PDF: ReBUILD
(2017)Objective
To assess the efficacy and safety of clemastine fumarate as a remyelinating therapy for chronic demyelinating optic neuropathy in patients with relapsing multiple sclerosis
Study Summary
• Clemastine fumarate reduced VEP P100 latency delay by 1.7 ms/eye (p=0.0048) in crossover analysis; 3.2 ms in delayed-treatment analysis (p=0.0001)
• Effect was sustained after treatment cessation, suggesting structural remyelination rather than transient pharmacological effect
Intervention
Clemastine fumarate 5.36 mg orally twice daily (10.72 mg/day total)
Inclusion Criteria
Relapsing MS (McDonald criteria), disease duration <15 years, VEP P100 latency ≥118 ms in at least one eye, RNFL thickness >70 µm on OCT, stable immunomodulatory therapy, no optic neuritis within 6 months or >5 years prior
Study Design
Arms: Crossover design - Group 1: Clemastine 90 days then placebo 60 days; Group 2: Placebo 90 days then clemastine 60 days
Patients per Arm: 25 per group (50 total)
Outcome
• 16-26% of patients showed >6 ms latency improvement on treatment vs 3-6% on placebo
• LCLA improved 0.9 letters/eye (p=0.085, crossover); 1.6 letters (p=0.022, delayed-treatment)
Bottom Line
Clemastine fumarate significantly reduced VEP P100 latency delay by 1.7 ms/eye (p=0.0048) in the prespecified crossover analysis and by 3.2 ms/eye (p=0.0001) in the post-hoc delayed-treatment analysis. The sustained improvement after treatment cessation suggests structural remyelination rather than a transient pharmacological effect. This is the first randomized controlled trial to document efficacy of a remyelinating drug for chronic demyelinating injury in multiple sclerosis, providing proof-of-concept that myelin repair is achievable even after prolonged damage.
Major Points
- First successful RCT demonstrating drug-induced remyelination in a chronic neurodegenerative condition
- Primary endpoint met: VEP P100 latency reduced by 1.7 ms/eye (95% CI 0.5-2.9; p=0.0048) using crossover analysis
- Post-hoc delayed-treatment analysis showed 3.2 ms reduction (95% CI 1.8-4.7; p=0.0001), better reflecting the sustained treatment effect
- Clinical benefit was sustained into the second epoch after treatment cessation (group 1), indicating structural repair rather than transient pharmacological effect
- 16% of group 1 and 26% of group 2 showed >6 ms latency improvement on treatment vs 3% and 6% on placebo
- Low-contrast letter acuity (LCLA) showed trend toward improvement: 0.9 letters/eye (p=0.085) crossover; 1.6 letters (p=0.022) delayed-treatment
- Clemastine promotes oligodendrocyte precursor cell differentiation via off-target antimuscarinic effects, confirmed in vitro, animal models, and human cells
- 92% of patients were on concurrent immunomodulatory therapy with no interaction detected
- Fatigue was the main adverse event (worsened on treatment, p=0.017); no serious adverse events
- MRI measures (MWF, MTR, FA) showed no changes - high variability limits utility as remyelination biomarkers
Study Design
- Study Type
- Single-center, double-blind, randomized, placebo-controlled, crossover trial (phase 2)
- Randomization
- Yes
- Blinding
- Double-blind. All patients and investigators were masked to group assignment for the duration of the trial including evaluation of all data and outcomes. UCSF investigational pharmacy performed randomization. VEP quality assessment and P100 latency marking done by masked investigators not involved in clinical assessment, in batch at study completion.
- Sample Size
- 50
- Follow-up
- 150 days (5 months)
- Centers
- 1
- Countries
- USA
Primary Outcome
Definition: Shortening of P100 latency delay on full-field, pattern-reversal visual-evoked potentials (VEPs). VEPs recorded monocularly with check size 64-min of arc, at least 100 averages per recording. P100 latency defined by masked investigators. Baseline = average of screening and baseline values.
| Control | Intervention | HR/OR | P-value |
|---|---|---|---|
| See crossover analysis | See crossover analysis | - (0.5-2.9 (crossover); 1.8-4.7 (delayed-treatment)) | 0.0048 (crossover); 0.0001 (delayed-treatment) |
Limitations & Criticisms
- Small sample size (n=50) at single center limits generalizability
- Crossover design assumption of no carryover was violated, complicating interpretation (sustained effect in epoch 2 for group 1)
- Standard crossover analysis underestimates treatment magnitude due to sustained carryover effect
- Cannot determine optimal treatment duration - different exposure times (90 vs 60 days) showed similar efficacy
- Long-term durability of remyelination effect beyond 2 months is unknown
- Cannot assess if higher doses would provide enhanced benefit or if effects extend to other CNS pathways
- MRI biomarkers (MWF, MTR, FA) showed too much variability to detect biological effects
- LCLA improvement confounded by learning effect over the course of the trial
- Fatigue adverse effect may limit real-world tolerability at this dose
- Trial cannot distinguish if visual pathway is more amenable to repair than other pathways
- Most patients (56%) had prior optic neuritis - cannot fully extrapolate to subclinical demyelination
- Sex imbalance between groups (76% vs 52% female) not fully balanced by randomization
Citation
Lancet 2017; Published Online October 10, 2017