Krista Vandenborne

Given heterogeneous progression of Duchenne muscular dystrophy (DMD), accurately characterizing individual disease trajectories is critical for effective clinical trials and therapeutic development. In this study, we applied machine learning (ML) approaches leveraging disease progression models and baseline magnetic resonance (MR)‐derived radiomic features to quantify individual‐level DMD progression trajectories and support quantitative characterization of disease progression. Radiomic features were extracted from segmented soleus (SOL) and vastus lateralis (VL) fat‐fraction maps, generating 810 features per individual, with additional functional and demographic measures incorporated as covariates following feature selection. Multiple ML regressors were trained and validated to estimate MR‐spectroscopy fat‐fraction trajectory descriptive parameters: DPT ₅₀ (age at half‐maximal increase), DP _max (extent), γ (rate), S ₀ (extrapolated measure at age 0). Key radiomic features were identified through feature importance analyses, followed by downstream analyses, including unsupervised clustering. The tree‐based XGBoost regression model demonstrated best future longitudinal fat‐fraction trajectories (SOL: mean absolute error (MAE) = 0.04, R ² = 0.94; VL: MAE = 0.06, R ² = 0.95). Subsequent downstream analyses highlighted potential key biomarkers clinically associated with DMD progression. Clustering analysis revealed clinically meaningful slow‐ and fast‐progression subgroups with interpretable decision criteria (SOL: DPT ₅₀ = 12.21 vs. 14.83; γ = 8.07 vs. 7.54, VL: 10.37 vs. 14.21; γ = 7.72 vs. 7.09) that can inform clinical trial designs and therapeutic effects within homogenous groups. Individual‐level feature importance interpretation further shows the possibility of tailoring personalized intervention strategies. These findings highlight the potential of ML‐based trajectory modeling approaches leveraging baseline MRI radiomics for advancing AI‐assisted clinical trial development and precision medicine, charting a new direction for MRI‐guided disease modeling in neuromuscular disorders.

Introduction: Myotonic dystrophy type 1 is the most prevalent muscular dystrophy in adults, characterized by weakness, impaired functional abilities, and myotonia. However, little is known about the relationship between quantitative MRI measures (fat fraction and T₂ relaxation time) and clinical findings of the upper extremity. This study assessed forearm muscle structure in patients with myotonic dystrophy using quantitative MRI and correlated these measures with strength, function, and handgrip myotonia. Materials and Methods: Eighteen adults with myotonic dystrophy type 1 underwent MRI using three-point Dixon and T₂ spin echo imaging of the forearm. Results: The average fat fraction and T₂ relaxation time were greatest in the flexor digitorum profundus (26.7% and 55.6 ms, respectively). Correlations were found between quantitative MRI values and clinical tests of strength (r = −0.61 to −0.92, p < 0.01), function (r = −0.64 to −0.83, p < 0.01), and handgrip myotonia (r = 0.48, p < 0.05). Overall, the anterior forearm fat fraction values showed higher correlations with strength and function compared to those of the posterior forearm. Discussion: Our results support the use of quantitative MRI measures to assess forearm disease pathology and show potential to monitor the effectiveness of therapeutic treatments in patients with myotonic dystrophy type 1.

