Assessing a Mitochondrial Disease Treatment via a Novel Statistical Technique for Accelerometer Data

Therapeutic development for mitochondrial diseases, rare genetic disorders with pathogenic defects of oxidative phosphorylation, is hindered by unsatisfactory outcome measures. To address this problem, we provide the first clinical application of a novel, bias-adjusted outcome measure of acceleration across a range of subjects' activities to assess nucleoside therapy for thymidine kinase 2 deficiency, an ultra-rare autosomal recessive mitochondrial disease.

Data were collected from treated patients in an ongoing phase 2 clinical trial who served as their own controls. If there is a treatment effect, time-in-activity curves for these patients will increase over successive clinic visits. We used a combination of functional data analysis and longitudinal mixed-effects linear regression, adjusted for age and gender, to test for the effect of treatment length on time-in-activity.

For 14 patients with at least two assessments 6 months apart, we found a significant overall improvement of time-in-activity due to treatment. Improvement was especially significant at two individual activity levels within the range (0.14 and 2 g). In longitudinal analyses, using data on time-in-activity at these two levels for all clinic visits of 19 subjects, the effect of treatment length on time-in-activity was highly significant at both 0.14 g (0.04, CI 0.01-0.08, p = 0.023) and 2 g (0.01, 0.00-0.02, p = 0.013).

This small-N exploratory analysis using a new accelerometer-based activity measure featuring powerful data reduction and adjustment for circadian rhythms and other biases finds that nucleoside therapy may increase activity levels in thymidine kinase 2 deficiency patients.