Reimbursement Review

Omaveloxolone (Skyclarys)

Sponsor: Biogen Canada Inc.

Therapeutic area: Friedreich’s ataxia

This multi-part report includes:

Clinical Review

Pharmacoeconomic Review

Ethics Review

Clinical Review

Abbreviations

Executive Summary

An overview of the submission details for the drug under review is provided in Table 1.

Table 1: Background Information on the Application Submitted for Review

NOC = Notice of Compliance.

Introduction

Friedreich’s ataxia (FA) is a rare autosomal recessive ataxia caused by loss-of-function mutations from trinucleotide repeat expansions in the FXN gene located on chromosome 9q13.^1,2 FA accounts for up to one-half of all hereditary ataxia cases and affects approximately 1 in 30,000 to 1 in 50,000 individuals.² In Canada, estimates suggest that 300 to 1,000 patients are affected.^3-5 FA typically presents in childhood or adolescence with a complex neurologic phenotype characterized by progressive gait ataxia. Additional cerebellar signs including dysarthria and dysphagia; peripheral motor and sensory neuropathy combined with pyramidal signs; and, in advanced disease, visual and hearing impairment.^6-8 Most affected individuals develop hypertrophic cardiomyopathy, which in some cases may precede the onset of ataxia. Diabetes mellitus and skeletal abnormalities, such as pes cavus and scoliosis, are also common. The disease shortens life expectancy considerably because of cardiac complications, with a mean age at death of 37 years. Currently, there are no approved disease-modifying therapies available in Canada for FA. Management focuses on supportive care, rehabilitation, and symptomatic treatment of complications.^9,10

The objective of this report is to review and critically appraise the evidence submitted by the sponsor on the beneficial and harmful effects of omaveloxolone, administered at a recommended dose of 150 mg orally once daily (as 3 capsules of 50 mg each), in the treatment of FA in adults and adolescents aged 16 years and older.

Perspectives of Patients, Clinicians, and Drug Programs

The information in this section is a summary of the input provided by the patient and clinician groups who responded to our call for input and from clinical experts consulted for the purpose of this review.

Patient Input

Four patient groups — Muscular Dystrophy Canada (MDC), the National Ataxia Foundation (NAF), the Ataxia Canada – Claude St-Jean Foundation, and Friedreich’s Ataxia Research Alliance (FARA) — provided input. MDC identified and contacted adults living with FA and parents of children aged 16 years and over living with FA to participate in a health care experience survey and semistructured virtual interviews. The organization gathered insights from 85 individuals (aged 16 years to 70 years) with a confirmed FA diagnosis on diagnostics delays, gaps in treatment, emotional and social effects, and access to care and support systems. NAF conducted a survey of its community members with FA living in different provinces across Canada and received 14 responses (from 9 people with FA and 5 caregivers). Ataxia Canada gathered the experiences of patients and caregivers in Canada through a combination of interviews and a survey and received 85 responses. Feedback from FARA regarding the disease experience was drawn from 2 sources: a white paper that included views from parents of children with FA living in the US, highlighting the importance of pediatric inclusion in clinical trials, and a patient-focused drug development meeting that included 145 patients and caregivers, during which participants were polled about their current disease state, experience with different symptoms, and perspectives on future treatments.

Ataxia Canada and MDC highlighted that patients with FA experience significant impacts with respect to coordinating and/or maintaining balance; mobility and scoliosis; productivity at home and work; independence and social participation; and mental health. Additionally, FARA noted that patients and caregivers indicated experiencing neurologic symptoms (e.g., regular falls and trouble with balance, walking, and coordinating hands and/or arms) and fatigue. NAF, Ataxia Canada, and FARA indicated that people with FA and their caregivers spend hours engaging in symptom-based therapies (such as occupational therapy, speech therapy, and physiotherapy), visiting medical specialists, and obtaining mobility devices, which can be challenging, expensive, and time-consuming.

The patient groups agreed that there is a significant need for a treatment that is a cure for FA or can reverse its effects. Patients are also seeking a treatment that slows and/or stops disease progression, promotes better symptom management, helps them regain and/or preserve mobility, increases energy levels, improves health-related quality of life (HRQoL) and independence, prevents complications (such as scoliosis or diabetes), and reduces the physical and emotional burden on families and caregivers. Currently, individuals require genetic testing to confirm the presence of biallelic mutations of the FXN gene. MDC and NAF reported that the testing process was easy and provided clarity about the diagnosis but was often emotionally challenging.

Four respondents (1 each from MDC and NAF and 2 from Ataxia Canada) shared their experiences with accessing omaveloxolone through clinical trials. All 4 respondents (2 of whom were already wheelchair users) highlighted that being on the drug slowed their disease progression. The viewpoints gathered by FARA from individuals with FA and parents of individuals with FA who participated in the MOXIe trial included slowed progression, longer retention of motor function (i.e., improved endurance, speech, and ability to walk, stay upright, and eat), better ability to cope with fatigue, and minimal side effects (i.e., transient elevation of liver enzymes, cholesterol, headache, nausea, and diarrhea).

Clinician Input

Input From Clinical Experts Consulted for This Review

The information in this section is based on the input received from a panel of 4 clinical specialists consulted by Canada’s Drug Agency (CDA-AMC) for the purpose of this review.

The clinical experts emphasized that there are currently no approved disease-modifying treatments for FA in Canada; thus, there is a critical unmet need. Current management relies on supportive care and symptom management, which do not alter the underlying progressive disease course.

The experts indicated that omaveloxolone would represent the first disease-modifying therapy for FA and be positioned as a first-line option for eligible patients aged 16 years and older with genetically confirmed disease. They noted that it would be used alongside existing supportive care measures rather than replacing them.

Regarding patient selection, the experts suggested that while all patients with confirmed FA could theoretically benefit, those at earlier disease stages may show more discernible stabilization before irreversible neurologic damage occurs. However, they emphasized that nonambulatory patients should not be excluded, given that preserving upper limb and/or bulbar function could still provide meaningful benefit.

With respect to assessing treatment response, the experts acknowledged challenges in translating the clinical trial measures to routine practice. While the modified Friedreich’s Ataxia Rating Scale (mFARS) was used in trials and natural history studies, it is not commonly employed in clinical settings. Furthermore, the experts suggested monitoring for 2 years to establish efficacy but noted that even partial slowing of disease progression may be valuable. According to the experts, treatment discontinuation should be based primarily on safety and tolerability rather than on lack of improvement alone, given the progressive nature of FA, the variability of the disease, its rate of progression, and the absence of alternative disease-modifying options.

The experts emphasized that diagnosis and treatment should be guided by specialists experienced in FA management, such as neuromuscular or movement disorder neurologists. They also noted that, while regular monitoring is important, requiring intensive specialized outcome measures (such as standardized scales designed for clinical trials that require trained users for reliable administration) could limit equitable access to treatment.

Clinician Group Input

A single clinician group input was received from the Neuromuscular Disease Network for Canada. Input from 4 clinicians familiar with clinical trials of FA treatments, specifically omaveloxolone, was gathered through 1-to-1 submissions and group discussions. The group noted that the primary therapeutic goals are to slow disease progression, preserve or enhance function, extend survival, and improve well-being.

The clinician group indicated that omaveloxolone is poised to be incorporated into the current treatment paradigm. However, the group also highlighted that there is not enough evidence to establish which patients are most likely to respond to omaveloxolone. Regarding the outcomes used to determine a patient’s response, the clinician group indicated standardized tests used in neurologic exams (i.e., mFARS) and functional assessments (such as the Friedreich’s Ataxia Rating Scale – Activities of Daily Living [FARS-ADL]). Measurements every 6 months in the first year and then annually were noted as reasonable and practical. In terms of a clinically meaningful response to treatment, the group noted that there should be an improvement in patient function and well-being. For example, this could be reflected by just a 1-point improvement in upright stability score compared to expected progression, indicating preserved balance in the short term and predicting delayed loss of ambulation.

The clinician group noted that omaveloxolone may be discontinued due to lack of efficacy (determined after 1 year, based on Clinical Global Impression of Change [CGIC] and Patient Global Impression of Change [PGIC]) or side effects (i.e., evidence of organ dysfunction). Additionally, it was recommended that a statin be prescribed to manage the cardiovascular risk factors associated with increased low-density lipoprotein (LDL) (a common side effect of omaveloxolone); this was deemed a better approach than discontinuing omaveloxolone. Lastly, the clinician group highlighted that people with FA must be treated at specialized centres that offer comprehensive interdisciplinary care, regardless of omaveloxolone treatment. For patients without easy access to such centres, care should be managed by a neurologist who is knowledgeable about the disease and its management.

Drug Program Input

Drug plans submitted questions concerning comparators as well as the initiation, renewal, discontinuation, and generalizability of omaveloxolone. The clinical expert panel convened by CDA-AMC provided advice on the potential implementation issues raised by the drug program. Refer to Table 4 for more details.

Clinical Evidence

Systematic Review

Description of Studies

One pivotal, phase II, randomized controlled trial (RCT) (the MOXIe Part 2 trial; N = 103) was included in the review to evaluate whether omaveloxolone 150 mg once daily improved modified mFARS scores compared to placebo after 48 weeks of treatment in patients aged 16 years to 40 years with an mFARS score of 20 to 80 and genetically confirmed FA. The trial included secondary end points assessing changes in activities of daily living (ADLs), upper limb function (9-hole peg test [9-HPT]), mobility (25-foot timed walk test), and frequency of falls.

Despite randomization, there were imbalances in baseline characteristics between the treatment groups. Compared to the placebo group, the omaveloxolone group had a higher proportion of males (60% males versus 33% females), more patients with cardiomyopathy (48% versus 29%), higher baseline mFARS scores (40.94 versus 38.77), and more patients with guanine-adenine-adenine 1 (GAA1) ( i.e., the shorter expanded allele) repeat lengths of greater than or equal to 675 (███ ██ ██%). The mean age was similar between groups (approximately 24 years), and the majority of patients in both groups were ambulatory (93%), ██████ █████ ████ ████████ █████████ ███████. Other disease characteristics were generally balanced, including the mean age of FA onset (15 years), disease duration (4.8 years), and prevalence of conditions like scoliosis (74%) and sensory neuropathy (49%).

Efficacy Results

Change in mFARS Score at Week 48

The mFARS measures neurologic function across 4 domains: bulbar function, upper limb coordination, lower limb coordination, and upright stability. Scores range from 0 to 93, with higher scores indicating greater impairment. Assessments were conducted at baseline and week 48.

In the prespecified primary analysis population (full analysis set [FAS]), patients receiving omaveloxolone had a mean baseline mFARS score of 40.94 points (standard deviation [SD] = 10.39 points) and showed a mean improvement (decrease) from baseline of █████ ██████ ███ █████) at week 48. The placebo group had a mean baseline score of 38.77 points (SD = 11.03 points) and showed a mean worsening (increase) from baseline of █████ ██████ ███ █████). The mixed model for repeated measures (MMRM) estimate of mean difference in change from baseline between the omaveloxolone and placebo groups was −2.40 points (95% confidence interval [CI], −4.31 to −0.50 points; P = 0.0141).

In the all-randomized population, which included patients with severe pes cavus, omaveloxolone treatment improved mFARS scores by an estimated mean difference of −1.93 points relative to placebo (95% CI, −3.70 to −0.15 points; P = 0.0342). The mean baseline scores were █████ ███ █████) for omaveloxolone ███ █████ ███ █████) for placebo, with mean changes from baseline ██ █████ ███ █████) and ████ ███ █████), respectively.

Change in Performance on the 9-HPT

The 9-HPT measures upper-extremity function based on the time taken to place and remove 9 pegs in a pegboard. The test was performed at baseline and week 48, with faster times indicating better function. Results are reported as the reciprocal of average time (1 per second) for the dominant hand.

In the FAS population, patients receiving omaveloxolone had a mean reciprocal of average time (1 per second) baseline value of ██████ ███ ███████) and showed little to no change from baseline (mean change ████████ ██ █████) at week 48. The placebo group had a mean reciprocal of average time (1 per second) baseline value of ██████ ███ ███████) and showed little to no change (mean change ████████ ██ ███████). The MMRM estimated mean difference between groups was ██████ ████ ███ ███████ ██ ███████ | | ██████).

Change in Performance on a 25-Foot Timed Walk Test

The 25-foot timed walk test measures mobility based on the time taken to walk 25 feet. Assessments were conducted at baseline and week 48, with results reported as the reciprocal of average walk time (1 per second). Higher values indicate better function.

In the FAS population, patients receiving omaveloxolone had a mean baseline score of ██████ ███ ███████) and showed a decline from baseline of –██████ ███ ███████) at week 48. The placebo group had a mean baseline score of ██████ ███ ███████) and showed a decline of –██████ ███ ███████). The MMRM estimated mean difference between groups was 0.0058 (95% CI, –0.0099 to 0.0214; P = 0.4635).

Frequency of Falls at Week 48

Falls were recorded daily by patients in a study diary throughout the 48-week treatment period. A fall was defined as “the patient unintentionally coming to rest on the ground or at a lower level.”

In the FAS population, patients receiving omaveloxolone reported a mean of ████ █████ ███ █████) during treatment compared to ████ █████ ███ █████) in the placebo group. The Poisson estimated difference in the incidence rate of falls between the omaveloxolone and placebo groups was █████ ████ ███ █████ ██ █████ | | ██████).

ADLs at Week 48

The ADL assessment included 9 questions evaluating speech, swallowing, cutting food and/or handling utensils, dressing, performing personal hygiene activities, falling, and walking, as well as bladder function and quality of sitting position. The total scores range from 0 to 36, with higher scores indicating greater impairment. Assessments were conducted at baseline and week 48.

In the FAS population, patients receiving omaveloxolone had a mean baseline ADL score of 10.738 (SD = 4.7663) and showed an improvement (decrease) from baseline of ██████ ██████ ███ ██████) at week 48. The placebo group had a mean baseline score of 9.869 (SD = 4.8339) and showed a worsening (increase) of ██████ ██████ ███ ██████). The MMRM estimated mean difference between groups was –1.30 points (95% CI, █████ ██ ██████ P = 0.0420).

Harms Results

All patients in both treatment groups experienced at least 1 adverse event (AE) during the 48-week trial. Common AEs occurring more frequently with omaveloxolone than with placebo included nausea (33.3% versus 13.5%), abdominal pain (21.6% versus 5.8%), diarrhea (19.6% versus 9.6%), fatigue (21.6% versus 13.5%), and increased liver enzymes (alanine aminotransferase [ALT] = 37.3% versus 1.9%; aspartate aminotransferase [AST] = 21.6% versus 1.9%). Serious adverse events (SAEs) occurred in 9.8% of patients receiving omaveloxolone versus 5.8% of patients receiving placebo. Treatment discontinuations due to AEs were more frequent in the group receiving omaveloxolone (7.8% versus 3.8%). Notable harms of special interest included liver enzyme elevations, which occurred more often in the group receiving omaveloxolone. No deaths were reported during the study period.

Critical Appraisal

Overall, the MOXIe trial demonstrated acceptable internal validity, benefiting from its randomized, double-blind, placebo-controlled design, proper allocation concealment, validated primary outcome measure, and appropriate statistical methods. Key limitations affecting internal validity included baseline imbalances between treatment groups despite randomization, with the omaveloxolone group having characteristics suggesting more advanced disease (i.e., higher mFARS scores, longer GAA1 repeat lengths, and a greater proportion of patients with cardiomyopathy). Furthermore, the imbalance in the male-to-female ratio between the groups receiving omaveloxolone and placebo may also bias the results against omaveloxolone, as observed in subgroup analyses showing better responses among males. The impact of some of these imbalances was explored in post hoc analyses that suggested a potential underestimation of treatment effect in the primary end point. Additionally, a higher discontinuation rate in the group receiving omaveloxolone versus placebo (13.7% versus 3.8%), primarily due to AEs, raises concerns about potential bias from missing data. The absence of an established minimum clinically important difference for mFARS also creates uncertainty in interpreting the clinical significance of the observed treatment effect. All outcomes other than the primary outcome either include the null or are outside the statistical testing hierarchy.

Overall, the external validity of the MOXIe Part 2 trial was limited by eligibility restrictions, the exclusion of patients with significant cardiac issues, and the capping of the number of patients with severe pes cavus. The sponsor justified the decision to cap the number of patients with severe pes cavus based on evidence from the MOXIe Part 1 trial suggesting that these patients may represent a different subtype of FA and also because their severe pes cavus would likely interfere with their ability to perform assessments that required standing or pedalling. Clinical experts involved in this review suggested that all patients with FA have some degree of pes cavus, which has not been shown to be clinically prognostic. The trial excluded patients aged younger than 16 years despite most patients being diagnosed at around 11 years of age; this exclusion created uncertainty about treatment effects in pediatric populations. However, the current Health Canada indication restricts the population to patients aged 16 years or older. This limitation to external validity is of limited impact if the drug is prescribed according to the indication. The restriction to patients with mFARS scores ranging from 20 to 80 and the exclusion of those with significant cardiac issues limits generalizability to patients with more severe disease. Furthermore, the outcomes used in the trial, particularly mFARS scores, are not routinely implemented in clinical practice, a situation that creates challenges when it comes to translating trial results to real-world assessment of treatment response. While the absence of Canadian sites was noted, clinical experts did not consider this a major limitation, given the similarity of the patient population and treatment approaches across countries.

GRADE Summary of Findings and Certainty of the Evidence

The selection of outcomes for the Grading of Recommendations Assessment, Development and Evaluation (GRADE) assessment was based on the sponsor’s Summary of Clinical Evidence, consultation with clinical experts, and input received from patient and clinician groups and public drug plans. The following list of outcomes was finalized in consultation with expert committee members:

• change in mFARS score

• change in 9-HTP performance

• change in 25-foot timed walk test results

• change in frequency of falls

• change in ADLs.

Table 2: Summary of Findings for Omaveloxolone vs. Placebo for Patients With FA

9-HPT = 9-hole peg test; ADL = activities of daily living; CDA-AMC = Canada’s Drug Agency; CI = confidence interval; FA = Friedreich's ataxia; mFARS = modified Friedreich’s Ataxia Rating Scale; MID = minimal important difference; NA = not applicable; RCT = randomized controlled trial; vs. = versus.

Note: Study limitations (which refer to internal validity or risk of bias), inconsistency across studies, indirectness, imprecision of effects, and publication bias were considered when assessing the certainty of the evidence. All serious concerns in these domains that led to the rating down of the level of certainty are documented in the table footnotes.