Importance: Delandistrogene moxeparvovec is a recombinant adeno-associated virus rhesus isolate serotype 74 vector-based gene transfer therapy for the treatment of Duchenne muscular dystrophy (DMD) in patients with a confirmed pathogenic variant of the DMD gene. In a subset of patients in the EMBARK (A Gene Transfer Therapy Study to Evaluate the Safety and Efficacy of Delandistrogene Moxeparvovec [SRP-9001] in Participants With DMD) randomized clinical trial, changes in muscle health and pathology were assessed to evaluate the therapeutic impact of the treatment on disease progression.Objective: To determine the effect of delandistrogene moxeparvovec on muscle quantitative magnetic resonance (QMR) measures of disease progression in patients in the EMBARK trial.Design, Setting, and Participants: This was a phase 3, double-blind, placebo-controlled (October 2021-September 2023; week 52 cutoff date: September 13, 2023), multicenter randomized clinical trial that included 131 patients. Patients were randomized, and 125 were treated with either delandistrogene moxeparvovec (n = 63) or placebo (n = 62). The current study focused on a subset of patients who underwent muscle QMR imaging.Intervention: Single-administration intravenous delandistrogene moxeparvovec (1.33 × 10¹⁴ vector genome/kg) or placebo.Main Outcomes and Measures: Change from baseline to week 52 in muscle MR was a prespecified exploratory end point. Proton MR spectroscopy (MRS) and 8-point Dixon MR imaging (MRI) measured muscle fat fraction (FF); multislice spin echo MRI measured transverse relaxation time (T₂). MRS FF was measured in the soleus and vastus lateralis. MRI FF and T₂ were measured in 5 leg muscle locations important for ambulation. A post hoc global statistical test combining all muscles and modalities assessed overall treatment effect.Results: In this exploratory EMBARK analysis, 39 male participants (delandistrogene moxeparvovec, n = 19; placebo, n = 20; mean [SD] age, 6.10 [1.04] years; mean [SD] baseline North Star Ambulatory Assessment total score, 22.99 [3.71] points) underwent muscle MRI. Treated patients showed less disease progression vs placebo on MR measures. Across muscles and modalities, magnitudes of FF change favored delandistrogene moxeparvovec; between-group differences in least-squares mean change ranged from −1.01 (95% CI, −2.79 to 0.77; soleus) to −0.71 (95% CI, −3.21 to 1.80; vastus lateralis) for MRS FF and −3.09 (95% CI, −7.62 to 1.45; vastus lateralis) to −0.44 (95% CI, −4.01 to 3.12; hamstrings) for MRI FF. T₂ reductions (improvements; 4 of 5 muscles) were observed in treated patients vs increases (worsening; all muscles) in placebo patients; within-group differences in least-squares mean change ranged from −1.06 (95% CI, −2.10 to −0.02; soleus) to 0.17 (95% CI, −1.76 to 2.10; biceps femoris) in the delandistrogene moxeparvovec group and from 1.12 (95% CI, 0.08-2.16; soleus) to 2.94 (95% CI, 0.84-5.03; quadriceps) in the placebo group. The global statistical test supported treatment benefit (P = .03).Conclusions and Relevance: Results reveal that QMR outcomes consistently favored delandistrogene moxeparvovec across muscle groups, with treatment leading to decreased fat accumulation and improved T₂ vs placebo over 52 weeks. Consistent with treatment effects on functional outcomes observed in the EMBARK trial, these results suggest stabilization or less progression of muscle pathology with delandistrogene moxeparvovec—adding to the totality of evidence supporting disease stabilization or slowing of disease progression with delandistrogene moxeparvovec.Trial Registration: ClinicalTrials.gov Identifier: NCT05096221

Quantitative model‐based clinical trial simulation tools play a critical role in informing study designs through simulation before actual execution. These tools help drug developers explore various trial scenarios in silico to select a clinical trial design to detect therapeutic effects more efficiently, therefore reducing time, expense, and participants' burden. To increase the usability of the tools, user‐friendly and interactive platforms should be developed to navigate various simulation scenarios. However, developing such tools challenges researchers, requiring expertise in modeling and interface development. This tutorial aims to address this gap by guiding developers in creating tailored R Shiny apps, using an example of a model‐based clinical trial simulation tool that we developed for Duchenne muscular dystrophy (DMD). In this tutorial, the structural framework, essential controllers, and visualization techniques for analysis are described, along with key code examples such as criteria selection and power calculation. A virtual population was created using a machine learning algorithm to enlarge the available sample size to simulate clinical trial scenarios in the presented tool. In addition, external validation of the simulated outputs was conducted using a placebo arm of a recently published DMD trial. This tutorial will be particularly useful for developing clinical trial simulation tools based on DMD progression models for other end points and biomarkers. The presented strategies can also be applied to other diseases.

Evaluations of treatment efficacy in Duchenne muscular dystrophy (DMD), a rare genetic disease that results in progressive muscle wasting, require an understanding of the ‘meaningfulness’ of changes in functional measures. We estimated the minimal detectable change (MDC) for selected motor function measures in ambulatory DMD, i.e., the minimal degree of measured change needed to be confident that true underlying change has occurred rather than transient variation or measurement error. MDC estimates were compared across multiple data sources, representing >1000 DMD patients in clinical trials and real-world clinical practice settings. Included patients were ambulatory, aged ≥4 to 80% confidence in true change were 2.8 units for the North Star Ambulatory Assessment (NSAA) total score, 1.3 seconds for the 4-stair climb (4SC) completion time, 0.36 stairs/second for 4SC velocity and 36.3 meters for the 6-minute walk distance (6MWD). MDC estimates were similar across clinical trial and real-world data sources, and tended to be slightly larger than MCIDs for these measures. The identified thresholds can be used to inform endpoint definitions, or as benchmarks for monitoring individual changes in motor function in ambulatory DMD.

