Topline data: In the combined Phase 2 REPAIR program, 12 weeks of CNM-Au8 increased the brain NAD+/NADH ratio by 8.65% in 39 evaluable patients across relapsing MS, non-active progressive MS, and Parkinson’s disease (absolute change +0.449 units; 95% CI 0.093–0.805; p=0.0148; percent change p=0.0006). In MS-only cohorts (n=26), percent change was +9.49% (95% CI 1.14%–17.85%; p=0.0275), while the absolute change trended but did not reach nominal significance (+0.480; p=0.058). Secondary analyses showed increased brain NAD+ and decreased NADH fractions in the full population (p=0.0058) and in MS (p=0.0232). Treatment-emergent adverse events were transient and mostly mild to moderate. At baseline, greater disability correlated with lower NAD+/NADH (EDSS ρ=-0.429; p=0.0127), and cognitive processing speed (SDMT) and upper limb function (9-HPT) correlated with higher brain ATP peaks (ρ=0.542; p=0.0009 and ρ=-0.513; p=0.0032, respectively).
Core news: Clene presented the combined REPAIR-MS and REPAIR-PD findings at ECTRIMS 2025 and disclosed feedback from a Type B end-of-Phase 2 MS meeting in which the FDA acknowledged limitations of EDSS and signaled openness to alternative primary endpoints, including cognition, for future MS trials. REPAIR was an open-label, sequential-group, investigator-blinded magnetic resonance spectroscopy study using 31P-MRS to quantify bioenergetic indices as mechanistic readouts over 12 weeks with a six-week safety follow-up.
Strategy and tension: The company is leaning into a mechanistic biomarker narrative—improving neuronal redox state and ATP-related measures—to justify moving beyond EDSS in progressive MS, where disability accrual is slow and EDSS sensitivity is widely debated. The baseline correlations to EDSS, SDMT, and 9-HPT provide a translational bridge from bioenergetics to clinical function, positioning cognition-centric endpoints as more aligned with the drug’s proposed mechanism and potentially more responsive over feasible trial timelines. The tension is clear: the signal is biologically coherent but comes from a small, uncontrolled, open-label dataset. Percent-change statistics are stronger than absolute-change p-values in MS-only analyses, and the readouts are surrogate in nature. Any registrational strategy will have to convert this biomarker narrative into durable, clinically meaningful benefits under randomized, blinded conditions.
Who’s affected and how: For sponsors, FDA’s openness to non-EDSS primaries in MS—particularly cognition—could reshape endpoint selection in progressive populations and reduce dependence on long-duration disability composites. CROs and imaging core labs stand to benefit if 31P-MRS remains a key secondary, but they will need harmonized acquisition protocols, scanner calibration, and centralized QC to manage inter-site variability. Sites without spectroscopy capability may be excluded, tightening the footprint and elevating start-up complexity. If SDMT or upper limb function measures are elevated in hierarchy, rater training, practice-effect mitigation, and potential use of validated digital tools will become operational priorities. Regulators gain a test case for broader MS endpoint modernization, while patients in non-active progressive MS could see trials that better capture cognition and function changes that matter day-to-day.
What’s next: Watch the Phase 3 design choices—primary endpoint selection (SDMT or composite), duration sufficient to test durability, and whether non-active progressive MS is the target population. Expect MRS to be retained as a mechanistic secondary rather than a primary efficacy measure. Key risks include translation of short-term biomarker gains into clinically meaningful benefit, control of practice effects in cognitive testing, and ensuring assay/site consistency if spectroscopy is included. Comparator strategy, background therapy rules, and global feasibility will signal how ambitious the program is and how many sites can realistically participate. If Clene can replicate a functional signal with clean safety in a randomized setting, the regulatory path will clarify; if not, the mechanistic story may remain supportive rather than decisive.
Jon Napitupulu is Director of Media Relations at The Clinical Trial Vanguard. Jon, a computer data scientist, focuses on the latest clinical trial industry news and trends.