^aNo published between-group MID was identified. Clinical experts consulted by CDA-AMC identified 2 points as the potential clinically meaningful threshold. Rated down 1 level down for imprecision because the upper bound of the 95% CI suggests no clinically meaningful difference, while the lower bound of the 95% CI suggests benefit.

^bNo published between-group MID was identified, and the clinical experts consulted by CDA-AMC were unable to estimate a threshold for clinically important effects; therefore, the null was used. Rated down 2 levels for very serious imprecision because the lower CI suggests harm while the upper CI suggests benefit of little to no difference.

^cNo published between-group MID was identified, and the clinical experts consulted by CDA-AMC identified 1 point as the potential clinically relevance threshold. Rated down 1 level for imprecision because the upper bound of the CI suggests no clinically meaningful difference, while the lower bound of the CI suggests clinically meaningful difference. Not rated down for imprecision; there is a between-group difference of less than the null and a CI that excludes the null.

Source: Sponsor’s Summary of Clinical Evidence.¹¹

Long-Term Extension Study

Description of Study

One ongoing, open-label extension (OLE) study was included to assess the long-term safety and tolerability of omaveloxolone in patients with FA following the completion of Part 1 or Part 2 of the MOXIe trial. The study enrolled 149 patients from the MOXIe Part 1 and MOXIe Part 2 trials, including 106 patients who were omaveloxolone-naive (referred to as the placebo-omaveloxolone group) and 43 who had previously received omaveloxolone (referred to as the omaveloxolone-omaveloxolone group). All patients received omaveloxolone 150 mg daily, with interim analysis data available up to 144 weeks. Baseline characteristics were generally balanced between groups; however, the placebo-omaveloxolone group had a higher proportion of males than females (55.7% versus 37.2%) and patients with pes cavus (█████ ██ ████%).

Efficacy Results

At 144 weeks, the mean changes from baseline in mFARS score were 3.37 points (SD = 4.939 points) in the placebo-omaveloxolone group and 2.28 points (SD = 5.896 points) in the omaveloxolone-omaveloxolone group. ADL scores showed mean increases (indicating worsening) of 1.873 points in the placebo-omaveloxolone group and 1.286 points in the omaveloxolone-omaveloxolone group at week 144. Outcomes for additional functional measures, including the 9-HPT and 25-foot timed walk test, showed similar patterns.

Harms Results

The safety profile in the OLE was consistent with that of the controlled trial. Common AEs included coronavirus infection (18.8%), increased ALT (18.8%), headache (18.1%), upper respiratory tract infection (16.8%), nausea (16.1%), and fatigue (13.4%). SAEs occurred in 8.7% of patients (7.5% in the placebo-omaveloxolone group, 11.6% in the omaveloxolone-omaveloxolone group); ████ discontinued due to AEs (███% in the placebo-omaveloxolone group, ███% in the omaveloxolone-omaveloxolone group). Liver enzyme elevations remained a notable AE of special interest, but appeared to be manageable with monitoring. No deaths were reported.

Critical Appraisal

The main limitations of the long-term extension study are the lack of a control group, the open-label design with subjective outcomes, and potential selection bias (given that the study enrolled only patients who had completed the original trials). There is also a risk of attrition bias because the number of patients contributing to the analyses declined steadily over time; the final outcome measures are based on less than half the number of originally enrolled patients. In addition, the COVID-19 pandemic affected study visits and treatment continuity, with 14.8% of patients experiencing treatment interruptions. The use of historical controls for contextualizing progression rates, while informative, has inherent limitations due to potential differences in patient populations and assessment methods.

The clinical experts consulted for this review suggested that, aside from the exclusion of pediatric patients, the eligibility criteria of the OLE resulted in a study population comparable to patients in Canada. However, the trial’s strict inclusion and exclusion criteria, including constraints around cardiac involvement, may have led to a cohort that was healthier than what is typically encountered in routine clinical practice involving the treatment of FA in Canada. Furthermore, the study did not include any sites in Canada, which may reduce the generalizability and applicability of the results to Canadian practice.

Indirect Comparisons

No indirect treatment comparisons were submitted for this review.

Studies Addressing Gaps in the Evidence From the Systematic Review

Description of Studies

A propensity score–matched analysis compared long-term outcomes between patients in the MOXIe trial extension (N = 136) and matched controls from the Friedreich’s Ataxia Clinical Outcome Measures Study (FACOMS) natural history database (N = 136). Patients were matched on key characteristics, including age, sex, baseline mFARS score, age at FA onset, and baseline gait score, with a mean follow-up of approximately 2.5 years in both cohorts.

Efficacy Results

• The estimated 3-year change from baseline in mFARS score was 6.61 points (standard error [SE] = 0.65 points) in matched FACOMS patients compared to 3.00 points (SE = 0.66 points) in patients in the MOXIe trial extension, representing a statistically significant difference of −3.61 points (95% CI, −1.79 to −5.43 points) favouring omaveloxolone. This suggests that patients in the MOXIe OLE experienced slower increases in mFARS scores than patients included from the matched FACOMS group (indicative of slowed disease progression).

Harms Results

Safety outcomes were not assessed in this analysis.

Critical Appraisal

The choice of study design was considered appropriate, given the constraints of the rare disease. The real-world evidence (RWE) used comparative evidence to describe treatment efficacy in a treatment-naive population; however, the choice of baseline in the FACOMS group could introduce indication bias due to unmeasured confounding factors. While the timing of treatment initiation was not an issue, the primary analysis cohort — patients who had completed 48 weeks of follow-up — may have differed systematically from patients in the FACOMS group. Limited bias due to exposure or outcome misclassification was noted; however, measurement error at year 3 could slightly favour omaveloxolone. Propensity score–matching variables were sufficient, and diagnostic results showed comparability between the FACOMS and MOXIe cohorts. However, the estimation of progression at 3 years relied on a missing at random (MAR) assumption, with missing outcome analyses not provided, raising concerns about dropout due to AEs.

Conclusions

Evidence from 1 phase II RCT suggests that, compared to placebo, omaveloxolone likely results in slower decline in neurologic function, as measured by mFARS scores over 48 weeks in patients with FA aged 16 years and older. It likely results in improvements in ADL measures, such as activities involving upper limb function and mobility, and may reduce the frequency of falls. While this is suggestive of functional relevance, with the clinical experts suggesting 2-point and 1-point potentially meaningful thresholds for mFARS and ADL, respectively, the actual clinical significance of these observations is uncertain due to the lack of an established minimal clinically important difference in mFARS and ADL scores. Other functional outcomes, including upper limb function, mobility, and frequency of falls, did not show significant improvements with treatment.

Comparison with natural history controls suggest that the treatment benefit of omaveloxolone may be maintained for up to 3 years. However, these studies do not support clear conclusions to be drawn about whether disease progression will continue to be slower among patients treated with omaveloxolone than among matched controls. A long-term extension study suggests that benefit may be observed within the first year, given that treatment-naive patients were observed to have better responses. However, these findings are limited by the open-label design and potential selection bias, and the magnitude of these group differences is uncertain. The safety profile appears manageable, with liver enzyme elevations being the primary concern requiring monitoring. Common adverse effects, including gastrointestinal symptoms and fatigue, were generally mild to moderate in severity.

Important evidence gaps remain, particularly regarding efficacy in pediatric patients (who account for a significant proportion of newly diagnosed cases), efficacy in patients with more advanced disease or significant cardiac involvement, and long-term comparative efficacy (i.e., beyond 3 years). Additionally, while the mFARS captures aspects of neurologic function, there is limited evidence for outcomes identified as important by patients, such as HRQoL, fatigue, and ability to maintain independence. Given that FA is a progressive, debilitating condition with no approved disease-modifying treatment, even a modest slowing of progression could be meaningful to patients. However, uncertainty remains about the magnitude and durability of benefit.

Introduction

The objective of this report is to review and critically appraise the evidence submitted by the sponsor on the beneficial and harmful effects of omaveloxolone, administered at a recommended dose of 150 mg orally once daily (as 3 capsules of 50 mg each), for the treatment of FA in adults and adolescents aged 16 years and older.

Disease Background

The contents of this section have been informed by materials submitted by the sponsor and by clinical expert input. The following information has been summarized and validated by the review team.

FA is a rare autosomal recessive ataxia caused by loss-of-function mutations from trinucleotide repeat expansions in the FXN gene located on chromosome 9q13.^1,2 The majority of patients have an expanded guanine-adenine-adenine (GAA) trinucleotide repeat in intron 1 of both alleles of the FXN gene.² A longer GAA trinucleotide repeat translates to more FXN silencing, leading to an early onset of disease and rapid disease progression.^1,12 FA accounts for up to one-half of all hereditary ataxia cases.² It occurs in approximately 1 in 30,000 to 1 in 50,000 individuals,² and the GAA triplet repeat expansion that causes FA is found only in individuals of European, North African, Middle Eastern, or Indian origin.² Estimates based on 2010 Canadian data suggest that there are 300 to 750 patients with FA in Canada. More recent consultations with Ataxia Canada suggest that the number may be as high as 1,000.⁵

Common neurological symptoms to present at disease onset include gait instability (i.e., ataxia), which may manifest as clumsiness and/or increased falls, or dysmetria, which may manifest as problems with hand skills and fine motor tasks. While less common, patients may also present with sensory loss or dysarthria.⁶ Patients may also present with non-neurologic symptoms (i.e., cardiac involvement, diabetes mellitus, and skeletal abnormalities); however, this presentation is much less common, accounting for fewer than 10% of cases.^6,7 Onset of FA typically occurs in childhood or adolescence, generally appearing between 5 years and 20 years of age,⁸ with a mean age at onset of 10 years to 15 years. Age at disease onset is inversely correlated with the number of GAA repeats¹ and is an important predictor of overall disease severity and speed of progression.²

Diagnosis of FA typically occurs during childhood or adolescence.^7,8 It is based on clinical suspicion of symptoms (the majority of which are neurologic) and confirmed by genetic testing.⁹ Other suggestive clinical findings include musculoskeletal features, such as scoliosis or pes cavus; hypertrophic nonobstructive cardiomyopathy; optic atrophy and/or deafness; and endocrinological features, such as glucose intolerance or diabetes mellitus.¹³ Testing options include single-gene testing or a multigene panel.¹³ Single-gene testing is targeted at identifying abnormal GAA expansion in intron 1 of FXN; however, if only 1 expanded allele is detected (which happens in approximately 4% of cases), sequence analysis of FXN is also performed.^8,13 Multigene testing, in which other genes of interest are assessed in addition to FXN, is not recommended as a first-line strategy in typical cases. However, it may be helpful for atypical presentations.¹³ Genetic testing is readily available in Canada. Patients require a referral for a genetic test from a neurologist or geneticist.

Upon presentation of symptoms, a diagnosis of FA is typically confirmed within 2 years to 3 years.⁶ For patients who present with non-neurologic symptoms, the time to diagnosis may be longer because this is an uncommon presentation. In a European prospective, observational, natural history study of more than 600 patients with FA conducted by the European Friedreich's Ataxia Consortium for Translational Studies, the median time to diagnosis was 5 years for patients with non-neurologic onset of symptoms.⁶ Even after controlling for the effect of age at examination, age at onset, and presentation of symptoms before or after 1996 (when the FXN gene was discovered), a diagnostic delay of approximately 7 years was observed among patients.⁶ Another natural history analysis of patients with FA found that those with early onset declined 50% more quickly than patients with a more typical onset and twice as quickly as those with an intermediate onset.¹⁴ Results from a 4-year cohort analysis in the consortium study found that patients experienced progressively worse functioning and reduced ability to perform ADLs. Loss of ambulation was also identified as an important driver of disease progression, particularly contributing to declines in speech and upper limb function.¹⁵

FA considerably shortens life expectancy; patients with FA have a mean age of death of 37 years. Those with late-onset disease tend to have a longer life expectancy, with many living until the 40 years to 50 years.^9,10 Common causes of death due to FA include cardiac complications (60% of deaths), pneumonia, aspiration, diabetic coma, stroke, and trauma sequelae.^9,10,16 FA can result in significant morbidity and mortality. Because patients are at greater risk of developing comorbidities (e.g., cardiomyopathy and diabetes) and physical disabilities (e.g., scoliosis), they face greater limitations in their ability to engage in social and physical activities. They also experience more fatigue, pain, unsteadiness, emotional issues, and eating and drinking difficulties. They may feel an overall lack of control and experience losses of independence and autonomy, resulting in substantial impairments to their quality of life.

Standards of Therapy

Contents within this section have been informed by materials submitted by the sponsor and clinical expert input. The following has been summarized and validated by the review team.

According to clinicians consulted by CDA-AMC for this review, there are currently no approved disease-modifying therapies available in Canada for the treatment of FA. Therefore, management of FA focuses on supportive, rehabilitative, and symptomatic measures. The latest set of FA treatment guidelines was published in 2022.^17,18 At the time of publication, these guidelines stated that there was no approved pharmacological treatment for FA, but that research into potential therapeutic drugs had advanced considerably in the past 2 decades.¹⁸ The guidelines also stated that, despite significant progress in the search for disease-modifying drugs, the chronic, progressive course of FA cannot yet be significantly slowed;¹⁸ these guidelines were published before omaveloxolone became available. Management of FA has historically focused on symptomatic and supportive care.^8,9,19 As FA progresses, affected individuals experience worsening gait and limb ataxia, motor weakness, reflex and sensory loss, impairments in speech and swallowing, hearing loss, reduced visual acuity, and bladder dysfunction.⁷ Currently, given that there are no pharmacological treatments for weakness or ataxia,⁸ management approaches focus on orthotics, adaptive equipment, walking aids, wheelchairs, and physical therapy.¹⁹ Some neurologic features can be managed through selective pharmacological therapies. For example, spasticity can be managed with baclofen, tizanidine, or botulinum toxin; neuropathic pain can be managed with gabapentin or pregabalin; and urinary urgency can be managed with anticholinergic drugs, such as oxybutynin.⁸

Non-neurologic features of FA can include cardiac complications, diabetes mellitus, and skeletal abnormalities (including scoliosis and foot abnormalities).⁷ There are no FA-specific medications to prevent or treat cardiac disease progression; therefore, cardiomyopathy is typically managed according to general cardiology guidelines,⁸ including with antiarrhythmic drugs, anti–cardiac failure medication, anticoagulants, and pacemaker insertion.¹³ Diabetes mellitus in individuals with FA may be treated with diet changes and, if necessary, oral hypoglycemic medications or insulin.^7,13

Patients with FA may experience curvature of the spine and abnormalities of the feet. Nonsurgical interventions, such as physical therapy and bracing to stabilize the spine during growth, can be used to address these complications; however, in severe cases, surgery may be used.^8,19 Orthopedic surgeons, neurologists, physiatrists, or rehabilitation specialists should be consulted when making decisions to address such complications. Vision and hearing problems can be treated with corrective devices.¹⁹ Speech therapy can help patients maximize their verbal communication skills.¹⁹

Input from clinicians consulted by CDA-AMC for the purpose of this review suggests that nonpharmacological interventions form the foundation of FA management. These include physiotherapy, occupational therapy, speech and language therapy, orthotic supports for gait and upper limb function, and adaptive equipment to help maintain mobility and independence. Routine cardiac evaluations and interventions (e.g., medications for cardiomyopathy, heart rhythm management) are an integral component of care, given the high prevalence of cardiac involvement in FA. Psychological support and genetic counselling should also be offered to patients and their families. Pharmacological options in current Canadian practice are limited to symptomatic and adjunctive treatments. The use of coenzyme Q10 and vitamin E, while not strongly evidence-based, is common due to their theoretical antioxidant benefits and relatively benign safety profiles. As per the clinical experts consulted by CDA-AMC, these are generally well-tolerated, over-the-counter interventions thought to potentially support mitochondrial function or reduce oxidative stress; however, high-quality evidence demonstrating meaningful clinical impact is lacking. No other drugs are routinely used in Canada specifically to slow or halt disease progression in FA. Any off-label use of drugs without a Health Canada indication is done on a case-by-case basis, typically informed by limited and inconclusive research data; this may sometimes be facilitated through special access processes.

Drug Under Review

The key characteristics of omaveloxolone for the treatment of FA are summarized in Table 3.

FA is associated with the inhibition of nuclear factor erythroid 2-related factor 2 (Nrf2), which is involved in cellular response and oxidative stress.²⁰ The suppression of Nrf2 in FA results in oxidative damage, leading to cell death and tissue degradation. Activation of Nrf2 by omaveloxolone has been shown to restore Nrf2 levels, increase Nrf2 activity, rescue mitochondrial dysfunction, and restore redox balance.²⁰ The precise mechanism by which omaveloxolone exerts therapeutic effects in patients with FA is unknown.²⁰

The recommended dose of omaveloxolone is 150 mg (3 capsules of 50 mg each) taken orally once daily. The Health Canada–approved indication is for the treatment of FA in patients aged 16 years and older. The sponsor’s reimbursement request aligns with the Health Canada indication. Omaveloxolone has been approved by regulatory agencies in the US, European Union, and Switzerland. It is currently being reviewed by health technology assessment agencies in England (National Institute for Health and Care Excellence) and Australia (Pharmaceutical Benefits Advisory Committee).

Table 3: Key Characteristics of Omaveloxolone

ALT = alanine aminotransferase; AST = aspartate aminotransferase; BNP = B-type natriuretic peptide; Nrf2 = nuclear factor erythroid 2-related factor 2.

Source: Draft product monograph for Skyclarys.²⁰

Perspectives of Patients, Clinicians, and Drug Programs

The full patient and clinician group submissions received are available in the consolidated patient and clinician group input document for this review on the project website.

Patient Group Input

This section was prepared by the review team based on the input provided by patient groups.

CDA-AMC received 4 patient group submissions: 1 each from MDC, NAF, Ataxia Canada – Claude-St-Jean Foundation, and FARA. MDC supports people affected by muscular dystrophies and related muscle diseases in Canada. Its mission is to enhance the lives of those affected by neuromuscular disorders by providing support through all stages of disease progression so individuals have the appropriate tools to navigate the challenges. Its programs and services help patients and caregivers navigate the system. The organization also shares educational materials and evidence-based information on new treatments, provides financial assistance, and works toward improving the overall well-being and quality of life of patients and their family members. NAF was established specifically to help people with ataxia and their families, with a mission to accelerate the development of treatments by partnering with pharmaceutical companies, researchers, and clinicians in search of new treatments or a cure while providing vital services for families affected by ataxia. Ataxia Canada – Claude-St-Jean Foundation is a charity and community of women, men, adults, teenagers, and children with various forms of ataxia across the country. Its mission is to improve the well-being of people with familial ataxia and support research initiatives. FARA is a national, public, not-for profit organization dedicated to supporting scientific research in search of treatments and a cure for FA. It promotes public awareness and aligns with clinicians, patients, scientists, government agencies, pharmaceutical companies, and other organizations committed to finding a cure.

