

The registry labels the primary estimates least-squares means and net differences. In general, least-squares means are model-based group means, and a net difference is a modeled contrast between groups. They should not be confused with the raw mean change calculated directly from every available observation. The reported model-based analysis is the sponsor’s own analysis of record for these comparisons.

How do the recalculations differ?

For selected secondary continuous outcomes, the posted recalculations compare summary means using an unadjusted approach. These are simple comparisons based on the reported group summaries, rather than the full repeated-measures model. They are useful as a transparent, approximate way to examine the scale and uncertainty of a difference, but they do not reproduce adjustments or modeling choices used in the sponsor’s analysis. Where a sponsor analysis is available, it should take precedence for the trial’s reported inferential result; the recalculation is supplementary.

The distinction matters because the registry’s Part 2 MMRM result for peak work is a model-based least-squares mean difference, whereas the recalculation uses the posted means and standard deviations. Such estimates need not match exactly. Differences can arise from the model, included observations, or adjustment choices. No discrepancy flag is recorded for the primary comparison, but the secondary summaries should still be interpreted according to the method attached to each estimate.

Why do over-dispersion and variability matter?

In general, variability describes how spread out participant measurements are around a group summary. Greater variability makes it harder to estimate a group mean or difference precisely, particularly when the analyzed groups are small. For outcomes such as falls, a few participants with high counts can contribute substantially to the mean and spread. If counts vary more than a simple statistical model expects, over-dispersion can make standard errors too small if it is ignored. The posted falls summary gives means and standard deviations, but the registry entries do not establish that a count-specific model or over-dispersion adjustment was used for that recalculation.

The sample-size figures are planning illustrations, not observed power. They use the observed effect as a starting point and assume a two-sided significance level of 0.05, equal allocation, and a normal approximation. The effect to detect is expressed as a fraction of the observed effect. These assumptions are not the trial’s documented design justification. The protocol’s sample-size rationale and any interim analyses are not reported in the registry record. Planning illustrations cannot show whether the completed trial had a particular probability of detecting an effect.

10. Analyses Reported by the Sponsor

The sponsor’s primary analysis uses MMRM and reports least-squares mean differences between Part 1 omaveloxolone groups and placebo. The reported net differences range from -0.13 to 0.1 W/kg; the reported 95% confidence intervals for the dose-specific comparisons include zero. The across-group result is reported as 0, with a 95% confidence interval from -0.08 to 0.08. Those sponsor analyses are the primary inferential results for the registered endpoint.

The unadjusted recalculations based on posted summaries are not needed to replace those primary results. They are simple checks based on available summaries and do not recreate the MMRM. For the secondary outcomes, the same distinction applies: where the registry provides an MMRM result, that model-based result is the sponsor’s analysis; a summary-statistic recalculation is supplementary. If a discrepancy between a recalculation and the sponsor’s result were flagged, the sponsor’s own analysis would take precedence. The record does not flag a discrepancy for this trial.

The recalculated estimates should therefore be used to understand scale and uncertainty, not as an alternative official analysis. The registry record does not report all model details needed to reproduce the sponsor’s estimates exactly. The sponsor’s analysis and the recalculation also answer slightly different computational questions when one adjusts for repeated measures or other model terms and the other does not.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary endpoint and 8 secondary outcomes, and reports 22 analyses. Part 1 also involved several dose groups compared with placebo. This creates a multiplicity issue: when many comparisons or outcomes are examined, the chance of at least one result appearing notable by chance can increase. The interpretation of individual p-values or confidence intervals therefore depends partly on whether and how the analysis plan controlled the family of tests.

A multiplicity adjustment, a prespecified hierarchy, or a clear distinction between confirmatory and exploratory outcomes can help explain how multiple findings should be weighed. Such a strategy is not reported in the registry record. It is also not reported whether interim analyses took place, or what stopping or dose-selection rules applied. The brief study description says Part 1 comparisons were used to decide which dose to use in Part 2, but the precise decision procedure is not reported.

Missing-data methods also affect interpretation. The primary endpoint description defines the Part 1 intention-to-treat population as all enrolled participants in Part 1, whether or not they received study drug, while the number analyzed means participants with data available for that outcome. The actual analyzed group counts are small. The registry identifies MMRM, but it does not report the detailed missing-data assumptions, the amount or pattern of missing primary endpoint data, or sensitivity analyses assessing departures from those assumptions. Thus, the population label alone does not establish that every randomized participant contributed an observed Week 12 endpoint.

Part 2 and extension results involve different follow-up and treatment arrangements from the Part 1 primary comparison. The record does not provide a full account of how missing observations or treatment changes were handled for every endpoint. The protocol’s sample-size justification is also not reported. The planning illustrations are based on assumed conditions and observed effects, and should not be treated as evidence of observed power or as a record of the trial’s original design assumptions.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: drug withdrawal syndrome (5.9%), burns third degree (5.9%), left ventricular failure (2.3%), sinus tachycardia (2.3%), constipation (2.3%).

The registry lists serious adverse events, other adverse events, and deaths by study group. These are counts of participants with events in the recorded groups, not estimates of the probability that treatment caused an event. Across the listed groups, the number at risk and the number with any event vary, so comparisons need to be read with those denominators in mind. The record lists 0 deaths in each reported group.

For serious adverse events, the Part 1 placebo group has 2 affected participants among 17 at risk; the other Part 1 groups list 0 affected participants. In Part 2, 5 of 51 participants in the omaveloxolone group and 3 of 52 in the placebo group are listed as affected. In the extension groups, the registry lists 12 of 106 and 13 of 43 participants with serious adverse events. These raw counts are descriptive. The extension groups reflect prior treatment histories and are not a simple randomized comparison with one another.