The study aimed to provide quantitative information on the utilization of MRI transverse relaxation time constant (MRI-T₂) of leg muscles in DMD clinical trials by developing multivariate disease progression models of Duchenne muscular dystrophy (DMD) using 6-min walk distance (6MWD) and MRI-T₂. Clinical data were collected from the prospective and longitudinal ImagingNMD study. Disease progression models were developed by a nonlinear mixed-effect modeling approach. Univariate models of 6MWD and MRI-T₂ of five muscles were developed separately. Age at assessment was the time metric. Multivariate models were developed by estimating the correlation of 6MWD and MRI-T₂ model variables. Full model estimation approach for covariate analysis and five-fold cross validation were conducted. Simulations were performed to compare the models and predict the covariate effects on the trajectories of 6MWD and MRI-T₂. Sigmoid I_max and E_max models best captured the profiles of 6MWD and MRI-T₂ over age. Steroid use, baseline 6MWD, and baseline MRI-T₂ were significant covariates. The median age at which 6MWD is half of its maximum decrease in the five models was similar, while the median age at which MRI-T₂ is half of its maximum increase varied depending on the type of muscle. The models connecting 6MWD and MRI-T₂ successfully quantified how individual characteristics alter disease trajectories. The models demonstrate a plausible correlation between 6MWD and MRI-T₂, supporting the use of MRI-T₂. The developed models will guide drug developers in using the MRI-T₂ to most efficient use in DMD clinical trials.

Although regulatory agencies encourage inclusion of imaging biomarkers in clinical trials for Duchenne muscular dystrophy (DMD), industry receives minimal guidance on how to use these biomarkers most beneficially in trials. This study aims to identify the optimal use of muscle fat fraction biomarkers in DMD clinical trials through a quantitative disease‐drug‐trial modeling and simulation approach. We simultaneously developed two multivariate models quantifying the longitudinal associations between 6‐minute walk distance (6MWD) and fat fraction measures from vastus lateralis and soleus muscles. We leveraged the longitudinal individual‐level data collected for 10 years through the ImagingDMD study. Age of the individuals at assessment was chosen as the time metric. After the longitudinal dynamic of each measure was modeled separately, the selected univariate models were combined using correlation parameters. Covariates, including baseline scores of the measures and steroid use, were assessed using the full model approach. The nonlinear mixed‐effects modeling was performed in Monolix. The final models showed reasonable precision of the parameter estimates. Simulation‐based diagnostics and fivefold cross‐validation further showed the model's adequacy. The multivariate models will guide drug developers on using fat fraction assessment most efficiently using available data, including the widely used 6MWD. The models will provide valuable information about how individual characteristics alter disease trajectories. We will extend the multivariate models to incorporate trial design parameters and hypothetical drug effects to inform better clinical trial designs through simulation, which will facilitate the design of clinical trials that are both more inclusive and more conclusive using fat fraction biomarkers.

Microdystrophin gene transfer using recombinant adeno-associated virus serotype rh74 (rAAVrh74) driven by a skeletal and cardiac muscle-specific promoter with enhanced cardiac expression (MHCK7) (SRP-9001) showed promise in an open-label gene transfer study in Duchenne muscular dystrophy (DMD), with robust transgene expression on muscle biopsy (74%-96% of fibers).¹ However, biopsy data reflect a small sample of muscle, whereas muscle quality in large muscle groups can be objectively and noninvasively measured using quantitative magnetic resonance imaging (qMRI) and spectroscopy (qMRS). qMRI and qMRS are powerful tools for monitoring disease progression and therapeutic response in DMD, and they correlate with and identify future loss of ambulatory function.^2-5 We hypothesized that qMRI and qMRS data (muscle fat fraction and bulk MRI transverse relaxation time [qT₂]) would be reduced in 3 boys treated with SRP-9001 compared with an age-matched natural history cohort treated with standard of care and a control group of individuals without DMD.

Duchenne muscular dystrophy (DMD) is a muscle degenerative disorder that manifests in early childhood and results in progressive muscle weakness. Physical therapists have long been an important component of the multidisciplinary team caring for people with DMD, providing expertise in areas of disease assessment, contracture management, assistive device prescription, and exercise prescription. Over the last decade, magnetic resonance imaging of muscles in people with DMD has led to an improved understanding of the muscle pathology underlying the clinical manifestations of DMD. Findings from magnetic resonance imaging (MRI) studies in DMD, paired with the clinical expertise of physical therapists, can help guide research that leads to improved physical therapist care for this unique patient population. The 2 main goals of this perspective article are to (1) summarize muscle pathology and disease progression findings from qualitative and quantitative muscle MRI studies in DMD and (2) link MRI findings of muscle pathology to the clinical manifestations observed by physical therapists with discussion of any potential implications of MRI findings on physical therapy management.

In a multicenter study, MRI and hydrogen 1 MR spectroscopy demonstrated early and progressive muscle involvement of the upper extremities in participants with Duchenne muscular dystrophy. Background Upper extremity MRI and proton MR spectroscopy are increasingly considered to be outcome measures in Duchenne muscular dystrophy (DMD) clinical trials. Purpose To demonstrate the feasibi...