Neuromuscular service support staff at MDC identified and contacted adults living with FA and parents of children aged 16 years or older with FA to participate in a health care experience survey and semistructured virtual interviews. The survey was shared through e-blasts, personalized invites, and patient online groups in Canada. It gathered insights into diagnosis delays, gaps in treatment, emotional and social effects, and access to care and support systems. MDC received data from a total of 85 individuals (40 males and 45 females) who had received a confirmed diagnosis of FA between the ages of 16 years and 70 years from all provinces in Canada. NAF conducted a survey of its community members with FA in provinces across Canada. A total of 14 responses were received, including from 9 people with FA (living in British Columbia, Manitoba, and Ontario) and 5 caregivers (living in New Brunswick, Ontario, and Quebec). Ataxia Canada gathered the experiences of patients and caregivers in Canada through a combination of interviews and a survey. Of the 85 responders (who represented every Canadian province, including the Northwest Territories), most patients (72%) had experienced disease onset before the age of 15 years, and 50% had been diagnosed between the ages of 8 years and 15 years. Feedback from FARA regarding disease experience was drawn from 2 sources: a white paper that included views from parents of children with FA living in the US, highlighting the importance of pediatric inclusion in clinical trials; and a patient-focused drug development meeting that included 145 patients and caregivers (most of whom resided in the US, with 3% living in Canada) who were asked about their current disease state, experience with different symptoms, and perspectives on future treatments. Views regarding patients’ experiences with omaveloxolone were collected by FARA in January 2021 as part of the FA Community Response Letter, which was shared through the organization’s website, email lists, and social media channels. The letter was signed by 74,070 individuals from 118 countries; of these, 3,157 were from Canada. Written testimony was provided by individuals with varying durations of experience with FA (from very recent diagnoses to diagnoses 15 or more years ago). Respondents included individuals living with FA (n = 1,924); parents of children with FA (n = 688); those who had experience taking omaveloxolone through clinical trials (MOXIe trial, n = 70); and family members who had a loved one with FA who had participated in the MOXIe trials (n = 148).

FA is a recessive neurodegenerative disease caused by biallelic mutations in the FXN gene. It is a progressive, debilitating, life-shortening condition. People with the condition progressively lose ambulation and the ability to accomplish ADLs 2 decades to 3 decades following initial symptom onset. Ataxia Canada and MDC highlighted that patients with FA have significant challenges and/or impairments in coordination and/or balance, mobility and scoliosis, productivity at home and work, independence and social participation, and mental health. Additionally, FARA noted that patients and caregivers report fatigue and neurological symptoms (i.e., trouble with balancing, walking, and coordinating hands and/or arms and having regular falls) as having the most impact on daily quality of life. Currently, there is no approved medication for FA in Canada. NAF, Ataxia Canada, and FARA indicated that people with FA and their caregivers spend hours engaging in symptom-based therapies (such as occupational therapy, speech therapy, and physiotherapy), visiting medical specialists, and obtaining mobility devices, all of which can be challenging, expensive, and time-consuming. Ataxia Canada noted that individuals with FA use anticoagulation treatments for permanent, persistent, or paroxysmal atrial fibrillation as well as muscle relaxants (i.e., baclofen, tizanidine, benzodiazepines, dantrolene sodium, and/or intramuscular botulinum toxin injection) to manage symptoms. Current management strategies do not slow progression; this was noted as a major limitation to their effectiveness. MDC noted that the majority of the patients (n = 61) had no prior experience with treatments for FA; however, many reported using coenzyme Q10 and creatinine to manage symptoms or said they had been prescribed beta blockers to support heart function. Very few respondents (n = 3) had access to targeted treatments for FA through participation in clinical trials. MDC and NAF noted that health disparities for people with FA and their caregivers can arise because some do not have access to specialist clinics, live far away from ataxia centres or movement disorder clinics, and/or may find it difficult to travel as their disease progresses.

Four respondents (1 each from MDC and NAF, 2 from Ataxia Canada) shared their experiences with accessing omaveloxolone through clinical trials. All highlighted that being on the drug had slowed their disease progression (2 of these respondents were already wheelchair-dependent). In addition, 1 respondent highlighted feeling a slight reversal or progression of neurological symptoms such as maintaining balance, writing, swallowing, talking, and stamina. Mild side effects were noted at the beginning of treatment, such as rapid heart rate (infrequently) for 6 weeks and occasional diarrhea for 6 months. The viewpoints gathered by FARA of individuals with FA and parents of individuals who participated in the MOXIe trial indicated that patients who received omaveloxolone through the trial experienced slowing progression, longer retention of motor function (i.e., improved endurance, speech, and ability to walk, stay upright, and eat), and better ability to cope with fatigue. They reported minimal side effects; these included headache, nausea, diarrhea, and transient elevation of liver enzymes and cholesterol.

Currently, individuals require genetic testing to confirm the presence of biallelic mutations of the FXN gene. Ataxia Canada, NAF, and MDC highlighted that the majority of their respondents had undergone genetic testing. MDC and NAF highlighted that the process was easy, especially for people with family members who had been diagnosed with FA. While the process was often emotionally challenging, it provided clarity about the diagnosis. Additionally, MDC noted that some respondents found certain tests, such as nerve conduction studies or shock therapy, to be uncomfortable. Ataxia Canada noted the use of the mFARS scale to evaluate a patient’s status (i.e., assessment of their bulbar function, upper limb coordination, lower limb coordination, and ability to stand and walk). However, the organization highlighted that only a few clinicians may be familiar with using the scale and that it becomes less sensitive to change as patients transfer to wheelchair use. Therefore, use of the scale should be limited in such cases for determining treatment eligibility.

The patient groups agreed that there is a significant need for a treatment that can cure FA or reverse its effects. Patients are seeking a treatment that slows and/or stops disease progression, offers better symptom management, helps them regain and/or preserve their mobility, increases their energy levels, and improves their HRQoL and independence while reducing the physical and emotional burden on families and caregivers and preventing complications like scoliosis and diabetes. MDC noted that patients were concerned about the cost of the treatment (if it would not be covered), which could add to their existing financial burden (i.e., the costs involved in travelling to specialist centres and managing symptoms through various therapies). Furthermore, NAF highlighted those respondents with FA had a moderate to major impact on their and their family members’ quality of life.

Clinician Input

Input From Clinical Experts Consulted for This Review

All CDA-AMC review teams include at least 1 clinical specialist with expertise regarding the diagnosis and management of the condition for which the drug is indicated. Clinical experts are a critical part of the review team and are involved in all phases of the review process (e.g., providing guidance on the development of the review protocol, assisting in the critical appraisal of clinical evidence, interpreting the clinical relevance of the results, and providing guidance on the potential place in therapy). In this review, a total of 3 clinical experts were consulted (2 pediatric specialists and 1 who treats adults). In addition, as part of the review of omaveloxolone, a panel of 4 clinical experts from across Canada (3 pediatric specialists and 1 who treats adults) was convened to characterize patients’ unmet therapeutic needs, assist in identifying and communicating situations in which there are gaps in the evidence that could be addressed through the collection of additional data, promote the early identification of potential implementation challenges, gain further insight into the clinical management of patients living with the condition, and explore the potential place in therapy of the drug (e.g., potential reimbursement conditions). A summary of these discussions follows.

Unmet Needs

The experts emphasized that currently, there are no approved treatments in Canada that can halt, slow, or reverse the underlying neurodegenerative process in FA. The standard of care consists of nonpharmacological interventions (e.g., physiotherapy, occupational therapy, speech therapy, nutritional support, and adaptive devices) and symptomatic management (e.g., through coenzyme Q10 and vitamin E, management of cardiomyopathies, and treatments for spasticity); none of these offer a confirmed disease-modifying effect. These strategies do not alter the disease trajectory or prevent progressive loss of independence. Despite multidisciplinary involvement and comprehensive supportive care, patients and families continue to face the certainty of ongoing neurologic decline.

Consequently, the largest gap in the current therapeutic landscape for FA is the lack of any proven disease-modifying drug. A drug that could provide durable stabilization or slow the progression of the disease would represent a paradigm shift and meet a critical, longstanding unmet need in the management of FA.

Place in Therapy

Omaveloxolone would represent the first therapy in Canada to specifically target the underlying pathophysiology of FA rather than to solely manage its symptoms. This would mark a significant departure from the current treatment landscape, which is limited to supportive and symptomatic measures. As such, the introduction of omaveloxolone would be expected to fill a critical gap. This would position it as a first-line disease-modifying option for patients who meet the indicated age threshold (≥ 16 years) and have a confirmed genetic diagnosis of FA. Patients would not be required to try other treatments first, given that there are no alternatives that modify disease progression; omaveloxolone would likely be offered as soon as a patient is eligible. It would be used in conjunction with existing supportive strategies — not replacing them, but potentially providing the disease-targeted component. This also applies broadly to adults with FA who continue to experience progressive disability.

Patient Population

FA is a rare but genetically definable disorder. Diagnosis involves genetic testing for GAA repeats in the FXN gene and is usually straightforward. While rare ataxic mimics exist, misdiagnosis is uncommon. Patients are typically identified in childhood (often around the age of 11 years and sometime as early as 8 years, according to the pediatric clinical experts); by age 16 years, many have advanced manifestations. Although pediatric age of onset is the most common, a sizable proportion of patients are diagnosed in their twenties, thirties; diagnosis beyond these age groups is rare. Earlier onset tends to predict a more severe and rapidly progressive course, but there is wide heterogeneity. The age of disease onset is the most influential factor in determining the prognosis of patients. Patients who experience adult onset typically show slower progression, fewer cardiac complications, and longer survival. While trial data have focused on patients aged 16 years to 40 years, experts believe that all patients with genetically confirmed FA could, in theory, benefit, with those at the earlier stages (i.e., with lower disease burden) potentially showing more discernible stabilization. However, the experts acknowledge that there is no clinical evidence to substantiate the theoretical benefit in the pediatric population or in patients aged older than 40 years. Thus, the most suitable patients may be those who still retain considerable function (i.e., before there has been irreversible neurologic damage).

The clinical expert specializing in adult patients observed that those with advanced or late-onset disease may meaningfully benefit from slowed progression or preservation of upper limb or bulbar function. Nonambulatory patients should not be excluded; preserving upper limb or bulbar function could still be meaningful and translate into better communication and preserved autonomy that can be measured through some domains of the ADL tool (i.e., eating and personal hygiene). Although the magnitude of measurable benefit might be greatest in those with milder disease severity, all patients with FA technically need disease-modifying interventions.

Assessing the Response to Treatment

According to clinical experts, it is challenging to harmonize how treatment response is measured in clinical trials and in clinical practice. The clinical trial used the mFARS as a primary outcome. While this is a valid research tool, it is time-consuming and not routinely employed in standard clinic visits. Other trial measures (e.g., the 25-foot timed walk test and 9-HPT) are not standard in routine care and can be impractical or risky (e.g., timed walking tests might increase fall risk).

The clinical experts reported that in practice, clinicians rely on a combination of observed clinical stability (e.g., in gait, upper limb coordination, and bulbar function), reports from patients and/or their families, and annual or semiannual neurologic assessments. A clinically meaningful response might simply be a slowing of progression or a stabilization over 12 months rather than the improvement of any specific metric. “Buying time” is meaningful. The experts expressed concerns about relying on patient-reported outcomes alone, given the high unmet need in this patient population. Tools that are objective and standardized, but practical, are preferred.

Furthermore, given that no validated minimal clinically important difference for these scales exists in real-world practice, and that natural progression is heterogeneous, objective definitions of response may be elusive. However, clinical experts believe that a reduction or halting of expected annual declines (e.g., of approximately 2 mFARS points per year) could be considered a meaningful success. For patients who are nonambulatory, changes in ADL scores (e.g., ability to perform personal care independently or speak more clearly) may be more relevant. However, similarly, the experts were not aware of an established minimal important difference for ADL; they suggested that a change of 1 point could be clinically informative.

Discontinuing Treatment

According to clinical experts, the decision to stop therapy should be guided primarily by safety and tolerability rather than by a lack of improvement or disease progression alone. This perspective arises from the recognition that even a partial slowing of progression is valuable, and that discontinuing the only available treatment that could provide some stabilizing effect may be detrimental to the patient’s long-term function. AEs — particularly persistent or severe liver function abnormalities (i.e., ALT and/or AST elevations) — or severe allergic reactions might prompt discontinuation. However, absent severe toxicity, it would be hard to justify stopping therapy in a disease with no alternatives. The experts favoured individualized clinical judgment over rigid stopping rules.

However, while the experts did not necessarily view discontinuation based on efficacy outcomes as a favourable approach, they noted the need to objectively monitor the progress of patients. Lack of response could be determined after 2 years of treatment. Data on disease progression in individuals who discontinued omaveloxolone after prolonged use are not yet published, but they will start to become available, mostly from US-based cohorts. These data will be important in developing future guidance about omaveloxolone discontinuation.

Prescribing Considerations

According to clinical experts, diagnosis and treatment should be guided by specialists experienced in FA, such as neuromuscular or movement disorder neurologists. While no official companion diagnostic beyond genetic testing is necessary, confirmed genetic diagnosis is integral to ensuring appropriate patient selection. This therapy would likely be prescribed and monitored in specialized clinics capable of assessing neurologic function and coordinating care, rather than in general community settings.

Experts underscored the importance of regular liver function monitoring (e.g., due to elevated transaminases observed in the clinical trials). Similarly, the noted increases in LDL cholesterol may require additional monitoring or management (e.g., statins). Patients with cardiac issues (e.g., advanced cardiomyopathy) were largely excluded from the trials; however, the experts do not view such patients as necessarily ineligible. Rather, additional monitoring by a cardiologist may be required.

Requiring intensive or frequent specialized outcome measures for reimbursement (e.g., mFARS assessments) could limit equitable access, favouring patients who live closer to large centres with more resources. Experts acknowledged the need for quantifiable measures to monitor progression, but suggested annual intervals would be sufficient to establish a quantifiable measure of disease progression.

Additional Considerations

The experts noted that the efficacy of omaveloxolone beyond the study duration remains uncertain due to lack of robust evidence on long-term efficacy. In terms of evaluating outcomes important in clinical trials, the experts recognized that the mFARS and other ataxia scales are not routinely used in clinical practice. Frequency of falls, upper limb function, bulbar function, and stabilization of disease are more meaningful in practice; these are at least partially captured in the FARS-ADL scale. This scale tightly correlates with the mFARS, supporting the meaningfulness of mFARS and providing an additional useful standardized assessment tool. Patient quality of life measures, while potentially informative, may be less reliable because of bias and high patient expectations.

Experts highlighted strong patient demand: many individuals with FA may seek immediate access to omaveloxolone, given the historical absence of any disease-modifying option. The experts also noted that a number of therapeutics within the development pipeline may be approved in the coming years, potentially creating opportunities for combination or sequential therapy.

Clinician Group Input

This section was prepared by the review team based on the input provided by clinician groups.

A single submission was received from the Neuromuscular Disease Network for Canada. The network was launched in January 2000, bringing together Canada’s leading clinical, scientific, technical, and patient expertise to improve care, research, and collaboration in neuromuscular disease. Its mission is to improve the treatment of all neuromuscular diseases for all people in Canada. Input from 4 clinicians familiar with clinical trials of omaveloxolone for FA was gathered from 1-to-1 submissions and group discussions.

Treatment strategies for FA remain complex, requiring interdisciplinary management by a team of medical and health professionals. FA management addresses not only motor impairment, but also associated challenges, such as vision and hearing loss, psychological and cognitive issues, cardiomyopathy, diabetes, and skeletal abnormalities. Current treatment focuses on rehabilitation and managing complications. The primary therapeutic goals are to slow disease progression, preserve or enhance function, extend survival, and improve patient well-being.

The clinician group indicated that omaveloxolone is poised to be incorporated into the current treatment paradigm. The experts highlighted that all patients with FA would benefit from an intervention to slow disease progression, but that people with milder symptoms and in earlier stages of FA were most likely to benefit. In addition, the need applies to ambulatory patients, given that treatment may extend their ability to walk, and to nonambulatory patients, given that treatment could help them maintain upper limb function, speech for a longer period. Currently, there is not enough evidence to establish which patients are most likely to respond to omaveloxolone.

Regarding the outcomes used to determine a patient’s response, the clinician group indicated that standardized tests are used for neurologic exams (i.e., the mFARS) and functional assessments (i.e., the FARS-ADL). The use of simple clinician- and patient-related outcome measures like the CGIC and PGIC was also recommended. Measurements every 6 months in the first year and then annually were noted as reasonable and practical. In terms of a clinically meaningful response to treatment, the group noted that there should be an improvement in patient function and well-being (for example, this could be reflected by just a 1-point improvement in upright stability score, indicating better balance in the short-term and predicting delayed loss of ambulation).

The clinician group highlighted that omaveloxolone may be discontinued due to lack of efficacy (to be determined after a year, based on CGIC and PGIC) or side effects (i.e., evidence of organ dysfunction). Transient increases in aminotransferases without bilirubin changes and transient increases in NT-ProBNP without evidence of heart failure do not require discontinuation. Additionally, in the context of increased LDL (a common side effect of omaveloxolone), it was recommended that a statin be prescribed to manage cardiovascular risk factors rather than discontinuing omaveloxolone. Lastly, the clinician group highlighted that people with FA must be treated at specialized centres that offer comprehensive interdisciplinary care, regardless of omaveloxolone treatment. For patients without easy access to such centres, care should be managed by a neurologist knowledgeable about the disease and its management. Equipping providers with the necessary educational tools for prescribing and monitoring the drug will be crucial for ensuring effective treatment.

Drug Program Input

The drug programs provide input on each drug being reviewed through the reimbursement review processes by identifying issues that may affect their ability to implement a recommendation. The implementation questions and corresponding responses from the clinical experts consulted for this review are summarized in Table 4.

Table 4: Summary of Drug Plan Input and Clinical Expert Response

CDEC = Canadian Drug Expert Committee; FA = Friedreich’s ataxia; mFARS = modified Friedreich’s Ataxia Rating Scale.

Clinical Evidence

The objective of this clinical review is to review and critically appraise the clinical evidence submitted by the sponsor on the beneficial and harmful effects of omaveloxolone 150 mg orally once daily for the treatment of FA in adults and adolescents aged 16 years and older. The focus will be on comparing omaveloxolone to relevant comparators and identifying gaps in the current evidence.

