Clinical trials in DMD: Ten years on, what have we learned? How can we optimize future trial design?

Location: Hoofddorp, The Netherlands, with virtual participation of a few participants

Title: 269th ENMC International Workshop on: Clinical trials in DMD: Ten years on, what have we learned? How can we optimize future trial design?

Organisers: Prof N. Goemans (Belgium), Prof C. McDonald (US), Prof. E. Mercuri (Italy) and Prof F. Muntoni (UK)

Participants: Francesco Muntoni (UK); Nathalie Goemans (Belgium); Craig McDonald (USA, via web-link); Eugenio Mercuri (Italy); Jean-Yves Hogrel (France); Michela Guglieri (UK); Laurent Servais (UK); Tina Duong (USA); Suzanne Hendrix (USA); Erik Henricson (USA); Terina Martinez (USA); James Signorovitch (USA); Susan Ward (USA); Elizabeth Vroom (The Netherlands); Patricia Furlong (USA); Mindy Leffler (USA); Matthew Klein (USA, via web-link); Michael Binks (USA); Paolo Bettica (Italy); Ana Ferreiro (France); Karin Naarding (The Netherlands); Georgia Stimpson (UK); Erik Niks (The Netherlands); Frank Van Ieperen (The Netherlands).

Translations of this report by:

Italian: Paolo Bettica
Dutch: Karin Naarding
French: Gérald Perret

Description and aims of the workshop The 269^th ENMC workshop was held from the 9^th to the 11^th of December 2022 and brought together 24 representatives of patients, advocacy groups, industry, and neuromuscular and clinical experts from 5 European countries and from the United States.

This workshop was organized to discuss participants’ experiences of clinical trials in Duchenne muscular dystrophy (DMD) over the last decade, both those that failed to meet the expected clinical endpoints and also recent successes. The workshop did not consider aspects related to trials failed due to toxicological issues but was focused on learning from the clinical trial design, patient recruitment and execution to mitigate risks for future studies and improve clinical trial success rate for patients with DMD.

The first part of the workshop was focused on the improved understanding of the variability in disease progression rates in DMD. This mix of trajectories results in high variance in clinical trials. Drug companies have sought to reduce variance by using strict inclusion and exclusion criteria to enroll patients who are in a more predictable disease phase: slowly declining but still able to walk. In the last few years, prognostic models have proved to be important in accounting for variance in 48-week trials, and this has also allowed researchers to compare the rate of disease progression in patients involved in several placebo-controlled studies to the rate of progression observed in multiple Real World Data (RWD) cohorts. This has led to the success that in recent trials a limited and similar number of patients in placebo and treatment arm lost ambulation, thus most patients could contribute to the primary ambulatory endpoint of the study.

Moreover, by considering a combination of functional measures, use of steroids, height, weight, and BMI at baseline, it is possible to explain up to 40% of the variability in disease progression rates at one year. Understanding these prognostic factors can guide clinical trial design and can increase power when applied to analysis of trial results.

Another part of the discussion was focused on the role of RWD in clinical trials for DMD. Data from RWD models can add value in different ways, such as in predicting the likely outcomes of the trial via simulations or in the systematic collection and storing of data to better understand long-term drug efficacy. High quality RWD, if appropriately curated and ideally also qualified, could also be used to serve to enrich the placebo arm of RCT studies.

Sponsors representatives discussed their experiences, and hurdles, in the execution of clinical trials. An important point that was brought up relates to the fact that regulators will often require that in order for a clinical trial to be considered successful, it must not only meet its primary endpoint but also the various secondary endpoints. While consistency between the primary and secondary endpoints strengthens evidence of a drug working, as different endpoints can follow different trajectories, it may be that not all endpoints are as responsive to treatment. There was concern that the potential discordance of primary and secondary trial outcomes could prove a problem in future trials. Composite endpoints were discussed as a potential solution for this, which will be investigated further.

One session was devoted to the definition of the minimal clinically important difference and the various methodologies that are recommended to detect this. There was emphasis in this session that a patient experience-centered approach is always preferable. The discussion also focused on the minimal detectable change that can be observed for multiple endpoints and how to take into account these different measures.

An interesting session was devoted to novel concepts in measuring disease progression.

Experts from another field of medicine (Alzheimer disease) brought up their experience using the concept of “time saving”, in which changes in disease progression are transformed into time savings for a given treatment period. In this way, true disease progression can be measured against time as a gold standard, and this brings concepts that are more easily appreciated by the regulators, the payers and the patients.

More specific DMD presentations focusing on measuring DMD disease progression and time to events using readily available clinical parameters, such as the impact of time to rise from the floor on subsequent loss of ambulation. This information can be used to prepare a prognostic score of disease progression that could be used as endpoint in clinical trials.

Novel outcome measures were also discussed, ranging from the validity and reliability of the Duchenne video assessment, to provide additional information on the quality of the execution of different manoeuvres, which is not captured by the current scale system; and the recent success of qualifying maximal stride velocity (using accelerometer devices) as outcome measure for clinical trials.

Members of the advocacy groups and parents of boys with DMD stressed the burden of clinical trials, especially when frequent visits to the treatment centers are required. In some circumstances, this can lead to disenfranchising the involvement of the local care centers, with some confusion of responsibilities between the clinical care team and the experimental therapy center. The recommendation of the advocacy groups is to have a clear care pathway document at the inception of the clinical trial were discussed. Concerns regarding potentially biased recruitment strategies were also voiced, followed by recommendations for robust recruitment standard operative procedures.

A strongly-voiced concern relates to the duration of placebo-controlled studies, in which DMD boys could be randomized for 12, 18 or even 24 months to receive a placebo. Moreover, there was concern over the number of muscle biopsies that are required of children in trials, particularly at baseline and including in those in the placebo arms.

There was additional discussion on strategies that could ensure that rapidly deteriorating children could be switched from the placebo to the active treatment arm, a strategy called a rescue protocol. Similar strategies are in use in other diseases, but they depend on the trial outcome measure being objective. Deterioration of muscle MRI was suggested as a potential objective measure, and this will require further discussion. As for the need to have a baseline muscle biopsy, which is typically required in dystrophin restoration clinical trials, the participants suggested that it would be very valuable if previously collected muscle biopsy data from multiple trials could be collated and analysed. This data on the proportion of children with baseline dystrophin levels below the lower limit of detection could then be presented for discussion to regulators.

Finally, several investigators raised concerns over operational issues related to clinical trial data acquisition. In particular it was felt that data storage is not always optimally aligned with the reality of the data capture on the ground, which lead to errors in data reporting and further work verifying individual inconsistencies.

As specific deliverables of this workshop, the participants agreed to take forward discussions on a number of items that would benefit from further alignment between the different academic centers and registries (for example patient reported outcome measures, currently not systematically used in most of the RWD efforts), and on how to engage regulators to obtain guidance on specific topics that were prioritized during the discussion. These efforts will be pursued in the course of 2023.

A full report of the conference is published in Neuromuscular Disorders (PDF).