Background: Duchenne muscular dystrophy (DMD) is a progressive X-linked disorder that leads to loss of ambulation and reduced life expectancy in males. Magnetic resonance imaging (MRI) biomarkers such as fat fraction and T2 relaxation time offer quantitative, noninvasive measures of disease progression, yet their adoption in therapeutic trials has been inconsistent.
Objective: To assess trends in MRI biomarker use across DMD clinical and observational studies and to identify predictors of their adoption using machine learning (XGBoost).
Methods: A systematic review of 85 DMD studies (2010–2022) was conducted to evaluate trial design, intervention type, duration, and endpoint selection. MRI and functional endpoint usage were compared, and an XGBoost model was applied to identify variables most predictive of imaging biomarker inclusion.
Results: Studies incorporating imaging biomarkers lasted on average 4.5 years, approximately 11 months longer than trials using only functional endpoints. Among biologic intervention trials (n=28), 93% included ambulatory endpoints, while only 13.3% (n=4) used imaging biomarkers. Small molecule trials and natural history studies each accounted for about 30% of imaging biomarker use. After the 2018 FDA guidance on DMD drug development, new trials including imaging biomarkers declined despite prior regulatory encouragement. The XGBoost model identified trial duration and start year as the strongest predictors of biomarker inclusion.
Conclusions: Although MRI biomarkers can objectively quantify disease progression, their use in DMD interventional trials remains limited. The decline in adoption following the 2018 FDA guidance suggests persistent regulatory and logistical barriers. Standardization of MRI acquisition and analysis, alongside hybrid functional–biomarker endpoints, may facilitate broader clinical and regulatory acceptance.
Disclosure: This abstract represents an encore presentation of previously published work titled “Use of imaging biomarkers and ambulatory functional endpoints in Duchenne muscular dystrophy clinical trials: Systematic review and machine learning-driven trend analysis” by Todd et al., published online in the Journal of Neuromuscular Diseases on July 29, 2025. It will also be presented at the University of Florida, College of Pharmacy’s Annual Research Showcase in Gainesville, FL, on February 9, 2026. The content has been adapted for presentation at the 2026 MDA Conference.