A summary of the clinical evidence included by the sponsor in the review of omaveloxolone is presented in 4 sections, with our critical appraisal of the evidence included at the end of each. The first section, the Systematic Review, includes pivotal studies and RCTs that were selected according to the sponsor’s systematic review protocol. The review team’s assessment of the certainty of the evidence in this first section (using the GRADE approach) follows the critical appraisal of the evidence. The second section includes 1 sponsor-submitted, long-term extension study. The third section includes indirect evidence from the sponsor. The fourth section includes additional studies that were considered by the sponsor to address important gaps in the systematic review evidence.

Included Studies

Clinical evidence from the following are included in the review and appraised in this document:

• 1 pivotal study (an RCT) identified through a systematic review

• 1 long-term extension study

• 1 additional study addressing gaps in evidence.

Systematic Review

The contents in this section have been informed by materials submitted by the sponsor. The following information has been summarized and validated by the review team.

Description of Studies

Characteristics of the included studies are summarized in Table 5.

Table 5: Details of Studies Included in the Systematic Review

9-HPT = 9-hole peg test; ALT = alanine aminotransferase; AST = aspartate aminotransferase; CGIC = Clinical Global Impression of Change; CoQ10 = coenzyme Q10; DBL = database lock; eGFR = estimated glomerular filtration rate; FA = Friedreich’s ataxia; FA-ADL = Friedreich’s ataxia – Activities of Daily Living; mFARS = modified Friedreich’s Ataxia Rating Scale; PGIC = Patient Global Impression of Change; SF-36 = Short Form (36) Health Survey; ULN = upper limit of normal.

Source: MOXIe Part 2 trial Clinical Study Report.²⁴ Details are from the sponsor’s Summary of Clinical Evidence.¹¹

The MOXIe trial was conducted in 2 parts:

• Part 1 was a randomized, placebo-controlled, double-blind, dose-ranging study to evaluate the safety, efficacy, pharmacokinetic (PK) activity, and pharmacodynamic (PD) activity of omaveloxolone at various doses (2.5 mg, 5 mg, 10 mg, 20 mg, 40 mg, 80 mg, 160 mg, and 300 mg) in patients with FA. Part 1 enrolled a total of 69 patients with a mean age of 26 years and a relatively equal proportion of female (54%) and male (46%) patients. The study concluded that a 160 mg dose of omaveloxolone was preferable for safety, efficacy, and PKs. Omaveloxolone is available as 2.5 mg, 10 mg, and 50 mg capsules. A dose of 150 mg was selected for further investigation in Part 2 because it was expected to provide a similar pharmacologic response to 160 mg while also reducing the number of capsules patients are required to take each day.^25,26 A detailed summary of Part 1 of the MOXIe trial is not provided for this report because the direct comparative evidence for omaveloxolone 150 mg versus placebo is informed through Part 2.

• Part 2 was a multicentre, randomized, placebo-controlled, double-blind, parallel-group study to evaluate the safety and efficacy of omaveloxolone 150 mg in patients with FA.

The key evidence informing the efficacy and safety of omaveloxolone is based on Part 2 of the MOXIe trial. The MOXIe trial was conducted at 11 sites across 5 countries. The trial consisted of a 48-week treatment phase followed by a 4-week follow-up phase after patients received their last doses; thus, patients’ safety follow-up visits occurred at 52 weeks. Following randomization, patients self-administered omaveloxolone orally once daily for 48 weeks. The 150 mg dose of omaveloxolone was determined from Part 1 of the trial based on reviews of efficacy, safety, PK, and PD data by a Data Safety Monitoring Board (DSMB) and the sponsor. A total of 103 patients were randomized into Part 2 of the MOXIe trial, with 51 patients in the omaveloxolone group and 52 patients in the placebo group. Patients were randomized 1 to 1 through a centralized interactive web response system (IWRS) to receive omaveloxolone 150 mg or placebo. Randomization was stratified by pes cavus status (severe pes cavus versus no severe pes cavus, determined using a flashlight test developed by the sponsor). Patients in the severe pes cavus stratum were not to comprise more than 20% of patients enrolled in the MOXIe Part 2 trial.

The number of patients with severe pes cavus was capped based on observations made during the MOXIe Part 1 trial. This decision was based on the observation that omaveloxolone did not lead to significant improvements in key end points, including mFARS and exercise testing, among patients with severe pes cavus. The lack of improvement was attributed to the requirement of these end points to achieve balance and coordination, which may be difficult for patients who have less surface area contact of their foot with the floor. The clinical diagnosis of pes cavus was defined as a loss of lateral support. Loss of lateral support was determined by having a patient stand barefoot and bear weight. If light from a flashlight could be seen under the patient’s arch, then the patient was diagnosed as having pes cavus, per the study’s definition.²⁷

Patients who were enrolled into Part 1 and Part 2 of the MOXIe trial were eligible to roll into an OLE study to assess the long-term safety and tolerability of omaveloxolone in patients with FA.

Populations

Inclusion and Exclusion Criteria

Key inclusion criteria mandated that participants have genetically confirmed FA, exhibit a moderate level of functional disability (i.e., mFARS score from 20 to 80), and maintain stable exercise regimens. Additionally, eligible patients had to demonstrate adequate cardiac and renal function (i.e., left ventricular ejection fraction of at least 40% and estimated glomerular filtration rate of at least 60 mL/min/1.73 m²). Exclusion criteria included clinically significant cardiac disease (i.e., beyond the mild to moderate cardiomyopathy that is typically associated with FA). Furthermore, patients could be excluded if they were using antioxidant supplements or medications that could interfere with cytochrome P450 enzymes or if they had recently participated in other interventional studies.

Interventions

Patients randomized into the MOXIe Part 2 trial received either omaveloxolone 150 mg or placebo. Omaveloxolone was self-administered by patients and taken orally as 3 capsules of 50 mg once daily in the morning on an empty stomach. Patients tracked their doses using diaries and were instructed to record the date and time of each dose.

Because the study was double-blinded, all patients, investigators, site personnel, and laboratories with direct involvement in the conduct of the study (or their designees) were blinded to treatment assignments. The only individuals with access to treatment assignments for the MOXIe Part 2 trial included individuals who maintained the IWRS or DSMB, members of the unblinded statistical team providing data to and analyzing data for the DSMB, and safety personnel without direct involvement in the conduct of the study who were assigned to report unblinded data to regulatory authorities as required.

A list of permitted and excluded concomitant medications was prespecified. Permitted medications included antibiotics, daily multivitamins, pain medications, other medications for the management of comorbidities, and contraceptives. Excluded medications included other investigational therapies, anticoagulants (except baby Aspirin up to 81 mg), specified medications within 14 days of study day 1 (i.e., antioxidant supplements and antispasticity drugs), and specified medications within 7 days of study day 1 (i.e., herbal preparations, sensitivity substrates for cytochrome P450 2C8 or 3A4, moderate or strong inducers of cytochrome P450 3A4, and substrates for p-glycoprotein transporter). Patients requiring chronic medications were instructed to maintain their doses and dose schedules throughout the MOXIe trial period, as medically feasible. Those requiring intermittent medications were advised to avoid taking them on days when PK samples were collected, when medically feasible.

Outcomes

A list of efficacy end points assessed in this Clinical Review report is provided in Table 6, followed by descriptions of the outcome measures. Summarized end points are based on the outcomes included in the sponsor’s Summary of Clinical Evidence as well as on any outcomes identified as important to this review, according to the clinical experts consulted for this review and input from patient and clinician groups and public drug plans. Using the same considerations, the selected end points were considered those most relevant to inform expert committee deliberations and the list of end points was finalized in consultation with members of the expert committee. All summarized efficacy end points were assessed using GRADE. Select notable harms outcomes considered important for informing expert committee deliberations were also assessed using GRADE. The available clinical trial assessing the efficacy and safety of omaveloxolone had a predefined time point for reported end points at week 48. Consequently, this time point was prioritized as the most clinically relevant. Outcomes that were determined to be the most clinically relevant and had potential applicability to clinical practice were prioritized. Prioritization of outcomes was based on input from clinicians, patients, sponsors, and health economics experts. Key secondary outcomes in the trial (PGIC at week 48 and CGIC at week 48) are presented in Appendix 1, but are not part of the outcomes that have undergone the GRADE process.

Table 6: Outcomes Summarized From the Studies Included in the Systematic Review

9-HPT = 9-hole peg test; FA-ADL = Friedreich’s ataxia – Activities of Daily Living; mFARS = modified Friedreich’s Ataxia Rating Scale.

^aStatistical testing for these end points was adjusted for multiple comparisons (e.g., hierarchal testing).

Source: MOXIe Part 2 trial Clinical Study Report.²⁴ Details are from the sponsor’s Summary of Clinical Evidence.¹¹

Friedreich’s Ataxia Rating Scale

The mFARS includes 4 domains from the Friedreich’s Ataxia Rating Scale (FARS) examination (i.e., bulbar function, upper limb coordination, lower limb coordination, and upright stability),but excludes the neurologic section. The mFARS is calculated as the full neurologic FARS assessment minus section D (peripheral nervous system). A higher score on the FARS or mFARS examination signifies more severe physical impairment, and a reduction in FARS or mFARS score signifies improvement in functioning. The change in mFARS score from baseline (relative to placebo) was the primary end point of the MOXIe Part 2 trial. Previous studies assessing the natural history of FA suggest that upright stability is a key measure correlated to the progression of patient’s disease and ultimate loss of ambulation, while upper limb coordination is associated with progression among nonambulatory patients.^14,28

9-Hole Peg Test

The 9-HPT is a brief, standardized, quantitative test of upper-extremity (arm and hand) function. Both the dominant and nondominant hands were tested twice (with 2 consecutive trials of the dominant hand followed immediately by 2 consecutive trials of the nondominant hand). Longer test times reflect more impairment in the patient’s upper-extremity function; thus, a decrease from baseline suggests improvement.

25-Foot Timed Walk Test

The 25-foot timed walk test is a quantitative mobility and leg function performance test based on the time (in seconds) needed to complete a 25-foot walk. Longer test times reflect more impairment in the patient’s ability to walk; decreases in time from baseline suggest improvement. If a patient used an assistive device for the 25-foot timed walk test on day 1, the same assistive device was used for all 25-foot walk assessments.

Falls Diary

Throughout Part 2 of the study, patients were instructed to record any instances of falls in a study diary. A fall was defined as “the patient unintentionally coming to rest on the ground or at a lower level.” Items tracked in the falls diary included, but were not limited to, the date and time of each fall, the location of the fall, the activity that preceded the fall, the perceived cause of the fall, and whether an injury was sustained during the fall. Patients were provided the fall diary at screening and were to record fall incidents from screening to week 48.

Activities of Daily Living

Patients were instructed to answer 9 questions on the ADL survey to assess 9 concepts: speech, swallowing, cutting food and handling utensils, dressing, personal hygiene, falling, walking, quality of sitting position, and bladder function. The total score can range from 0 (no ADL impairments) to 36 (most severe ADL impairments).

Harms

AEs, SAEs, withdrawal due to AEs, deaths, and AEs of special interest were reported.

Table 7: Summary of Outcome Measures and Their Measurement Properties

9-HPT = 9-hole peg test; ADL = activities of daily living; CI = confidence interval; FA = Friedreich’s ataxia; FA-ADL = Friedreich’s ataxia – Activities of Daily Living; FARS = Friedreich’s Ataxia Rating Scale; FARS-ADL = Friedreich’s Ataxia Rating Scale – Activities of Daily Living; ICC = intraclass correlation coefficient; MCID = minimal clinically important difference; MID = minimal important difference; mFARS = modified Friedreich’s Ataxia Rating Scale; PGIC = Patient Global Impression of Change; PROM = patient-reported outcome measure; SARA = Scale for the Assessment and Rating of Ataxia; SCA3 = spinocerebellar ataxia type 3; RFC1 = replication factor C subunit 1.

Statistical Analysis

Sample-Size and Power Calculation

Patients Without Severe Pes Cavus

The protocol sample-size calculations were based on 100 patients (i.e., the total planned enrolment), 80 of whom would be without severe pes cavus. A mixed model with repeated measures based on the 80 patients without severe pes cavus had approximately 85% power to detect a 2.0-point difference between omaveloxolone and placebo in change from baseline in mFARS at week 48 under the following model specifications:

• ████████ ████████████ ██████ ██ ███ ███ ███ ███ ███ ███ ██████ ████████ ████████ ██████████ █████████ ████ ███████████ █ ███████ ████████████ ██ ███ ████ ███████ ██ ███ Two-sided type I error rate of 0.05

• a difference of 2.0 points between the change in mFARS in the omaveloxolone and placebo groups

• an SD of change in mFARS of 3.5 points.

Patients With Severe Pes Cavus

███ ███████ ████ ██████ ███ █████ ████ ██ ████████ ███ █████████████ ███ █████ ██ ██████ █ ██████████ ██ ███ ██████ ██ ██████ ████ ████████ ██ ██████ █████ ███ ███████████ ███ ███ ███████ ██████ ███ █████ █████ ██████ ███████ ███████ ███████ ███ █████ ██ ████ ███████ ███ ████████ ██ ██ ███ ███ ██ ████████ ██████ █ ██████████ ███████ ███████████ ███ ████ ████ ████ ███████ ████ ██████████ ██████████████ ██ ███ █████ █████ ████ ██ ███ ████████ █████████ ██ ██████ ███ █████ ███ ███████ ██ ██████ █ ████ ██ ███████ ████████ ████ ██ ███████ ███████ ███ ██████████ ██ ██████ █ ███████ █████ ████████ ███ ████ ███████ ██ █████ ████ █ ██████████ █████ ██ ████ █████ ███ █████████ █████ ████ ███ ███████ ███ █████████ ██ ██████ ███ █████ ███ ███ ███████ ██████████.²⁷

Statistical Testing

The primary objective of the trial was to assess change in mFARS score at week 48. Key secondary and secondary end points were tested hierarchically to maintain the family-wise overall type I error rate of 0.05 upon observation of statistical significance for the primary end point. Formal testing of the hierarchy continued as long as statistical significance was observed for each tested end point. If end points did not demonstrate statistical significance, formal testing of the subsequent end points was not to occur. Key secondary and secondary end points were tested in the following order:

• PGIC at week 48 (key secondary)

• CGIC at week 48 (key secondary)

• change in 9-HPT at week 48

• change in 25-foot timed walk test at week 48

• frequency of falls over 48 weeks

• change in peak work during maximal exercise testing at week 48

• change in ADL score at week 48.

All other end points were exploratory and presented with nominal significance levels.

Post Hoc Analysis

It was observed in the study that, while the distribution of many baseline characteristics was similar between treatment groups, some imbalances occurred despite randomization. The cohort receiving omaveloxolone had characteristics consistent with slightly more advanced disease than the cohort receiving placebo: higher mFARS scores (a mean score of 40.9 points versus 38.8 points), longer GAA1 repeat lengths (a mean of 739.2 versus 693.8), and, most importantly, a greater proportion of patients with a history of cardiomyopathy (48% versus 29%). To assess the impacts of these imbalances on the primary end point, mFARS pos-hoc analyses included MMRM evaluations with covariates, analyses of covariance at week 48 with and without imputation, and tipping point analysis. Correlations with ALT and AST were explored for the FAS and all-randomized populations. Additionally, ADL post hoc analyses included distributions of total scores and changes from baseline to weeks 24, 36, and 48 for FAS and all-randomized populations.

Subgroup Analyses

Subgroup analyses were conducted for the primary efficacy end point: change in mFARS score at week 48. The analyses were conducted among subgroups that had a sufficient number of patients to warrant these analyses:

• age: younger than 18 years, 18 years or older

• sex: female, male

• geographic location: US, other

• ethnicity: non-Hispanic or Latino, Hispanic or Latino

• race: white, nonwhite [wording from original source]

• GAA1 repeat length greater than or equal to 675: yes, no.

Subgroup analyses of geographic location, ethnicity, and race were not conducted because the subgroups were not sufficiently large. An additional subgroup analysis of ambulatory status (i.e., ambulatory or nonambulatory) was conducted post hoc. Subgroup analyses were not adjusted for multiplicity.

Table 8: Statistical Analysis of Efficacy End Points

9-HPT = 9-hole peg test; MAR = missing at random; MMRM = mixed model for repeated measures; mFARS = modified Friedreich’s Ataxia Rating Scale; PP = per protocol.

Source: MOXIe Part 2 trial Clinical Study Report.²⁴ Details are from the sponsor’s Summary of Clinical Evidence.¹¹

Analysis Populations

A summary of the analysis populations used in the MOXIe Part 2 trial are included in Table 9. The primary efficacy analyses were conducted using the FAS, while safety analyses were conducted using the safety population. Other analyses sets were used for exploratory and sensitivity analyses.

Table 9: Analysis Populations in the MOXIe Part 2 Trial

^aThis analysis set included patients with and without severe pes cavus.

Source: MOXIe Part 2 trial Clinical Study Report.²⁴ Details are from the sponsor’s Summary of Clinical Evidence.¹¹

Results

Patient Disposition

A summary of patient disposition in the MOXIe Part 2 trial is provided in Table 10. A total of 156 patients were screened for eligibility; of these, 103 patients were randomized, including 51 patients in the group receiving omaveloxolone and 52 patients in the group receiving placebo. Of the 103 patients, 20 patients had pes cavus (n = 10 in each group).

Among all patients, 94 patients (91%) completed treatment through week 48 of the study, including 44 patients out of 51 patients (86.3%) randomized to the group receiving omaveloxolone and 50 patients out of 52 patients (96.2%) randomized to the group receiving placebo. Seven patients (13.7%) in the group receiving omaveloxolone and 2 patients (3.8%) in the group receiving placebo discontinued study treatment due to AEs (7.8% versus 3.8%, respectively) and withdrawal by patient (5.9% versus 0%, respectively). Most patients (93%) completed the study follow-up through week 52.

The majority of protocol deviations occurring during Part 2 of the MOXIe trial were minor. A total of 22 patients in the group receiving omaveloxolone and 19 patients in the group receiving placebo had at least 1 major protocol deviation. The most common reason for a major protocol deviation was deviation from the study procedure (n = 13 in the group receiving omaveloxolone and n = 7 in the group receiving placebo).³⁴ Other protocol deviations included those related to the informed consent form (n = 5 versus n = 8 in the omaveloxolone and placebo groups, respectively), eligibility criteria (n = 3 versus n = 7), use of excluded concomitant medications (n = 3 versus n = 2), drug dispensing (n = 1 versus n = 0), visit window (n = 1 versus n = 0), and other (n = 1 versus n = 0).³⁴

Table 10: Summary of Patient Disposition From Studies Included in the Systematic Review

FAS = full analysis set; NA = not applicable; PP = per protocol.