The listed serious-event terms include drug withdrawal syndrome and third-degree burns, each reported as 5.9%, as well as left ventricular failure, sinus tachycardia, constipation, and diarrhoea, each reported as 2.3%. Those percentages do not establish a causal relationship to study treatment. Interpretation requires context about event definitions, timing, exposure, and clinical review; those details are not reported in the registry record. The adverse-event counts alone do not determine an individual’s risk or provide medical advice.

13. What the Result Does, and Does Not, Mean

The results show what was measured in this trial: change in cycling peak work over 12 weeks in Part 1, with additional functional and safety measures across later study parts. For the primary endpoint, the posted dose-versus-placebo estimates are not precise enough to establish a clear difference. The reported confidence intervals include zero, and the number of participants with primary endpoint data in the individual Part 1 groups is small.

That does not prove that omaveloxolone and placebo have identical effects. An interval that includes zero can also include nonzero differences. The important point is that the posted estimates leave uncertainty about the size and direction of any difference. Nor does a positive within-group change establish a treatment effect: the placebo group also has a change estimate, and the relevant question is the between-group contrast.

The Part 2 findings address a different dose and follow-up period than the primary Part 1 question. Secondary measures capture other aspects of function, daily life, and events, but they should not be treated as substitutes for the registered primary endpoint or combined into an overall benefit claim. The registry’s safety counts likewise describe events recorded during the study; they do not by themselves establish that an event was caused by the study drug.

These findings apply to the enrolled trial population and the recorded study conditions. They do not settle every question about longer-term outcomes, every measure of Friedreich’s ataxia, or treatment effects outside the study. The registry record does not report enough information to reconstruct every modeling and missing-data decision. Statistical interpretation should therefore keep the endpoint, study part, comparator, analyzed sample, and uncertainty attached to each result.

14. Why This Trial Matters Statistically

This trial is statistically interesting because its primary question sits within a multi-part, dose-ranging design. Part 1 compares several doses with placebo over 12 weeks, while Part 2 evaluates omaveloxolone 150 mg and placebo over 48 weeks. Those are related questions, but the Part 1 primary endpoint and Part 2 secondary measures are not one pooled experiment. Keeping the parts distinct prevents a later outcome or a different dose from being mistaken for the primary result.

The primary endpoint also illustrates why point estimates alone are not enough. The group estimates vary in direction, and the reported comparisons have confidence intervals that include zero. With only a few participants analyzed in each Part 1 group, a seemingly noticeable point estimate can remain highly uncertain. The interval and analyzed sample size are central to interpreting the result, not optional technical details.

The study assesses more than a single laboratory-like measure. Cycling performance, rating scales, timed tasks, global impressions, falls, and daily-living activities represent different types of evidence. That breadth may help describe multiple dimensions of the condition, while also making multiplicity and outcome selection important. Without a reported testing hierarchy or adjustment strategy, secondary findings should be interpreted cautiously.

Finally, the record distinguishes sponsor-model results from simple recalculations based on summary statistics. That distinction matters because adjusted repeated-measures analyses and unadjusted comparisons can yield different estimates. Showing both kinds of evidence can clarify what is directly reported and what is supplementary, provided neither is mistaken for the other.

15. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

Several features limit how much can be concluded from the posted results. The primary endpoint analysis includes small numbers of participants in the individual Part 1 groups, despite total study enrollment of 172. The total enrollment is not the same as the number contributing data to a specific endpoint, and the registry defines the analyzed count by available outcome data. This makes the primary estimates imprecise and limits the ability to distinguish modest differences.

Part 1 includes multiple dose comparisons, and the registry lists many secondary outcomes and analyses. The multiplicity approach is not reported, so the evidential status of individual secondary findings is difficult to assess fully. The protocol’s sample-size rationale, interim-analysis details, and the dose-selection procedure are also not reported in the registry record.

The registry identifies MMRM for the primary analysis but does not report enough model detail to reproduce it fully. Specific covariates, covariance structure, missing-data assumptions, and sensitivity analyses are not reported. For secondary outcomes, the available unadjusted calculations are based on summary statistics and do not replicate model-based analyses. The falls summary in particular should not be read as a full count-model analysis; the registry does not report whether over-dispersion was modeled.

The study has multiple parts with differing treatment exposure and follow-up. Part 2 and extension data therefore should not be directly treated as equivalent to the randomized Part 1 comparison. Finally, adverse-event counts are descriptive and do not establish causality. The registry record does not report all event-level context needed to interpret those counts in depth.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: A Study to Learn About the Effects and Safety of RTA 408 (Omaveloxolone) in People Aged 16 to 40 With Friedreich's Ataxia — NCT02255435.
• Registry record last updated 2026-09-03; status last verified 2026-08.
• Results first posted 2022-11-29.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

MOXIe was a completed phase 2 randomized, double-masked, parallel-group study sponsored by Biogen, enrolling 172 people with Friedreich’s ataxia. Its primary endpoint was change from baseline in peak work during maximal exercise testing at Week 12 in Part 1. Part 1 evaluated multiple omaveloxolone doses against placebo; Part 2 compared 150 mg with placebo over 48 weeks, followed by an optional extension.

The sponsor-reported Part 1 dose comparisons have 95% confidence intervals that include zero, and analyzed group sizes for the primary endpoint are small. Available Part 2 secondary results are also uncertain and should be interpreted by outcome and analysis method. The registry reports serious adverse events and deaths by group, but those descriptive counts do not establish causation. Details on multiplicity, missing-data assumptions, interim analyses, and the original sample-size justification are not reported in the registry record.