Review Finds Progress in Mitochondrial Disease Trials Where Drug, Genotype, and Endpoint Align

Key Takeaways

• A review in Biomedicine & Pharmacotherapy surveyed 159 ClinicalTrials.gov studies of primary mitochondrial diseases (PMDs), including 110 interventional trials, and concludes that drug development may be at a turning point.
• Four therapies have regulatory authorization: idebenone for LHON, taurine for MELAS, and the recent FDA approvals of Kygevvi (doxecitine and doxribtimine) for thymidine kinase 2 (TK2) deficiency and FORZINITY (elamipretide) for Barth syndrome.
• The authors say these successes came where mechanism, patient genotype, and endpoints were tightly aligned, while broad-population trials of plausible drugs often failed. They call the approvals proof of principle, not evidence that the broader challenges are solved.

Li-Tzu Wang and Han-Ying Jhuang of Taipei Medical University review more than two decades of clinical development in PMDs, a group of genetic disorders caused by variants in mitochondrial or nuclear DNA that impair cellular energy production. They note that progress has been limited by extreme genetic and clinical heterogeneity, small and geographically scattered patient populations, and a lack of validated pharmacodynamic biomarkers and disease-specific endpoints.

The authors searched ClinicalTrials.gov for studies registered from January 2010 to April 2026, identifying 159 studies after deduplication. Of these, 110 were interventional, and 37 programs met criteria for detailed review because they tested a defined pharmacological, nucleoside, gene-based, or cellular intervention in a confirmed PMD population. Twenty-five had a clear pharmacological or gene-based mechanism. Trial activity has grown, with 53 interventional trials registered for specific PMD subtypes from 2020 to 2026, compared with 49 in the whole of 2010 to 2019, and about 50 studies now active or recruiting. Trials starting in 2020 or later were also more likely to use randomized designs, genotype-restricted enrollment, patient-reported primary endpoints, biomarker secondary endpoints, and adaptive or crossover designs. The authors describe these percentages as descriptive syntheses, not a formal meta-analysis.

The approvals and failures show a pattern. Elamipretide missed both co-primary endpoints in the Phase 3 MMPOWER-3 trial in broad primary mitochondrial myopathy, yet earned FDA accelerated approval for Barth syndrome, where its target, cardiolipin, is central to the disease. Kygevvi was approved for TK2 deficiency on pooled evidence from one ongoing Phase 2 trial, two retrospective chart reviews, and an expanded-access program, without a conventional placebo-controlled trial. Taurine, approved in Japan for MELAS, showed a reduction in stroke-like episodes in a 10-patient open-label trial. Among the disappointments, mavodelpar failed both endpoints in the 213-patient STRIDE trial, in which the placebo group's walking distance improved substantially. Vatiquinone missed its seizure endpoint in a mixed basket of syndromes. The gene therapy lenadogene nolparvovec, for LHON, never won authorization after the European Medicines Agency reviewed trials that showed no statistically significant visual benefit over sham and improvement in both treated and untreated eyes. Newer strategies in earlier stages include NAD⁺ augmentation, soluble guanylate cyclase stimulation with zagociguat, mTOR inhibition with sirolimus, and heteroplasmy-targeting nucleases that so far remain preclinical.

The authors acknowledge that the review draws on a single registry and a curated subset of programs, so trials registered elsewhere may be missing, and that the field moves quickly, so findings after April 2026 aren't captured. They conclude that the next phase depends less on adding mechanisms than on improving trial design, with tighter molecular stratification, endpoints matched to the disease domain most likely to change, strong natural-history data, and careful use of biomarkers. They also stress that conditions such as Leigh syndrome, MELAS, and primary mitochondrial myopathy shouldn't be expected to follow a single development template.