Source: MOXIe trial Part 2 Clinical Study Report.²⁴ Details are from the sponsor’s Summary of Clinical Evidence.¹¹

Baseline Characteristics

A summary of baseline characteristics relevant to the review is provided in Table 11; baseline characteristics were summarized for the FAS and all-randomized population. In general, baseline characteristics in the FAS were similar between the groups receiving omaveloxolone versus placebo in the MOXIe Part 2 trial. Patients had a mean age of approximately 24 years and were mostly white (98%). Patients had a mean weight of 67 kg and a mean body mass index of ██ kg/m². Patients had a mean peak work of 1.13 watt/kg, ADL of 10, age of onset of 15.5 years, were approximately 4.8 years since onset, and had a mean GAA1 repeat length of 717 ████ ████ █████ ████ ██ ████████ (██%), with a GAA1 repeat length greater than or equal to 675. The majority of patients (93%) had ambulatory status, ████████ ██████ ████ █████ ██ ███ ████████ ████████ ████ ████ ██ █████████ ██████. Most patients also had a history of █████████ ███%) and a history of scoliosis (74%). Approximately ██% of patients had a history of sensory neuropathy; ██% of patients had a history of cardiomyopathy; 30% had a history of swallowing difficulties; and 23% had a history of scoliosis surgery.

Although randomization was conducted using IWRS, there were some imbalances in patient baseline characteristics. While the mean age in each treatment group was similar, the proportions of patients aged younger than 18 years versus 18 years or older differed; a greater proportion of patients in the group receiving placebo versus omaveloxolone were younger than 18 years of age (31% versus 17.5%, respectively), and a greater proportion of patients in the group receiving omaveloxolone versus placebo were aged 18 years or older (82.5% versus 69.0%, respectively). There was also a greater proportion of females in the group receiving omaveloxolone versus placebo (60.0% versus 33.35, respectively). The group receiving omaveloxolone also had a higher proportion of patients with a GAA1 repeat length greater than or equal to 675 (████% versus ████%, respectively), a history of cardiomyopathy (47.5% versus 28.6%), and a history of scoliosis surgery (30.0% versus 16.7%). Despite the relatively balanced reporting of use of assistive devices across both treatment groups, there was a higher proportion of patients in the group receiving omaveloxolone who required assistance with a walker (█████ versus ████%). A slightly higher proportion of patients in the group receiving placebo had a history of areflexia (████% versus ████%, respectively).

In the FAS and all-randomized population, the majority of patients (> 90%) had a relevant medical history event recorded at baseline. In the all-randomized population, the most common medical history preferred terms (occurring in > 25% of patients in the omaveloxolone group) (in the omaveloxolone group and placebo group, respectively) were █████████ ██████ ███ ███████ ██████████ ██████ ███ ███████ ███████ ██████ ███ ███████ ███ ████████ ██████ ███ ███████ Overall, the high proportion of comorbidities observed at baseline in both treatment groups, but particularly in the omaveloxolone group, highlights the considerable morbidity experienced by patients with FA.

Table 11: Summary of Baseline Characteristics From Studies Included in the Systematic Review

ADL = activities of daily living; BMI = body mass index; FA = Friedreich's ataxia; FAS = full analysis set; GAA1 = guanosine-adenine-adenine 1; Max = maximum; mFARS = modified Friedreich’s Ataxia Rating Scale; Min = minimum; SD = standard deviation.

Source: MOXIe Part 2 trial Clinical Study Report.²⁴ Details are from the sponsor’s Summary of Clinical Evidence.¹¹

Exposure to Study Treatments

Study Treatments

█████ ██ ██████████ ███ ████████ ███ ██████████ ██ █████ █████████ ██████ ███ █████ █████ ████ ██ ████ ███ ███████ █████████ ███████ █ ██████████ ████ ██ ██████ ██ ██ █████████████ ███ █████████ ██ █ ████ ██ █████ █████ ██████ █ ████ ██ █████ █████ ██ ███ ███████ ██████ █ ████ ██ █████ █████ ████ ████████ ██ ████████ ████ █ ████ ██ █████ █████ ████████ █████ ███ ████ ██████████ ██ ████ █████████ ██████ █████ ████ █████ ██ ████████ ██ ███ █████████████ █████ ███ █████ ██ ████████ ██ ███ ███████ █████ ██████ ████ ███████████ ███ █████ ██████ ██ █████ █████████ ███ ███████ ██████ ████ █████████ ██████ ████ █ ████ ██ ███ ██ ███ █████████████ █████ ███ █████ ██ ███ ███████ ██████ ████ ████ ██ █████ ████ ████████ ██ ██████ █████████ █████ ██████ ███ ███████ █████████ ███████

Concomitant Medications and Cointerventions

███ ████████ ██ ████████ ██████ ██ ███ ████ ███ █████ ███████████ ███ ███ ██████ ██████████ ███ ██ █████ █ ███████████ ███████████ ██ ███ ██████ ███████████ ███ ████ ██████ ██████ ██ ████ ██ ████████ ██ ███ █████████████ ██████ ███████████ ███████████ ████████ ███ ███ █████████████ █████ ███ ███████ ██████ █████████████ ██████ ███ █████ ████ █████████ ██████ ███ ██████ ███ ███████████ ██████ ███ ███████

Table 12: Summary of Patient Exposure From Studies Included in the Systematic Review

SD = standard deviation.

Source: MOXIe Part 2 trial Clinical Study Report.²⁴ Details are from the sponsor’s Summary of Clinical Evidence.¹¹

Efficacy

Change in mFARS Score at Week 48

In the primary analysis population (FAS), treatment with omaveloxolone improved mFARS score relative to placebo at week 48. Patients receiving omaveloxolone had a mean change from baseline of █████ ██████ ███ █████), while those receiving placebo had a mean change of █████ ██████ ███ █████). The MMRM estimate of mean difference in change from baseline between the groups receiving omaveloxolone and placebo was −2.40 points (95% CI, −4.31 to −0.50; P = 0.0141) favouring omaveloxolone. This was the primary outcome of the study and met the threshold for statistical significance.

In the all-randomized population, which included patients with severe pes cavus, omaveloxolone treatment improved mFARS scores by an estimated mean difference of −1.93 points relative to placebo (95% CI, −3.70 to −0.15; P = 0.0342).

Additional prespecified and post hoc analyses supported the primary analysis, demonstrating consistent benefits of omaveloxolone over placebo.

Change in Performance on 9-HPT

In the FAS population, patients receiving omaveloxolone had a mean change from baseline ██ ███████ ███ █ ███████), while those receiving placebo had a mean change ██ ███████ ███ █ ███████). The MMRM estimated mean difference in change from baseline between omaveloxolone and placebo was ██████ ████████████ ███ ███████ ██ ███████ | | ██████).

In the all-randomized population, which included additional patients, omaveloxolone treatment resulted in an estimated mean difference of ██████ 1 per second relative to placebo (95% CI, ███████ ██ ███████ | | ██████), indicating no significant difference between the groups.

Change in Performance on a 25-Foot Timed Walk Test

In the FAS population, omaveloxolone did not demonstrate a significant improvement in performance on the 25-foot timed walk test compared to placebo at week 48. Patients receiving omaveloxolone had a mean change from baseline in the reciprocal of average walk time of –██████ per second (SD ███████), whereas those receiving placebo had a mean change of –██████ per second (SD ███████). The MMRM estimated mean difference between omaveloxolone and placebo was 0.0058 per second (95% CI, ███████ ██ ██████; P = 0.4635).

In the all-randomized population, omaveloxolone treatment resulted in an estimated mean difference of ██████ per second relative to placebo (95% CI, ███████ ██ ███████ | | ██████), a nonsignificant difference between the treatment groups.

Frequency of Falls at Week 48

In the FAS population, treatment with omaveloxolone did not significantly reduce the frequency of falls compared to placebo at week 48. Patients receiving omaveloxolone reported a mean of ████ falls on treatment (SD █████), while those receiving placebo reported a mean of ████ falls (SD █████). The Poisson estimated difference in incidence rate of falls between omaveloxolone and placebo was █████ ████ ███ █████ ██ █████ | | ██████).

In the all-randomized population, omaveloxolone treatment led to an incidence rate difference of █████ compared to placebo (95% CI, █████ ██ █████ | | ██████), a nonsignificant difference in the frequency of falls between the groups.

ADLs at Week 48

In the FAS population, treatment with omaveloxolone improved ADL scores relative to placebo at week 48. Patients receiving omaveloxolone had a mean change from baseline of ██████ points (SD ██████), while those receiving placebo had a mean increase of █████ points (SD ██████). The MMRM estimated mean difference in change from baseline between omaveloxolone and placebo was –1.30 points (95% CI, █████ ██ █████; P = 0.0420) favouring omaveloxolone.

In the all-randomized population, which included patients with severe pes cavus, omaveloxolone treatment improved ADL scores by an estimated mean difference of █████ points relative to placebo (95% CI, –████ ██ █████ | | ██████).

Table 13: Summary of Key Efficacy Results From Studies Included in the Systematic Review

9-HPT = 9-hole peg test; ADL = activities of daily living; CI = confidence interval; FAS = full analysis set; mFARS = modified Friedreich’s Ataxia Rating Scale; MMRM = mixed model for repeated measures; SD = standard deviation.

^aAnalysis based on reciprocal of average time, dominant hand.

^bMean changes and P values estimated from MMRM analyses for 9-HPT, 25-foot timed walk test, peak work, and ADLs.

^cAnalysis based on reciprocal of average walk time.

^dComparison in the frequency of falls for the omaveloxolone versus placebo groups was estimated from the Poisson model, with the natural logarithm of time on study (days) included as an offset term.

^eTotal ADL score.

Source: MOXIe Part 2 trial Clinical Study Report.²⁴

Harms

Refer to Table 14 for harms data.

Adverse Events

Overall, all patients in the trial experienced at least 1 AE. The following AEs were reported in a greater proportion of patients in the omaveloxolone group than in the placebo group: nausea (33.3% versus 13.5%), abdominal pain (21.6% versus 5.8%), diarrhea (19.6% versus 9.6%), fatigue (21.6% versus 13.5%), ALT increase (37.3% versus 1.9%), AST increase (21.6% versus 1.9%), muscle spasm (15.7% versus 5.8%), back pain (13.7% versus 7.7%), and oropharyngeal pain (17.6% versus 5.8%). The following AE was reported in a greater proportion of patients in the placebo group than the omaveloxolone group: ligament sprain (15.4% versus 9.8%).

Serious Adverse Events

In the all safety population, SAEs were reported in 5 patients (9.8%) receiving omaveloxolone and in 3 patients (5.8%) receiving placebo. In the subgroup without pes cavus, SAEs occurred in | ████████ █████%) in the omaveloxolone group compared to | ████████ ████%) in the placebo group.

SAEs in the omaveloxolone group included anemia, various cardiac events (atrial fibrillation, palpitations, sinus tachycardia, ventricular tachycardia, and noncardiac chest pain), laryngitis, viral upper respiratory tract infection, and craniocerebral injury. In the placebo group, SAEs consisted of atrial fibrillation, gallbladder disorder, and ankle fracture.

Withdrawals Due to AEs

In the all safety population, treatment discontinuations due to AEs were more frequent in the group receiving omaveloxolone, with 4 patients (7.8%) in that group discontinuing treatment compared to 2 patients (3.8%) in the group receiving placebo. In the subgroup without pes cavus | | ████████ ████%) in the omaveloxolone group and | ████████ ████%) in the placebo group discontinued treatment due to AEs.

Mortality

No deaths were reported in either treatment group during the study.

Notable Harms

In the all safety population, elevations in liver enzymes were AEs of special interest associated with omaveloxolone treatment. Increases in ALT were reported in 19 patients (37.3%) receiving omaveloxolone compared to 1 patient (1.9%) on placebo. Similarly, AST elevations occurred in 11 patients (21.6%) in the omaveloxolone group versus 1 patient (1.9%) in the placebo group. In the subgroup without pes cavus, ALT increases were observed in ██ ████████ █████%) treated with omaveloxolone and | ███████ ████%) on placebo, while AST increases were reported in | ████████ █████%) receiving omaveloxolone compared to | ███████ ██████ on placebo. These findings indicate a higher incidence of liver enzyme elevations in patients treated with omaveloxolone compared to placebo.

Table 14: Summary of Harms Results From Studies Included in the Systematic Review

AE = adverse event; ALT = alanine aminotransferase; AST = aspartate aminotransferase; SAE = serious adverse event; NR = not reported.

Note: Multiple events were counted once for each patient for each summary. AEs were coded using MedDRA Version 14.1.

^aProportion greater than or equal to 10% in either treatment group in safety population.

Source: MOXIe Part 2 trial Clinical Study Report.²⁴ Details are from the sponsor’s Summary of Clinical Evidence.¹¹

Critical Appraisal

Internal Validity

The MOXIe Part 2 trial is a randomized, double-blind, placebo-controlled study appraised by the CDA-AMC review team as pivotal evidence to assess the use of omaveloxolone to treat FA exhibited several features that enhance its internal validity. The trial had an appropriate randomization design with clear inclusion and exclusion criteria; enrolled patients had genetic confirmation of FA. Central allocation through IWRS, coupled with a double-blind design and a placebo that matched the active treatment closely, ensured that patients, investigators, site personnel, and laboratory staff remained unaware of treatment assignments. It is possible that liver enzyme elevations in the group receiving omaveloxolone could have suggested treatment status, potentially leading to partial unblinding.

There were several imbalances in baseline characteristics across treatment groups in the trial. Specifically, the group receiving omaveloxolone had a higher proportion of males, a greater number of patients with a history of cardiomyopathy, higher baseline mFARS scores, and more individuals with GAA1 repeat lengths greater than or equal to 675. The imbalances in mFARS score, GAA1 repeat lengths, and cardiomyopathy history may suggest that patients in the group receiving omaveloxolone are at a more advanced stage of disease, potentially biasing efficacy against omaveloxolone. However, it is unclear if the higher proportion of males may create bias in favour of or against omaveloxolone.

Additional post hoc sensitivity analyses in the FAS population, performed to assess the impact of controlling for imbalances between the randomized cohorts in baseline disease characteristics (i.e., history of cardiomyopathy, GAA1 repeat length, and history of cardiomyopathy and GAA1 repeat length combined), showed a greater treatment effect than did the prespecified primary analysis. In addition, the prespecified subgroup analysis of mFARS scores based on sex suggests that male patients benefited more than female patients. This suggests that the imbalance in sex may bias the overall results against omaveloxolone.

Dropout rates were low overall, with 91% of participants completing the trial; reasons for discontinuation were clearly reported. However, discontinuations were more common in the group receiving omaveloxolone, generally due to AEs. This raises the possibility that missing data were not completely at random and that the MAR introduced bias in the results.

The use of mFARS as a primary end point was supported by its internal validity and reliability in measuring the physical performance of patients with FA; however, no minimal clinically important difference was established, nor is the measure routinely applied in clinical practice. The statistical analysis plan was prespecified and appropriate, employing MMRM methods and hierarchical testing procedures to control for multiple comparisons. Multiple sensitivity analyses supported the findings. The sample size was adequate to provide sufficient power for the primary outcome, based on the power calculation. However, it is unclear if that is the case for any outcomes beyond the primary outcome. The internal validity of secondary outcomes is further compromised by falling outside the statistical testing hierarchy. As such, all P values are considered nominal in nature.

External Validity

The clinical experts consulted during this review suggested that, aside from the exclusion of pediatric patients, the eligibility criteria of the MOXIe Part 2 trial were generally comparable to those applied to patients in Canada. However, the trial’s strict inclusion and exclusion criteria, including constraints around cardiac involvement, may have led to a cohort that was healthier than what is typically encountered in routine Canadian practice. Furthermore, and according to the pediatric clinical experts, most patients are diagnosed at an earlier age than 16 years (typically around 11 years). However, the clinical expert specializing in adult patients with FA expressed that the mean age of onset in the trial is representative of the population of patients in Canada. The exclusion of the pediatric population means there is no evidence of the safety and efficacy of omaveloxolone in that patient population. However, the approved indication is for patients aged 16 years and older. Finally, the eligibility criteria limited the patient population to those with mFARS scores ranging from 20 to 80. Therefore, considering the primary outcome, the generalizability of the results is limited to patients whose scores fall within this range.

The clinical experts noted that baseline characteristics were reflective of the patient population aged 16 years of age and older. However, the pediatric clinical experts also communicated that patients are typically diagnosed around the age of 11 years, which is lower than the mean age of onset in the MOXIe Part 2 trial (i.e., 15.5 years) and may suggest that patients in Canada could be at a more advanced stage than patients in the trial. On the other hand, the clinical expert specializing in adult patients expressed that the age of onset is representative of patients in Canada. Given these differing perspectives, the effect of uncertainty regarding the comparability of age of onset between trial participants and patients in Canada remains undetermined. Furthermore, the study did not include any sites in Canada, reducing the generalizability and applicability of the results to the Canadian practice. However, the clinical experts did not feel that the lack of Canadian sites is a significant limitation. This is due to the similarity of the patient population and treatment paradigm with other countries that had representing sites.

The study capped the number of patients with severe pes cavus in the study and excluded these patients from the FAS. The sponsor justified this decision on evidence from the MOXIe Part 1 trial that suggested patients with severe pes cavus may represent a different subtype of FA that likely interferes with the ability to perform assessments that require standing or pedalling. Clinical experts involved in this review suggested that all patients with FA have a certain level of pes cavus, which has not been suggested as clinically prognostic of FA prognosis. Results from the all-randomized population suggest that the inclusion of patients with severe pes cavus may lead to a decrease in the observed benefit of omaveloxolone compared to placebo; however, this may be a limitation of the mFARS as a measure of FA progression.

While the duration of the study is reasonable, at 48 weeks, the clinical experts noted that due to the heterogeneity of the disease course, a longer duration with a comparative efficacy outcome assessed at 2 years would have been appropriate. While the sponsor provided OLE studies, several limitations related to the design of OLE studies (outlined in the relevant section) prevent us from understanding the comparative efficacy of omaveloxolone over a period longer than 48 weeks.

