Though progress is being made in an indication with no known cure and only a handful of treatments, the recent wave of high-profile clinical studies in Duchenne muscular dystrophy (DMD) indicates that choosing the right trial endpoint remains an evolving paradigm.

Experts say it is notoriously difficult to evaluate the heterogeneous disease in a clinical setting, as subtle changes are heavily dependent on the stage of the disease. This, they say, is all the more reason to double down on choosing the right endpoint to evaluate any improvements.

Companies have taken different approaches to tackle this. While Capricor Therapeutics succeeded in demonstrating positive three-year data with the Performance of Upper Limb (PUL 2.0) assessment in the HOPE-2 open label extension study of the cell therapy CAP-1002, big name developers such as Pfizer and Sarepta Therapeutics notably failed to demonstrate improvements using the North Star Ambulatory Assessment (NSAA).

In June 2024, Pfizer shared that the pivotal Phase III study of its gene therapy, fordadistrogene movaparvovec, failed to demonstrate superiority against placebo when evaluated against the NSAA endpoint. In October 2023, Sarepta Therapeutics’ Elevidys (delandistrogene moxeparvovec) also failed its Phase III study in the same patient group, using NSAA as a primary endpoint. However, Elevidys has an accelerated approval in a certain patient population, which is dependent on results from a confirmatory study.

There have been “vogues” of different functional measures over time, where each outcome measure stood under the limelight for being more sensitive to change than its predecessor before being snubbed for the next popular measure, says Dr. John Brandsema, pediatric neurologist at the Children’s Hospital of Philadelphia in Pennsylvania.

Know your target population

In terms of designing outcome measures for DMD clinical trials, the key challenge is disease heterogeneity, says Brandsema.

Moreover, the condition affects cognitive functioning in a significant number of patients where cooperation with functional testing is further impaired, he explains.

Given the disease heterogeneity, the solution is to select a sensitive endpoint and build a study cohort around the endpoint, says Brandsema. Where the patient is on the ambulatory spectrum determines which endpoint should be selected for a clinical trial, adds Dr. Craig McDonald, a professor in the Departments of Paediatrics and Physical Medicine & Rehabilitation at the University of California, Davis.

When evaluating a very young ambulatory patient population, time to rise from the floor (TRF), the 4-stair climb test, and the 10-meter walk/run test are often employed as functional endpoints, as these functions are lost in “a very stereotypic, predictable fashion,” and in successive order, says McDonald.

Sarepta employed TRF and the 10-meter walk test as secondary endpoints in the Phase III EMBARK study, which missed its NSAA primary endpoint, while Milan, Italy-based ITF Therapeutics evaluated participants on the 4-stair climb test at 18 months in its Phase III study (NCT02851797) of Duvyzat (givinostat). Duvyzat was awarded FDA approval in March 2024. Brandsema and McDonald agree that building a study population around an endpoint is possible because the field has data from natural history studies and placebo arms of interventional trials. That data provides prognostic insight on the natural trajectory of the disease in a given individual based on their age, genotype, and functional status from when they are first assessed, says Brandsema.

“Once the 10-meter walk/run test value exceeds 10 seconds, we know that in natural history data, 100% of Duchenne patients will lose the ability to ambulate within the next two years. Similarly, once TRF exceeds five seconds, that predictably identifies a patient is truly in the functional decline phase of the disease,” says McDonald.

This is when functional endpoints such as performance of upper limb (PUL) scale should be employed, he adds. The field gravitates towards using measures like PUL or pulmonary spirometry measures in a non-ambulatory population. And PUL can be used in the transition between the late ambulatory stage and the non-ambulatory stage of DMD, explains McDonald.

However, the reliance on existing natural history data for the development of gene therapies needs to be cautioned, as these studies are enrolling patients who have not been thoroughly evaluated, says Tina Duong, MPT, PhD, director of Clinical Outcomes Research & Development, Neurology division at Stanford University. The standard of care for DMD has changed and the disease progression after treatments like steroids cannot be compared to historical cases without steroid use. Duong is a practicing physical therapist with expertise in conducting neuromuscular clinical research.

Using NSAA

Though NSAA has been used across several pivotal studies, Pfizer and Sarepta’s pivotal trials failed to show the statistically significant improvements needed to mark success with this measure.

It is the responsibility of the sponsor to conduct the necessary due diligence and select a de-risked patient population, says Duong. Patients under the age of seven years are more challenging to evaluate because they are younger and demonstrate inconsistent performance, says Duong.

The exact progression of developmental or milestone skills is relatively unknown in DMD patients as natural history studies have evaluated patients over the age of seven as opposed to under. In their pivotal studies, both Pfizer and Sarepta enrolled boys between the ages of 4–7 years old. Assuming that the NSAA would be applicable in a much younger patient population was a pitfall of the failed Phase III studies, says Duong.

Though NSAA is a good option for demonstrating efficacy in a long-term study of 18 months, it is not sensitive or granular enough to be employed in a 12-month study, says McDonald. The field is realizing that it is difficult to use NSAA in short-term trials to demonstrate treatment efficacy, while assessments like time-function tests are delivering efficacious results, he elaborates.

Emerging endpoints

The six-minute walk test (6MWT) was used for many trials about 10 years ago despite issues with patient cooperation, says Brandsema. Time-function tests, though useful and widely employed, lack the granularity that longer tests like 6MWT or a multi-category test like the NSAA have in terms of being more robust in identifying the small variances that occur on the day of testing. Time-function tests include 6MWT, TRF, and the 10-meter walk/run test, among others. Though NSAA is robust as a multi-category test, the assessment is still susceptible to the challenges of patient cooperation, Brandsema explains.

Natural history studies indicate that time-function tests are more responsive to treatment effect and more applicable to the 4–7 year old treatment group in DMD clinical studies, says McDonald. As patients under seven years of age are more challenging to evaluate, the field must also implement more frequent and consistent outcome evaluations that account for daily performance variabilities with the patient, says Duong.

“A new study design may look like a complement of some kind of real-world data collection, whether it be a wearable or a video remote measure that has sound psychometrics,” explains Duong.

Once regulatory agencies begin considering markers like dystrophin expression on muscle biopsy to extrapolate efficacy, they can be used in earlier phases of development, particularly in Phase II studies, says Brandsema. The challenge with using a biomarker-driven endpoint in a Phase III study is getting a signal that is clinically meaningful and translatable to the preservation of motor function, an outcome that is important to patients and families, he adds.

Though dystrophin evaluations are an effective surrogate biomarker, the dystrophin protein is not a contractile protein, meaning it is not involved in improving function; rather, it is important for muscle preservation, says McDonald. Stride Velocity 95^th Centile (SV95C) is emerging as a strong functional endpoint, while an effective secondary endpoint could be skeletal muscle magnetic resonance imaging (MRI) for fat fraction, McDonald elaborates.

"DMD clinical studies evolve with emphasis on building designs around relevant endpoints " was originally created and published by Clinical Trials Arena, a GlobalData owned brand.

The information on this site has been included in good faith for general informational purposes only. It is not intended to amount to advice on which you should rely, and we give no representation, warranty or guarantee, whether express or implied as to its accuracy or completeness. You must obtain professional or specialist advice before taking, or refraining from, any action on the basis of the content on our site.