Outcomes misaligned in mitochondrial encephalomyopathy, lactic acidosis, and stroke-like episodes (MELAS): implications for trial design

The m.3243A>G variant in the MT-TL gene is the most prevalent pathogenic variant in mitochondrial (mt) DNA in adults, associated with a wide clinical spectrum from asymptomatic individuals to Mitochondrial encephalomyopathy, lactic acidosis, and stroke-like episodes (MELAS) syndrome. Although pharmacological trials in mitochondrial disorders are increasing, the lack of validated endpoints remains a significant barrier to therapeutic development. This cross-sectional observational study aimed to evaluate patients with and without MELAS syndrome to identify factors associated with disease burden. Seventeen individuals genetically confirmed to harbour the heteroplasmic m.3243A>G pathogenic variant were enrolled: six who met the consensus-based diagnostic criteria for MELAS syndrome (median age: 30.0 (interquartile range [IQR]: 29.3–45.0) years). Ten patients who did not have a previous history of stroke-like episodes were assigned as ‘non-MELAS’ (age: 37.5 (32.8–48.3) years). Of these patients in the non-MELAS group, seven exhibited variable features of mitochondrial disease, including hearing loss, diabetes mellitus, migraine, and gastro-intestinal involvement, while the remaining three were asymptomatic. One patient was excluded from analysis due to a confirmed ischemic stroke unrelated to mitochondrial disease. Assessments included disease severity (Newcastle Mitochondrial Disease Adult Scale [NMDAS]) and patient-reported outcomes (PROs) of fatigue (Fatigue Impact Scale [FIS]), health-related quality of life (QoL) (Newcastle Mitochondrial-QoL [NMQ]), mental well-being (Warwick–Edinburgh Mental Wellbeing Scale [WEMWBS]), autonomic symptoms (The Composite Autonomic Symptom [COMPASS 31]), and physical activity (The International Physical Activity Questionnaire [IPAQ]. Performance outcomes included Timed-Up and Go, handgrip strength, cardiopulmonary exercise testing, and accelerometry. Age- and sex-matched healthy controls were included for comparison of accelerometry data (age: 35.5 (28.8–50.5) years). Despite comparable age and mtDNA heteroplasmy, patients with MELAS syndrome had significantly higher disease burden, reduced exercise capacity, and lower levels of objectively measured physical activity compared to non-MELAS and controls (P < 0.05–0.001). PROs did not significantly differ between MELAS syndrome/non-MELAS. While non-MELAS patients showed expected alignment between perceived and objective measures, MELAS syndrome patients demonstrated weak, absent, or paradoxical associations. This mismatch may reflect altered symptom perception, cognitive impairment, or disease-related adaptation. These findings underscore the complexity of disease expression in MELAS syndrome. Regulatory agencies encourage the inclusion of patient-centred endpoints; however, this study highlights the potential limitations of relying solely on PROs. The divergence between subjective and objective assessments supports the need for multidimensional outcomes that integrate both patient perspectives and objective measures to enhance the reliability and interpretability of clinical trials in primary mitochondrial disease.