The outcomes used in the MOXIe Part 2 trial were appropriate for a clinical study setting, but are not commonly implemented in clinical practice. According to the clinical experts, none of the outcomes implemented in the study are used in practice. Specifically, the primary outcome, mFARS, is not used in clinical practice. This may represent challenges in clinical implementation and assessment of treatment response, and further limits the interpretation of the study to current practice. Limitation of the generalizability of outcomes is reinforced by the lack of a clear and formal minimal important difference. However, the clinical experts supported the use of a 2-point threshold as a meaningful change in the mFARS score, based on estimates from 1 natural history study.

GRADE Summary of Findings and Certainty of the Evidence

Methods for Assessing the Certainty of the Evidence

For the pivotal studies and RCTs identified in the sponsor’s systematic review, GRADE was used to assess the certainty of the evidence for the outcomes considered most relevant to inform expert committee deliberations, and a final certainty rating was determined as outlined by the GRADE Working Group:^35,36

• High certainty: We are very confident that the true effect lies close to that of the estimate of the effect.

• Moderate certainty: We are moderately confident in the effect estimate. The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. We use the word “likely” for evidence of moderate certainty (e.g., “X intervention likely results in Y outcome”).

• Low certainty: Our confidence in the effect estimate is limited. The true effect may be substantially different from the estimate of the effect. We use the word “may” for evidence of low certainty (e.g., “X intervention may result in Y outcome”).

• Very low certainty: We have very little confidence in the effect estimate. The true effect is likely to be substantially different from the estimate of the effect. We describe evidence of very low certainty as “very uncertain.”

For RCTs: Following the GRADE approach, evidence from RCTs started as high-certainty evidence and could be rated down for concerns related to study limitations (which refers to internal validity or risk of bias), inconsistency across studies, indirectness, imprecision of effects, or publication bias.

When possible, certainty was rated in the context of the presence of an important (nontrivial) treatment effect; if this was not possible, certainty was rated in the context of the presence of any treatment effect (i.e., the clinical importance is unclear). In all cases, the target of the certainty of evidence assessment was based on the point estimate and its location relative to the threshold for a clinically important effect (when a threshold was available) or to the null. In the case of the outcomes presented in this report, a clinically important effect of 2 points was determined for mFARS, while the target of certainty was assessed based on the location of the point estimate relative to the null.

Results of GRADE Assessments

Table 2 presents the GRADE summary of findings for omaveloxolone versus placebo.

Long-Term Extension Study

Contents within this section have been informed by materials submitted by the sponsor. The following has been summarized and validated by the review team.

Description of Study

One OLE study (the Study 1402 extension) was included to assess the long-term safety and tolerability of omaveloxolone in patients with FA following the completion of Part 1 or Part 2 of the MOXIe trial. The OLE is ongoing; therefore, the preliminary results presented reflect an interim analysis based on the data cut-off date of March 24, 2022.

Part 1 of the MOXIe trial was a randomized, placebo-controlled, double-blind, dose-ranging study to evaluate the safety, efficacy, PK, and PD activity of omaveloxolone at various doses (2.5 mg, 5 mg, 10 mg, 20 mg, 40 mg, 80 mg, 160 mg, and 300 mg) in patients with FA. It included 69 patients, of whom 54% were female and 46% were male; the mean age was 26 years. The study concluded that a 160 mg dose of omaveloxolone was preferred for safety, efficacy, and PKs.

In the OLE, patients received omaveloxolone (150 mg) once daily until the drug became available through commercial channels or until patient withdrawal, whichever occurred first. Day 1 of the OLE was defined as the first day on which treatment was dispensed to a patient following completion of Part 1 or Part 2 of the MOXIe trial. All visits were enumerated with respect to this definition of day 1. Patients were to be scheduled for in-person assessments during treatment in the OLE on day 1, week 4, week 24, and every 24 weeks thereafter. Patients were also assessed by telephone contact on day 14. The end-of-study visit was scheduled 30 days after the last dose. The OLE was under way during the COVID-19 pandemic. Risk mitigation strategies were implemented to protect the health of patients while maintaining compliance with good clinical practice.

Given that patients in Part 2 of the MOXIe trial were followed continuously, their eligibility for the extension was assessed at the week 52 visit, and a separate screening visit was not required. For patients in Part 1, a screening visit to confirm eligibility for the extension was completed.

Populations

The inclusion criteria for the OLE consisted of previous enrolment in and completion of either the MOXIe Part 1 or Part 2 trials without any major protocol deviations. The exclusion criteria specified that patients should not have any uncontrolled comorbidities or clinically significant abnormality. Patients who discontinued treatment early in Part 1 or Part 2 of the MOXIe trial were not eligible to enrol in the OLE.

Patients were discontinued from treatment with omaveloxolone if they reported elevated transaminases (i.e., ALT or AST > 8 × the upper limit of normal [ULN]; ALT or AST > 5 × ULN for more than 2 weeks; ALT or AST > 3 × ULN and total bilirubin > 2 × ULN; or ALT or AST > 3 × ULN), with the appearance of fatigue, nausea, vomiting, right upper quadrant pain or tenderness, fever, rash, and/or eosinophilia (> 5%).

Interventions

All patients enrolled in the extension were assigned to receive oral omaveloxolone once daily at a dose of 150 mg (3 capsules of 50 mg each). Patients were treated with omaveloxolone until the drug became available through commercial channels or until patient withdrawal, whichever was sooner.

Patients were permitted to receive the following concomitant medications:

• antioxidant supplements

• antispasticity drugs.

• Patients taking medication chronically were to be maintained on those same doses and dose schedules throughout the study period, as medically feasible.

Medications not permitted during the OLE included:

• sensitive substrates for cytochrome P450 2C8 or 3A4 (e.g., repaglinide, midazolam, sildenafil)

• moderate or strong inhibitors or inducers of cytochrome P450 3A4 (e.g., carbamazepine, phenytoin, ciprofloxacin, grapefruit juice)

• substrates for p-glycoprotein transporter (e.g., ambrisentan, digoxin).

Outcomes

The primary objective of the OLE was to assess safety through AEs, SAEs, use of concomitant medications, laboratory test results, vital sign measurements, and weight. Assessment of long-term efficacy was a secondary objective. Efficacy outcomes assessed during the OLE included mFARS, PGIC, CGIC, 9-HPT, 25-foot timed walk test, and FARS-ADL.

A detailed description of these efficacy end points is provided in the Systematic Review section.

Statistical Analysis

Because Part 1 of the MOXIe trial was dose-ranging, patients who were treated with omaveloxolone received doses ranging from 2.5 mg to 300 mg. Based on the half-life of omaveloxolone 160 mg observed in Part 1 of the MOXIe trial (i.e., approximately 26 hours ± 13 hours), patients were not expected to have any lingering omaveloxolone drug effects more than 8 weeks after the date of their last dose. The last patient enrolled in Part 1 of the MOXIe trial completed their last visit on June 13, 2017. Patients in Part 1 of the MOXIe trial received treatment for a short period of time (i.e., 12 weeks) and were then off the study treatment for at least 12 months before enrolling in the OLE. Therefore, patients who were enrolled in the OLE from Part 1 of the MOXIe trial were considered treatment-naive and included in the placebo-omaveloxolone group.

Results of the March 24, 2022, interim analyses are presented using summary tables. Safety was monitored for both patient populations: those who participated in Part 1 of the study or who were initially randomized to placebo in Part 2 (known hereafter as the placebo-omaveloxolone group), and those who were initially randomized to omaveloxolone in Part 2 (known hereafter as the omaveloxolone-omaveloxolone group). Clinical outcomes were also summarized for all patients combined (known hereafter as the overall omaveloxolone group).

For patients initially randomized to placebo in Part 1 or Part 2 of the MOXIe trial, baseline values were defined as the last nonmissing assessment before the first study drug administration in the OLE, unless otherwise specified. If the first study drug administration in the OLE occurred after the date of drug dispensation in the extension, the last measurement before the first study drug administration was considered the day 1 measurement for the calculation of baseline. For patients in Part 2, extension day 1 (week 52) efficacy and safety results were pulled from the Part 2 analysis datasets if patients were not rescreened before the extension study.

Data are summarized using descriptive statistics. Continuous data are presented with the number of patients, mean, SD, median, minimum, and maximum (where applicable). Categorical data are summarized using frequency counts and percentages. If the mFARS assessment was not completed, there was no imputation; the imputation rules applied only to partially missing data. Data missing due to the COVID-19 pandemic were not imputed.

Results

Patient Disposition

A summary of the patient disposition during the OLE is reported in Table 15. A total of 149 patients were enrolled in the OLE, including ██ patients from Part 1 of the MOXIe trial and ██ patients from Part 2 of the MOXIe trial. All patients received at least 1 dose of omaveloxolone and were included in the safety population. All ██ patients enrolled from Part 1 of the MOXIe trial were considered treatment-naive upon enrolment in the OLE and were included in the placebo-omaveloxolone group. Of patients recruited from Part 2 of the trial, ██ patients had previously been randomized to placebo and were included in the placebo-omaveloxolone group, while 43 patients had previously been randomized to omaveloxolone and were included in the omaveloxolone-omaveloxolone group. More patients from the placebo-omaveloxolone group terminated the treatment early, either by patient choice or due to an AE.

Treatment Modifications Due to the COVID-19 Pandemic

█████████████ ███ ███████████ ███ ██ ███ ████████ █████████ █████████ ██ ████████ ███████ █████████ ██ ████████ ███████ ██ ███ ████████████ █████ ███ █ ████████ ███████ ██ ███ █████████ ██████ ███ ████ ██████ ██████ ███ █████████ ████████████ ███ ██████████████ ████ ███ ███████ ██ ███████ ███████ ██████ ██ ███ ████████████ █████ ███ █████ ██ ███ █████████ ███████ █████ ██████████ ███████ ███ █████████ ████████████ ███ ██ ███ ████████ ████████ ████████ ███████ ██████ █████ ██ ███ ████████████ █████ ███ █ ██ ███ █████████ ███████ ████████████ █████████ ██████ █████ ██████ ██████ ███ █████ █████ ██████ ██████ ███ ████ ████████ ████████ ██████ ███ ██ ████████ ████ ██ ████ ███ █████████ ██ ████████ ████████ ████████ ██ ████ ██ █████████ ██ ████████ ████████ ███ ████ ██ █████████ ██ ████████ ████████

Table 15: Patient Disposition

OLE = open-label extension.

Notes: Based on the half-life of omaveloxolone 160 mg in Part 1 (i.e., approximately 26 hours ± 13 hours), patients were not expected to have lingering drug effects more than 8 weeks after the date of their last dose. For this reason, all patients who participated in Part 1 of the study (who were then off the drug for a period of time ranging from 21 months to 49 months) were treated as placebo-omaveloxolone (i.e., omaveloxolone-naive) in the extension study analysis.

The placebo-omaveloxolone group consisted of patients initially randomized to placebo in Part 2 or participating in Part 1 of the study. The omaveloxolone-omaveloxolone group consisted of patients initially randomized to omaveloxolone in Part 2.

Source: MOXIe OLE Clinical Study Report.³⁷

Baseline Characteristics

Overall, patient characteristics were similar across treatment groups (Table 16). The mean age of patients was approximately 26 years; most were aged 18 years or older (92.6%). The mean mFARS was 42.66 points, and the mean ADL score was 12.50 points. The mean age at onset of FA was 15.2 years, and the mean length of time since onset of FA was 11.0 years. The majority of patients were without pes cavus (69.8%); 73 patients (49.0%) had a GAA1 repeat length greater than or equal to 675. However, more patients in the placebo-omaveloxolone group were male (55.7%), and more patients in the omaveloxolone-omaveloxolone group were female (62.8%).

Table 16: Baseline Characteristics of the Safety Population

ADL = activities of daily living; BMI = body mass index; FA = Friedreich’s ataxia; GAA1 = guanosine-adenine-adenine 1; mFARS = modified Friedreich’s Ataxia Rating Scale; OLE open-label extension; SD = standard deviation.

Note: The placebo-omaveloxolone group consisted of patients initially randomized to placebo in Part 2 or participating in Part 1 of the study. The omaveloxolone-omaveloxolone group consisted of patients initially randomized to omaveloxolone in Part 2.

^aThe sponsor confirmed that the reported mean (SD) for GAA1 repeat length did not include all 149 patients, given that there was variation across the MOXIe Part 2 trial sites in genetic testing kits and their capabilities; this difference resulted in some sites being less able to determine the specific GAA1 repeat length.

Source: MOXIe OLE Clinical Study Report.³⁷

Exposure to Study Treatments

██ ███ ███████ ████ ██████ ███ ████ ██████████ ████ █████████ █████ ███ ████████ ███ █████████ ██ ████ █ ████ ████████ ██ █████ ████ ██████ ████ █████ ██████████ ███ ████ █████ ██████ ███ ████ ████ ██ ████████████ █████ ███ █████████ █████ ███████ ███████ ████ ███ ██ ████████ ████████ █████████████ ██████ ███ ███ ██ ██ ████ ███ ███ █████ ██ ████████ ████████ █████████████ ██ ██ ████ ███ ████████ ████ ███ ████ ████████ ██ █████ ████ ██ treatment.

Efficacy

Modified FARS

Based on the natural history of FA progression, mFARS scores are expected to increase over time, indicating a patient’s worsening condition and neurologic function. Patients are expected to experience average increases of 1.91 points, 4.24 points, and 5.58 points in mFARS after 1 year, 2 years, and 3 years, respectively.³⁸ The results shown in Table 17 and Appendix 1 (Table 27) demonstrate that patients experienced an increase in mFARS scores over time, with patients in the omaveloxolone-omaveloxolone group experiencing a greater mean increase from baseline than patients considered treatment-naive at the start of the OLE (i.e., the placebo-omaveloxolone group); this can indicate decaying efficacy of treatment on disease progression. It should be noted that the sample sizes of the subgroups of patients with and without pes cavus were small, and that results at week 144 are based on less than half of the patients enrolled in the OLE.

Table 17: Results and Mean Change From Baseline in mFARS (Safety Population)

CFB = change from baseline; mFARS = modified Friedreich’s Ataxia Rating Scale; OLE = open-label extension; SD = standard deviation.

Note: The placebo-omaveloxolone group consisted of patients initially randomized to placebo in Part 2 or participating in Part 1 of the study. The omaveloxolone-omaveloxolone group consisted of patients initially randomized to omaveloxolone in Part 2.

Source: MOXIe OLE Clinical Study Report.³⁷

Activities of Daily Living

Results for the ADL survey are summarized in Table 18. Lower scores suggest improved function. Baseline values for the mean total ADL scores were ██████ in the placebo-omaveloxolone group and ██████ in the omaveloxolone-omaveloxolone group, indicating that patients in the placebo-omaveloxolone group may have had slightly worse function at baseline.

Changes from baseline in total ADL score for the placebo-omaveloxolone group at week 48, week 96, and week 144 were ██████ █████, and █████, respectively. Changes from baseline in total ADL score for the omaveloxolone-omaveloxolone group at week 48, week 96, and week 144 were ██████ ██████ ███ █████, respectively. Based on natural history data, the average changes at 1 year, 2 years, and 3 years are █████ █████ ███ ████, respectively.³⁸ Among all patients, the changes from baseline in total ADL score at week 48, week 96, and week 144 were ██████ ██████ ███ █████. Compared to natural history data at 3 years, patients treated with omaveloxolone had reduced long-term declines in total ADL scores; however, the sample sizes decreased over time.

9-Hole Peg Test

A longer time on the 9-HPT reflects more impairment on one’s upper-extremity function as a result of the disease; thus, a negative change from baseline suggests an improvement. Based on natural history data, an average change at 1 year is expected to be −0.001.³⁸ Patients in the both the omaveloxolone-omaveloxolone and placebo-omaveloxolone groups observed improvements from baseline Appendix 1 (Table 28).

25-Foot Timed Walk Test

Results for the timed 25-foot walk test are summarized in Appendix 1 (Table 29). Longer test times reflect more impairment of the patient’s ability to walk; thus, a decrease from baseline suggests improvement. At baseline, the mean value for the average reciprocal time (units in 1 per second) needed to complete the 25-foot walk test was ██████/second in the placebo-omaveloxolone group and ██████/second in the omaveloxolone-omaveloxolone group. At week 48, the mean changes from baseline were –██████/second in the placebo-omaveloxolone group and ███████/second in the omaveloxolone-omaveloxolone group. At week 96, the mean changes from baseline were –██████/second in the placebo-omaveloxolone group and | ██████/second in the omaveloxolone-omaveloxolone group. At week 144, the mean changes from baseline were –0.0294/second in the placebo-omaveloxolone group and ███████/second in the omaveloxolone-omaveloxolone group. The average change in natural history data at 1 year, 2 years, and 3 years ██ ██████ ██████ ███ █████, respectively.³⁸ Patients in both groups experienced improvements from baseline in their 25-foot timed walk test results over time.

Table 18: ADL Test by Study Visit (Safety Population)

CFB = change from baseline; omav = omaveloxolone; SD = standard deviation.

Note: The placebo-omaveloxolone group consisted of patients initially randomized to placebo in Part 2 or participating in Part 1 of the study. The omaveloxolone-omaveloxolone group consisted of patients initially randomized to omaveloxolone in Part 2.

Source: MOXIe OLE Clinical Study Report.³⁷

PGIC and CGIC

The PGIC and CGIC are 7-point scales that require the patient and clinician, respectively, to assess how much the patient’s illness has improved or worsened relative to baseline. PGIC and CGIC scores of less than 4 represent some measure of improvement; scores greater than 4 represent some measure of worsening; and a score of 4 represents no change.

Among all patients, the placebo-omaveloxolone group’s mean PGIC scores were ███ at week 48, ███ at week 96, and ███ at week 144. For the omaveloxolone-omaveloxolone group, mean PGIC scores were ███ at week 24, ███ at week 48 ███ at week 96, and ███ at week 144. Among all patients, mean CGIC scores at week 48, week 96, and week 144 for the placebo-omaveloxolone group were ████ ████ ███ ███, respectively. For the omaveloxolone-omaveloxolone group, mean CGIC scores at week 48, week 96, and week 144 were ████ ████ ███ ████ respectively.

