Abstract
OBJECTIVE: Duchenne Muscular Dystrophy (DMD) is characterized by progressive muscle wasting leading to early loss of motor function. Functional tests monitor disease progression and serve as clinical trial endpoints, but are influenced by maturation in younger patients, variability, and patient motivation. Blood biomarkers, that can predict disease progression and objectively evaluate treatment responses, offer a valuable alternative.In this study, we investigated whether longitudinal observations of biomarkers myostatin and the creatine/creatinine ratio (Cr/Crn) are associated with functional tests, such as 6-minute walk test, North Star Ambulatory Assessment (NSAA), 10-meter walk-run test velocity, Performance of Upper Limb (PUL2.0), and disease milestones like loss of ambulation (LoA), overhead reach. and hand-to-mouth function. METHODS: We used real-world longitudinal data from 74 DMD patients followed for up to 11 years with annual visits to the LUMC outpatient clinic, linked to 408 serum samples. Associations between biomarkers, functional tests, and clinical milestones were assessed using linear mixed models and time-dependent Cox models, respectively. RESULTS: Lower Cr/Crn levels and higher myostatin levels were associated with better functional performance and a less rapid decline in ambulation, given fixed treatment and BMI. Children with one-unit higher log2-myostatin levels had, on average, 4.73 points higher NSAA and 3.40 points higher PUL2.0 (both p < 0.001), and were 42% less likely to lose ambulation over the following year. Conversely, children with one-unit lower log2-ratio levels had, on average, 7.18 points higher NSAA and 11.40 points higher PUL2.0 (both p < 0.001), and were 3.67 times more likely to remain ambulant. We also proved that incorporating log2-myostatin and log2-Cr/Crn as endpoints could reduce the required sample size for clinical trials by more than half without compromising statistical power. For instance, to detect a yearly drop of 3 points in the NSAA with 80% power, recruitment requires almost 80 participants in a 1:1 randomized trial, in contrast to a little more than 50 patients for the respective value of log2-myostatin or log2-Cr/Crn. INTERPRETATION: These findings support the potential of myostatin and Cr/Crn as prognostic biomarkers to enhance trial design and endpoint in clinical and interventional trials for DMD.