Abstract
BACKGROUND: It is unclear how changes in quantitative muscle magnetic resonance imaging (MRI) relate to changes in clinical outcome in facioscapulohumeral muscular dystrophy (FSHD), although this information is crucial for optimal use of MRI as imaging biomarker in trials. We therefore assessed muscle MRI and clinical outcome measures in a large longitudinal prospective cohort study. METHODS: All patients were assessed by MRI at baseline and at 5-year follow-up, employing 2pt-Dixon and turbo inversion recovery magnitude (TIRM) sequences, after which fat fraction and TIRM positivity of 19 leg muscles were determined bilaterally. The MRI compound score (CoS) was defined as the mean fat fraction of all muscles weighted for cross-sectional area. Clinical outcome measures included the Ricci-score, FSHD clinical score (FSHD-CS), MRC sumscore (MRC-SS), and motor-function-measure (MFM). RESULTS: We included 105 FSHD patients [mean age 54 ± 14 years, median Ricci-score 7 (range 0-10)]. The median change over 5 years' time in the MRI-CoS was 2.0% (range -4.6 to +12.1; P < 0.001). The median change over 5 years' time in clinical outcome measures was small in all measures, with z-scores ranging from 5.0 to 7.2 (P < 0.001). The change in MRI-CoS correlated with change in FSHD-CS and Ricci-score (ρ = 0.25, respectively; ρ = 0.23, P < 0.05). The largest median increase in MRI-CoS was seen in baseline subgroups with an MRI-CoS 20-40% (6.1%), with ≥2 TIRM positive muscles (3.5%) or with an FSHD-CS 5-10 (3.1%). CONCLUSIONS: This 5-year study showed significant changes in MRI and clinical outcome measures and a significant correlation between changes in MRI-CoS and changes in clinical outcome measures. In addition, we identified subgroups of patients that are most prone to radiological disease progression. This knowledge further establishes quantitative MRI parameters as prognostic biomarkers in FSHD and as efficacy biomarkers in upcoming clinical trials.