Delayed-Start Analysis

An additional delayed-start analysis was conducted to evaluate the persistence of effect with omaveloxolone treatment by comparing the difference in mFARS scores at the end of the MOXIe Part 2 trial to the difference in mFARS scores after 72 weeks in the OLE. This analysis was based on data from 82 participants in the OLE FAS (placebo-omaveloxolone group: n = 42; omaveloxolone-omaveloxolone group: n = 40). The mean baseline mFARS score was higher in the omaveloxolone-omaveloxolone group (mean = 40.9; SD = 10.4) than in the placebo-omaveloxolone group (mean = 38.8; SD = 11.0). A noninferiority analysis using a single MMRM model was conducted using data from the 48-week MOXIe Part 2 trial and the OLE through extension week 144. Nearly parallel trajectories were observed between the 2 groups except from the data at week 48, which had many missing mFARS assessments due to interruptions caused by the COVID-19 pandemic (Figure 1). The change in mFARS scores of individuals in the omaveloxolone-omaveloxolone group were maintained relative to their extension baseline through week 144. There was a least squares mean difference of −2.91 at the end of the delayed-start period. There was an observed sustained treatment effect with patients who were initially treated with omaveloxolone during the MOXIe Part 2 trial and continued with omaveloxolone during the OLE (i.e., the omaveloxolone-omaveloxolone group) compared to patients who were initially randomized to placebo during Part 2, then started omaveloxolone during the OLE (i.e., the placebo-omaveloxolone group). However, sample sizes declined in both groups over time, indicating a risk of attrition bias. It is unclear whether those who were not benefiting from the treatment dropped out, skewing the results.

Figure 1: Change From Baseline in mFARS (FAS)³⁹

Ext. = extension; FAS = full analysis set; LS = least squares; mFARS = modified Friedreich’s Ataxia Rating Scale; MMRM = mixed model for repeated measures; omav = omaveloxolone; SE = standard error.

Notes: Part A of the figure depicts mean changes from baseline in mFARS score over time in the FAS for patients in the omaveloxolone-omaveloxolone (n = 40) or placebo-omaveloxolone (n = 42) groups, estimated using MMRM analysis.

Part B of the figure depicts observed mean changes from baseline in mFARS score over time in FAS for patients in the omaveloxolone-omaveloxolone (n = 40) or placebo-omaveloxolone (n = 42) groups.

Source: Lynch et al. Efficacy of Omaveloxolone in Friedreich's Ataxia: Delayed-Start Analysis of the MOXIe Extension. Mov Disord. 2023;38(2):313 to 320. Available from: https://movementdisorders.onlinelibrary.wiley.com/doi/10.1002/mds.29286. Reprinted in accordance with Creative Commons Attribution 4.0 International License (CC BY 4.0): https://creativecommons.org/licenses/by/4.0/deed.en.³⁹

Harms

A summary of harms data from the OLE is provided in Table 19. As in the pivotal trial, most patients (96.0%) experienced at least 1 AE. The most frequently occurring AEs were coronavirus infection (18.8%), ALT increase (18.8%), headache (18.1%), upper respiratory tract infection (16.8%), nausea (16.1%), diarrhea (12.8%), fatigue (13.4%), arthralgia (12.1%), contusion (12.1%), excoriation (12.1%), and vaccination complication (10.1%). As in the pivotal trial, SAEs and AEs leading to permanent treatment discontinuation occurred infrequently; these were reported in 8.7% and 6.7% of patients, respectively. No deaths were observed during the OLE. Overall, long-term treatment with omaveloxolone was considered to be safe and tolerable for patients with FA.

Refer to Table 19 for harms data.

Table 19: Summary of Harms Results From the Long-Term Extension Study

AE = adverse event; ALT = alanine aminotransferase; AST = aspartate aminotransferase; OLE = open-label extension; SAE = serious adverse event.

Note: The placebo-omaveloxolone group consisted of patients initially randomized to placebo in Part 2 or participating in Part 1 of the study. The omaveloxolone-omaveloxolone group consisted of patients initially randomized to omaveloxolone in Part 2.

^aAEs reported in greater than or equal to 3% of patients overall.

Source: MOXIe OLE Clinical Study Report.³⁷ Details are from the sponsor’s Summary of Clinical Evidence.¹¹

Critical Appraisal

Internal Validity

The long-term extension study was an open-label study without a comparator or placebo group. The lack of a control group precludes the ability to make causal statements about benefit and harm; increases in AE could not be interpreted relative to other interventions. The open-label nature of the study may increase the risk of bias in determining the magnitude of the subjective outcomes because the lack of blinding may influence patients’ expectations of the treatment, particularly for measures, including AEs, SAEs, and treatment-emergent AEs. Because completion of a pivotal trial was an eligibility criterion for enrolment in the extension study, patients who discontinued those trials due to AEs or lack of response were excluded. There is a risk of attrition bias, given that the number of patients contributing to the analyses declined steadily over time; final measures of the outcome are based on fewer than half of the enrolled patients. The COVID-19 pandemic also affected study visits and treatment continuity, with 14.8% of patients experiencing treatment interruptions.

External Validity

The clinical experts consulted for this review suggested that, aside from the exclusion of pediatric patients, the eligibility criteria of the OLE resulted in a study population comparable to patients in Canada. The trial’s strict inclusion and exclusion criteria, including constraints around cardiac involvement, may have led to a cohort that was healthier than what is typically encountered in routine Canadian practice. Furthermore, the study did not include any sites in Canada, which reduces the generalizability and applicability of the results to clinical practice in Canada.

Indirect Evidence

No indirect treatment comparisons were submitted for this review.

Studies Addressing Gaps in the Systematic Review Evidence

Contents within this section have been informed by materials submitted by the sponsor. The following has been summarized and validated by the review team.

The limitations related to the lack of long-term comparative efficacy (given that the MOXIe Part 2 trial was 48 weeks in length) are outlined in Table 20. An RWE study by Lynch et al. (2023)⁴⁰ used propensity score matching of patients from the FACOMS to provide an external control for the MOXIe trial and better understand the effects of omaveloxolone in FA. This study is summarized here.

Description of Studies

One propensity score–matched study was included to provide additional information about the longer-term efficacy of omaveloxolone compared to no treatment.

Table 20: Summary of Gaps in the Systematic Review Evidence

FA = Friedreich’s ataxia; FACOMS = Friedreich’s Ataxia Clinical Outcome Measures Study; mFARS = modified Friedreich’s Ataxia Rating Scale; OLE = open-label extension; RCT = randomized controlled trial.

Source: Sponsor’s Summary of Clinical Evidence.¹¹

Table 21: Details of Studies Addressing Gaps in the Systematic Review Evidence

FA = Friedreich’s ataxia; FACOMS = Friedreich’s Ataxia Clinical Outcome Measures Study; FARS = Friedreich’s Ataxia Rating Scale; mFARS = modified Friedreich’s Ataxia Rating Scale; OLE = open-label extension; vs. = versus.

Source: Sponsor’s Summary of Clinical Evidence.¹¹

Statistical Analysis

The data for this analysis are based on the interim analysis of the MOXIe extension trial (database lock: March 24, 2022). The data from the FACOMS were current as of March 24, 2021, and obtained from the Critical Path Institute.⁴³

Two analysis populations were assessed:

• the MOXIe extension population, which included patients from the MOXIe OLE study

• the natural history population, which included patients from the FACOMS (as described in Table 21).

In addition, because the MOXIe OLE study included patients who were considered omaveloxolone-naive as well as previously treated patients, a different propensity score–matching approach was used for each analysis population. The primary analysis of this propensity score matching consisted of the natural history population and patients divided as follows:

• The primary pooled group included all omaveloxolone-naive and -experienced patients.

• The secondary placebo-omaveloxolone group included patients who had been enrolled in Part 1; these patients had had long off-treatment periods (of at least last 21 months), which meant they were considered treatment-naive. This group also included patients who had been randomized to placebo in the MOXIe Part 2 trial, then treated with omaveloxolone during the OLE study.

• The secondary omaveloxolone-omaveloxolone group included patients who had been randomized to omaveloxolone in the MOXIe Part 2 trial and continued with omaveloxolone during the OLE.

Propensity Score Matching

The propensity score matching of patients from the FACOMS to those from the MOXIe OLE study was estimated using a logistic regression that was conditional on a given a set of covariates. The covariates used in the propensity score models included age, age of FA onset, sex, baseline mFARS score, and baseline gait score. These variables were used for matching because there was information available from both studies. It was noted that GAA1 repeat length may be a prognostic variable; however, it was not available for all patients. The FACOMS patients were matched to equivalent MOXIe OLE patients using optimal 1-to-1 matching without replacement. Diagnostics were then used to assess the similarity of the 2 groups and whether the propensity score model was adequately specified.

Primary Efficacy Analyses

The change from baseline in mFARS at year 3 was analyzed using an MMRM that included the following covariates: treatment group, baseline mFARS, visit, and interaction terms for visit-by-baseline and treatment group-by-visit. The model was fitted using restricted maximum likelihood, with a Toeplitz covariance structure (assuming that measurements taken closer together in time are more highly correlated than those taken farther apart). Mean changes from baseline, between-group differences in estimated changes from baseline, and associated P values estimated from the MMRM model are reported. Annual visits for the MMRM analysis were defined to align with the FACOMS assessment schedule. The mFARS assessment data collected closest to 1 year, 2 years, and 3 years after baseline were used for each of the annual assessments.

Secondary Efficacy Analyses

The same MMRM was used for both the primary and secondary analyses.

Results

Patient Disposition

A total of 136 patients had available data for key variables and were included in the propensity score–matched analysis. A total of 810 patients consented to have their data shared outside the core FACOMS; of these, 598 patients met the criteria for inclusion in the natural history study population and were considered potential matches for MOXIe OLE patients. After matching, 136 patients from the FACOMS natural history study were included. In total, the primary pooled population included 272 patients, with 136 patients from each of the MOXIe OLE and FACOMS.

Baseline Characteristics

A summary of baseline characteristics and covariates used for the propensity scores are reported in Table 22. Characteristics were balanced across the matched FACOMS (n = 136) and MOXIe extension (n = 136) groups. The trends observed in the primary pooled population (N = 272) were similar to those in the placebo-omaveloxolone (N = 190) and omaveloxolone-omaveloxolone (N = 182) groups (summarized in Supplementary Table 3 and Supplementary Table 4 of Lynch et al. [2023]).⁴⁰ The mean baseline mFARS scores were similar across the matched FACOMS patients (41.0; SD = 16.1) and MOXIe OLE patients (42.2; SD = 12.6).

Table 22: Demographics and Baseline Characteristics Used as Covariates for Propensity Score Calculation (Primary Pooled Population)

ADL = activities of daily living; BMI = body mass index; FA = Friedreich's ataxia; FACOMS = Friedreich’s Ataxia Clinical Outcome Measures Study; FARS = Friedreich’s Ataxia Rating Scale; GAA1 = guanosine-adenine-adenine 1; GAA2 = guanosine-adenine-adenine 2; mFARS = modified Friedreich’s Ataxia Rating Scale; SD = standard deviation.

Source: Sponsor’s Summary of Clinical Evidence.¹¹

Exposure to Study Treatments

The mean follow-up durations were similar among patients in both studies. Among the matched FACOMS patients, the mean follow-up was 2.54 years (SD = 0.786), while among patients in the MOXIe OLE, it was 2.60 years (SD = 0.524).

Efficacy

A summary of mean change in mFARS scores after 1 year, 2 years, and 3 years is provided in Table 23. There was an observed improvement among patients treated with omaveloxolone in the MOXIe OLE trial compared to matched FACOMS patients at each time point. Patients from the MOXIe OLE trial experienced slower increases in mFARS scores than matched FACOMS patients (i.e., indicative of slowed disease progression).

Change From Baseline in mFARS Score at Year 3

After 3 years, the estimated changes from baseline in mFARS scores were 6.61 (SE = 0.65) in matched FACOMS patients and 3.00 (SE = 0.66) in OLE patients, representing a statistically significant difference of −3.61 (95% CI, −1.79 to −5.43) in favour of omaveloxolone (Figure 2). These results were consistent among the placebo-omaveloxolone and omaveloxolone-omaveloxolone groups (Table 23).

Figure 2: Change in mFARS From Baseline Over Time (Primary Pooled Population)

CI = confidence interval; FA-COMS = Friedreich’s Ataxia Clinical Outcome Measures; LS = least squares; mFARS = modified Friedreich’s Ataxia Rating Scale; MMRM = mixed model for repeated measures; OLE = open-label extension.

Note: LS refers to MMRM estimates.

Source: Sponsor’s Summary of Clinical Evidence.¹¹

Table 23: Change in mFARS Over 3 Years

FACOMS = Friedreich’s Ataxia Clinical Outcome Measures Study; mFARS = modified Friedreich’s Ataxia Rating Scale; MMRM = mixed model for repeated measures; NA = not applicable; NR = not reported; SD = standard deviation; SE = standard error.

^aThe mean difference in change from baseline was estimated using an MMRM.

Source: Details are from the sponsor’s Summary of Clinical Evidence.¹¹

Harms

Harms data were not assessed in this study.

Critical Appraisal

The International Society for Pharmacoepidemiology (ISPE) RWE appraisal tool — modified for use by CDA-AMC according to the guidance documentation — was used to assess the quality of the propensity scored–matched analysis.⁴⁴ The Guidance for Real-World Evidence reporting was not completed before submission by the CDA-AMC; thus, it was not reviewable. Key excluded sections are noted as follows.

Setting and Context

No information was provided about differences in health systems, access to care, available health care resources, or other factors that may affect the care of patients with FA and, in turn, the applicability of findings to the Canadian context.

Data Specifications

No information was provided about data access, cleaning, or linkage.

Data Sources, Data Dictionary, and Variables

Detailed descriptions of data sources, data dictionary and variables were not provided. This means important variables were not captured, potentially affecting study results.

An assessment of how the RWE study follows the CDA-AMC Guidance for Reporting Real-World Evidence checklist was conducted.

Study Design

The study design was considered appropriate because it used comparative evidence and described the efficacy of treatment compared to no treatment in a treatment-naive population. However, the choice of the first visit as baseline in the FACOMS group can introduce indication bias because it is unclear whether there are unmeasured confounding factors that would cause patients to enroll in the FACOMS. While no issues were identified with the timing of treatment initiation, the primary analysis cohort includes patients who were randomized to omaveloxolone as well as patients who were randomized to placebo and completed 48 weeks of follow-up before initiating treatment. The patients who completed 48 weeks of follow-up may be systematically different from those included in FACOMS because the matching strategy did not adjust for these unmeasured criteria; bias may result.

Methods and Results

There is limited potential for bias due to exposure or for outcome misclassification due to study design. However, there is the possibility of measurement error in the results, particularly at year 3; the mean study day of collection of mFARS results was 2 months to 3 months later in matched FACOMS patients than in the MOXIe cohort, which may bias the results slightly in favour of omaveloxolone. The variables included in the propensity score matching were deemed sufficient by CDA-AMC clinical experts. The diagnostic results indicated that the FACOMS cohort constructed using matching was sufficiently comparable to the MOXIe cohort, based on the included covariates. Covariate values were nearly completely known; this reduces the risk of bias being introduced due to the exclusion of patients based on important but unknown information.

The estimation of progression at 3 years is reliant on a MAR assumption, while the numbers of observed scores at the 3-year mark are 77 (57%) in the omaveloxolone group and 83 (61%) in the matched FACOMS natural history study cohort. The sensitivity of the conclusion to the MAR assumption was not reported, which could be important. No analyses describing the causes of missing outcomes or dropouts are provided; this raises the question of whether the benefit observed is due to a depletion of the patients remaining in the study or AEs inducing dropout.

Discussion

Summary of Available Evidence

The primary evidence for omaveloxolone as a treatment for FA comes from 1 pivotal, phase II, randomized, double-blinded, placebo-controlled trial (the MOXIe Part 2 trial, N = 103) comparing omaveloxolone 150 mg (3 capsules of 50 mg each) to placebo over 48 weeks. Additional long-term information is provided from an OLE study (N = 77), with evidence from a real-world external control analysis comparing patients in the MOXIe OLE study (N = 136) to matched natural history controls from the FACOMS database (N = 136) over 3 years. The trial analyses evaluated differences in disease progression using mFARS scores at 48 weeks from baseline, while the comparison of the MOXIe OLE study to FACOMS evaluated changes in mFARS at year 1, year 2, and year 3.

The MOXIe Part 2 trial and MOXIe OLE study populations consisted primarily of young adults with a mean age of around 26 years and a mean of 14 years since diagnosis. Most, but not all, of the key baseline characteristics were generally well-balanced between treatment groups in the MOXIe Part 2 trial, with mean baseline mFARS scores of 40.1 and 37.9 in the groups receiving omaveloxolone and placebo, respectively, in the all-randomized population. In the external control analysis, propensity score matching was used to identify FACOMS controls with similar baseline characteristics to the MOXIe cohort across factors, including age, sex, GAA repeat length, disease duration, baseline mFARS score, and ambulatory status.

Interpretation of Results

Efficacy

The primary treatment goal was to identify a treatment benefit of omaveloxolone through change in mFARS score. The clinical experts consulted indicated that a change of 2 points in this score is considered clinically meaningful, but also that stabilization of disease over a long period would be considered beneficial, given the progressive nature of FA. The MOXIe Part 2 trial demonstrated a statistically significant improvement in mFARS scores after 48 weeks of treatment with omaveloxolone compared to placebo (i.e., a difference of −2.40 points; P = 0.014). In addition, according to the GRADE framework, the certainty of evidence is moderate, indicating that compared to placebo, omaveloxolone likely results in slower progression in mFARS scores. In the OLE study, a comparison of treatment-naive patients and patients maintaining treatment supports this finding but illustrates that the treatment effects may wain or decrease, given that the newly treated patients had better mFARS scores after 48 weeks. The RWE comparison, in which patients maintained an observed improvement in mFARS scores for up to 3 years compared to an external cohort, illustrated that treated and nontreated patients progressed at similar rates after the first year. While this provides supportive evidence of effectiveness in clinical practice, there are important limitations — including potential selection bias in the FACOMS cohort and missing data at later time points — that warrant cautious interpretation.

Key outcomes identified as important by patient groups included maintaining independence in ADLs, slowing disease progression, improving quality of life, managing fatigue and energy levels, and maintaining the ability to work and/or attend school. While the mFARS score captures aspects of physical function and disease progression (encapsulated in the mFARS end point), direct measures of quality of life, fatigue, and ADLs were either not assessed in the clinical trial or did not show significant treatment effects. This includes the key secondary outcomes of PGIC and CGIC. Beyond the primary outcomes, all outcomes tested in the MOXIe trial either included the null or fell outside the statistical testing hierarchy, suggesting that the only evidence of efficacy for omaveloxolone was seen in the mFARS end point. This represents a gap in the evidence for the outcomes most meaningful to patients. The clinical experts noted that improvements in mFARS would be expected to translate to functional benefits; however, this relationship has not been formally established.

Other notable evidence gaps include effectiveness in patients with more advanced disease, impact on non-neurologic manifestations, effects on disease progression in very early disease, and comparative long-term efficacy (i.e., beyond 3 years).

Harms

Common AEs included headache, nausea, fatigue, and abdominal pain. SAEs were numerically higher among patients receiving omaveloxolone than among those receiving placebo (9.8% versus 5.8%), with no deaths reported. Discontinuations due to AEs were also numerically higher among patients receiving omaveloxolone than among those receiving placebo (7.8% versus 3.8%).

The product monograph recommends monitoring liver function regularly and exercising caution when omaveloxolone is coadministered with CYP3A4 inhibitors or inducers. It also advises caution in patients with hepatic impairment. The clinical experts agreed that these risks can be managed effectively through routine monitoring and dose adjustments, as needed, in typical clinical practice. Patient groups have expressed that the observed side effects are acceptable, given the progressive nature of the disease and the limited treatment options available. Commonly experienced symptoms, such as headache and nausea, were considered manageable with supportive care measures.

However, long-term safety data (i.e., beyond 3 years) remain limited. Clinical experts have highlighted several areas for ongoing monitoring, including cardiovascular effects, liver function, and growth and development in pediatric populations. There is also interest in better understanding the potential effects of omaveloxolone on non-neurologic manifestations of the disease as more long-term data become available.

Other Considerations

There are currently no approved disease-modifying treatments for FA. A significant unmet need for such a treatment is consistently highlighted by patient groups and clinical experts. The oral administration of omaveloxolone was viewed favourably by patients compared to potential injectable therapies in development, although some noted challenges with the current requirement for 150 mg (3 capsules of 50 mg each) taken orally once daily.

Several ongoing studies are evaluating omaveloxolone, including:

• OLE studies following patients for up to 5 years

• studies in patients with more advanced disease

• pediatric studies

• studies examining cardiac effects.

Results from these studies will help address current evidence gaps.

The clinical experts noted that, while questions remain about long-term benefits, the available evidence supports offering this treatment option to eligible patients, given the serious nature of the condition and the lack of alternatives.

Conclusion

Evidence from 1 phase II RCT suggests that, compared to placebo, omaveloxolone likely results in slower decline in neurologic function as measured by mFARS scores over 48 weeks in patients with FA aged 16 years and older. It likely results in improvements in ADL measures, which include activities involving upper limb function, mobility, and frequency of falls. While this is suggestive of functional relevance, with the clinical experts suggesting 2-point and 1-point potentially meaningful thresholds for mFARS and ADL, respectively, the actual clinical significance of these observations is uncertain due to the lack of an established minimal clinically important difference in mFARS and ADL scores. Other functional outcomes, including upper limb function, mobility, and frequency of falls, did not show significant improvements with treatment.

Comparison with natural history controls suggests that the treatment benefit of omaveloxolone may be maintained for up to 3 years. However, these studies do not allow clear conclusions to be drawn about whether disease progression will continue to be slower among patients treated with omaveloxolone than among matched controls. A long-term extension study suggests that benefit may be observed within the first year, given that treatment-naive patients were observed to have better responses. However, these findings are limited by potential selection bias and the study’s open-label design; in addition, the magnitude of these group differences is uncertain. The safety profile of omaveloxolone appears to be manageable, with liver enzyme elevations being the primary concern requiring monitoring. Common adverse effects, including gastrointestinal symptoms and fatigue, were generally mild to moderate in severity.

Important evidence gaps remain, particularly regarding efficacy in pediatric populations (which represent a significant proportion of newly diagnosed patients), patients with more advanced disease or significant cardiac involvement, and long-term comparative efficacy beyond 3 years. Additionally, while mFARS captures aspects of neurologic function, there is limited evidence on outcomes identified as important by patients, such as HRQoL, fatigue, and ability to maintain independence. Given that FA is a progressive, debilitating condition with no approved disease-modifying treatment, even a modest slowing of progression could be meaningful to patients. However, uncertainty remains about the magnitude and durability of benefit.

References

1.Durr A, Cossee M, Agid Y, et al. Clinical and genetic abnormalities in patients with Friedreich's ataxia. N Engl J Med. 1996;335(16):1169-75. doi:10.1056/NEJM199610173351601 PubMed

2.Opal P, Zoghbi HY. Post T, ed. Friedreich Ataxia. UpToDate; 2023. Accessed October 29, 2024. http://www.uptodate.com

3.Buesch K, Zhang R. A systematic review of disease prevalence, health-related quality of life, and economic outcomes associated with Friedreich's Ataxia. Curr Med Res Opin. 2022;38(10):1739-1749. doi:10.1080/03007995.2022.2112870 PubMed

4.Polek B, Roach MJ, Andrews WT, Ehling M, Salek S. Burden of Friedreich's Ataxia to the Patients and Healthcare Systems in the United States and Canada. Front Pharmacol. 2013;4:66. doi:10.3389/fphar.2013.00066 PubMed

5.Ataxia Canada. Biogen Data on file [confidential information; sponsor supplied reference]. Biogen Canada, Inc.; 2024.

6.Indelicato E, Nachbauer W, Eigentler A, et al. Onset features and time to diagnosis in Friedreich's Ataxia. Orphanet J Rare Dis. 2020;15(1):198. doi:10.1186/s13023-020-01475-9 PubMed

7.Parkinson MH, Boesch S, Nachbauer W, Mariotti C, Giunti P. Clinical features of Friedreich's ataxia: classical and atypical phenotypes. J Neurochem. 2013;126 Suppl 1:103-17. doi:10.1111/jnc.12317 PubMed

8.Lynch DR, Schadt K, Kichula E, McCormack S, Lin KY. Friedreich Ataxia: Multidisciplinary Clinical Care. J Multidiscip Healthc. 2021;14:1645-1658. doi:10.2147/jmdh.S292945 PubMed

9.National Institute of Neurological Disorders and Stroke (NINDS). Friedreich Ataxia [sponsor supplied reference]. 2023. Accessed April 28, 2023. https://www.ninds.nih.gov/health-information/disorders/friedreich-ataxia

10.Tsou AY, Paulsen EK, Lagedrost SJ, et al. Mortality in Friedreich ataxia. J Neurol Sci. 2011;307(1-2):46-9. doi:10.1016/j.jns.2011.05.023 PubMed

11.Biogen Canada Inc. Sponsor Summary of Clinical Evidence for Omaveloxolone (Brand name TBD) for Friedreich’s Ataxia [internal sponsor's report]. In: Drug Reimbursement Review sponsor submission: Skyclarys (omaveloxolone), 50 mg oral capsule. Biogen Canada; 2023.

12.Silva AM, Brown JM, Buckle VJ, Wade-Martins R, Lufino MM. Expanded GAA repeats impair FXN gene expression and reposition the FXN locus to the nuclear lamina in single cells. Hum Mol Genet. 2015;24(12):3457-71. doi:10.1093/hmg/ddv096 PubMed

13.Bidichandani SI, Delatycki MB. Friedreich ataxia [sponsor supplied reference]. GeneReviews®; 2017.

14.Rummey C, Corben LA, Delatycki M, et al. Natural History of Friedreich Ataxia: Heterogeneity of Neurologic Progression and Consequences for Clinical Trial Design. Neurology. 2022;99(14):e1499-e1510. doi:10.1212/wnl.0000000000200913 PubMed

15.Reetz K, Dogan I, Hilgers RD, et al. Progression characteristics of the European Friedreich's Ataxia Consortium for Translational Studies (EFACTS): a 4-year cohort study. Lancet Neurol. 2021;20(5):362-372. doi:10.1016/S1474-4422(21)00027-2 PubMed

16.Burk K. Friedreich Ataxia: current status and future prospects. Cerebellum Ataxias. 2017;4:4. doi:10.1186/s40673-017-0062-x PubMed

17.Boghdeh NA, McGraw B, Barrera MD, et al. Inhibitors of the Ubiquitin-Mediated Signaling Pathway Exhibit Broad-Spectrum Antiviral Activities against New World Alphaviruses. Viruses. 2023;15(3):655. doi:10.3390/v15030655 PubMed

18.Corben LA, Collins V, Milne S, et al. Clinical management guidelines for Friedreich ataxia: best practice in rare diseases. Orphanet J Rare Dis. 2022;17(1):415. doi:10.1186/s13023-022-02568-3 PubMed

19.National Organization for Rare Disorders (NORD). Friedreich’s Ataxia. 2023. Accessed April 28, 2023. https://rarediseases.org/rare-diseases/friedreichs-ataxia/

20.Biogen Canada Inc. Skyclarys (omaveloxolone): capsule, 50 mg, oral [draft product monograph; sponsor supplied reference]. 2024. Accessed 2024.

21.Lynch DR, Chin MP, Delatycki MB, et al. Safety and Efficacy of Omaveloxolone in Friedreich Ataxia (MOXIe Study). Ann Neurol. 2021;89(2):212-225. doi:10.1002/ana.25934 PubMed

22.Lynch D. Correction to Safety and Efficacy of Omaveloxolone in Friedreich Ataxia (MOXIe Study). Ann Neurol. 2023;94(6):1190. doi:10.1002/ana.26808 PubMed

23.Hendrix S, Goldsberry A, Meyer C, Lynch D. Efficacy of Omaveloxolone in Friedreich' s Ataxia: Post-Hoc Analysis Using Global Statistics Test to Strengthen Secondary Endpoint Analyses. Neurology. 2023;100(17_supplement_2):1. doi:10.1212/WNL.0000000000203544

24.Reata Pharmaceuticals Inc. Clinical Study Report. MOXIe Part 2 CSR: RTA 408 (omaveloxolone) 408-C-1402 Part 2: A phase 2 study of the safety, efficacy, and pharmacodynamics of RTA 408 in the treatment of Friedreich’s ataxia [internal sponsor's report]. November 5, 2020.

25.Azevado M. Part One of Reata’s MOXIe Trial for Friedreich’s Ataxia Shows Promising Results [sponsor supplied reference]. Friedreich's Ataxia News: A Bionews Brand. 2017. https://friedreichsataxianews.com/news/part-one-of-moxie-trial-of-omaveloxolone-for-friedreichs-ataxia-shows-promising-results/

26.Reata Pharmaceuticals Inc. Clinical Study Report. RTA 408 (OMAVELOXOLONE) 408-C-1402 Part 1: A Phase 2 Study of the Safety, Efficacy, and Pharmacodynamics of RTA 408 in the Treatment of Friedreich’s Ataxia [internal sponsor's report]. September 29, 2020.

27.Reata Pharmaceuticals Inc. Protocol. RTA 408 (408-C-1402) A phase 2 study of the safety, efficacy, and pharmacodynamics of RTA 408 in the treatment of Friedreich's ataxia: Study Procedure Manual Version 13.0 [internal sponsor's report]. April 11, 2024.

28.Rummey C, Farmer JM, Lynch DR. Predictors of loss of ambulation in Friedreich's ataxia. EClinicalMedicine. 2020;18:100213. doi:10.1016/j.eclinm.2019.11.006 PubMed

29.Rummey C, Corben LA, Delatycki MB, et al. Psychometric properties of the Friedreich Ataxia Rating Scale. Neurol Genet. 2019;5(6):371. doi:10.1212/nxg.0000000000000371 PubMed

30.Rummey C, Zesiewicz TA, Perez-Lloret S, Farmer JM, Pandolfo M, Lynch DR. Test-retest reliability of the Friedreich's ataxia rating scale. Ann Clin Transl Neurol. 2020;7(9):1708-1712. doi:10.1002/acn3.51118 PubMed

31.Tai G, Corben LA, Woodcock IR, Yiu EM, Delatycki MB. Determining the Validity of Conducting Rating Scales in Friedreich Ataxia through Video. Mov Disord Clin Pract. 2021;8(5):688-693. doi:10.1002/mdc3.13204 PubMed

32.Fahey MC, Corben LA, Collins V, Churchyard AJ, Delatycki MB. The 25-foot walk velocity accurately measures real world ambulation in Friedreich ataxia. Neurology. 2007;68(9):705-6. doi:10.1212/01.wnl.0000256037.63832.6f PubMed

33.Traschutz A, Fleszar Z, Hengel H, et al. FARS-ADL across Ataxias: Construct Validity, Sensitivity to Change, and Minimal Important Change. Mov Disord. 2024;39(6):965-974. doi:10.1002/mds.29788 PubMed

34.Reata Pharmaceuticals Inc. Clinical Study Report Addendum. 408-C-1402 Part 2 Clinical Study Report. A phase 2 study of the safety, efficacy, and pharmacodynamics of RTA 408 in the treatment of Friedreich’s ataxia [internal sponsor's report]. November 5, 2020.

35.Balshem H, Helfand M, Schünemann HJ, et al. GRADE guidelines: 3. Rating the quality of evidence. J Clin Epidemiol. 2011;64(4):401-6. doi:10.1016/j.jclinepi.2010.07.015 PubMed

36.Santesso N, Glenton C, Dahm P, et al. GRADE guidelines 26: informative statements to communicate the findings of systematic reviews of interventions. J Clin Epidemiol. 2020;119:126-135. doi:10.1016/j.jclinepi.2019.10.014 PubMed

37.Reata Pharmaceuticals Inc. MOXIe OLE Interim CSR: RTA 408 (Omaveloxolone) 408-C-1402 Extension. A Phase 2 Study of the Safety, Efficacy, and Pharmacodynamics of RTA 408 in the Treatment of Friedreich’s Ataxia. Interim Clinical Study Report [internal sponsor's report]. 2021.

38.Patel M, Isaacs CJ, Seyer L, et al. Progression of Friedreich ataxia: quantitative characterization over 5 years. Ann Clin Transl Neurol. 2016;3(9):684-94. doi:10.1002/acn3.332 PubMed

39.Lynch DR, Chin MP, Boesch S, et al. Efficacy of Omaveloxolone in Friedreich's Ataxia: Delayed-Start Analysis of the MOXIe Extension. Mov Disord. 2023;38(2):313-320. doi:10.1002/mds.29286 PubMed

40.Lynch DR, Goldsberry A, Rummey C, et al. Propensity matched comparison of omaveloxolone treatment to Friedreich ataxia natural history data. Ann Clin Transl Neurol. 2024;11(1):4-16. doi:10.1002/acn3.51897 PubMed

41.De Michele G, Perrone F, Filla A, et al. Age of onset, sex, and cardiomyopathy as predictors of disability and survival in Friedreich's disease: a retrospective study on 119 patients. Neurology. 1996;47(5):1260-4. doi:10.1212/wnl.47.5.1260 PubMed

42.Lynch D, Goldsberry A, Rummey C, et al. Direct utility of natural history data in analysis of clinical trials: Propensity match-based analysis of Omaveloxolone in Friedreich ataxia using the FA-COMS dataset. medRxiv. 2022:2022.08.12.22278684. doi:10.1101/2022.08.12.22278684

43.Critical Path Institute (C-Path), Friedreich’s Ataxia Research Alliance (FARA). C-Path and FARA Launch Friedreich's Ataxia Integrated Clinical Database to Advance the Development of Treatments for FA. 2019. Accessed October 24, 2024. https://c-path.org/c-path-and-fara-launch-friedreichs-ataxia-integrated-clinical-database-to-advance-the-development-of-treatments-for-fa/

44.CADTH. Guidance for Reporting Real-World Evidence. 2023. Accessed December 27, 2024. https://www.cda-amc.ca/sites/default/files/RWE/MG0020/MG0020-RWE-Guidance-Report-Secured.pdf

45.Reata Pharmaceuticals Inc. Clinical Study Report: MOXIe Extension 408-C-1402_CSR_Tables [internal sponsor's report]. 2022.

Appendix 1: Detailed Outcome Data

Please note that this appendix has not been copy-edited.

Patient Global Impression of Change

The PGIC is a 7-point scale that requires patients to assess how much their illness has improved or worsened relative to a baseline state at the beginning of an intervention. The PGIC is assessed by completing the following statement “since I began trial treatment, my overall status is: very much improved (1), much improved (2), minimally improved (3), no change (4), minimally worse (5), much worse (6), and very much worse (7).” Patients completed this assessment following completion of the FARS neurologic assessment or the CGIC.

The PGIC measures the patient’s subjective perception of overall improvement or decline in their condition from baseline. It is scored from 1 (“very much improved”) to 7 (“very much worse”), with lower scores indicating better perceived improvement and higher scores indicating greater perceived worsening.

Table 24: Key Secondary Analysis — CGIC at Week 48

CGIC = Clinical Global Impression of Change; CI = confidence interval; LS = least squares; PGIC = Patient Global Impression of Change.

^aComparison of PGIC and CGIC for patients treated with omaveloxolone 150 mg and placebo was estimated using analyses of covariance, with the following fixed factors: site, pes cavus status, and treatment. Missing data were imputed using multiple imputation.

Source: MOXIe trial Part 2 Clinical Study Report.²⁴

Table 25: Key Secondary Analysis — PGIC at Week 48

CGIC = Clinician Global Impression of Change; CI = confidence interval; LS = least squares; PGIC = Patient Global Impression of Change.

^aComparison of PGIC and CGIC for patients treated with omaveloxolone 150 mg and placebo was estimated using analyses of covariance, with the following fixed factors: site, pes cavus status, and treatment. Missing data were imputed using multiple imputation.

Source: MOXIe trial Part 2 Clinical Study Report,²⁴ MOXIe trial Part 2 Clinical Study Report Erratum 1.³⁴

Clinical Global Impression of Change

The CGIC is a 7-point scale that requires the clinician to assess how much the patient's illness has improved or worsened relative to a baseline state at the beginning of an intervention (Guy, 1976). The CGIC is assessed by completing the following statement “Compared to the patient’s condition at the start of the trial, this patient’s overall status is: very much improved (1), much improved (2), minimally improved (3), no change (4), minimally worse (5), much worse (6), and very much worse (7).” Clinicians could take into account prior results of study assessments to influence their clinical judgment of the CGIC. Clinicians completed this assessment following the patient’s completion of the FARS neurologic assessment or the PGIC.

The CGIC reflects the clinician’s overall impression of how the patient’s condition has changed from baseline. Similar to the PGIC, it is scored on a 7-point scale (1 = “very much improved,” 7 = “very much worse”), where lower scores represent greater clinical improvement and higher scores represent greater clinical worsening.

Table 26: Duration of Study Treatment and Exposure to Omaveloxolone (Safety Population)

SD = standard deviation.

Note: The placebo-omaveloxolone group consisted of patients initially randomized to placebo in Part 2 or participating in Part 1 of the study. The omaveloxolone-omaveloxolone group consisted of patients initially randomized to omaveloxolone in Part 2.

Source: MOXIe OLE Clinical Study Report.³⁷

Table 27: mFARS Results and Mean Change From Baseline (Safety Population)