By Mindy Leffler
August 20, 2026
Five years ago, my husband and I moved our son into a college dormitory. To the casual observer, it was a milestone shared by thousands of families across the country—the standard transition of an 18-year-old into adulthood. But for us, the moment was heavy with uncertainty. Our son lives with Duchenne muscular dystrophy (DMD), a progressive, devastating condition that strips away muscle function. At the time, he was unable to walk, and due to his waning arm strength, we all quietly feared that his college independence would be short-lived.
Today, we are preparing to move him into that same dorm for his second year of master’s degree studies. He remains independent, vibrant, and focused on his future. For the past five years, his life has been sustained by Capricor Therapeutics’ experimental therapy, deramiocel. However, last week, that same drug faced a sobering reality check: a negative FDA advisory committee vote regarding its secondary outcome—the ability to stabilize heart function across a broad patient population.
The vote highlighted the "noisy" nature of data in rare disease trials, particularly when assessing patients whose cardiac health remains stable versus those already exhibiting signs of decline. This event serves as a microcosm of a much larger, more difficult conversation: How do we design clinical trials for rare diseases when every decision is a high-stakes gamble, and how can the FDA balance the rigidity of scientific pre-specification with the urgent, human need for regulatory flexibility?
The Anatomy of a Clinical Trial: A Cascade of "Tiny Decisions"
Designing a study for a rare disease is not a linear path; it is a series of hundreds of minute, complex, and often irreversible decisions. Sponsors operate in an environment where they are effectively blazing a trail, often without the luxury of established historical data to predict how a specific methodology will impact the final results.
When we talk about the "science" of a trial, we are often talking about the trade-offs. For instance, consider the challenge of inclusion criteria. If a sponsor narrows these criteria to focus on a specific, homogeneous population, they increase the probability of achieving statistical significance. However, this creates an enormous risk regarding enrollment; if the pool of eligible patients is too small, the study may never reach the numbers required to prove anything at all. In the rare disease space, this "enrollment risk" can jeopardize an entire program, effectively ending a potential treatment’s journey before it even reaches the finish line.
In the case of the Phase 3 study for deramiocel, the primary outcome focused on the performance of the upper limb—a vital metric for the quality of life in Duchenne patients. This measurement is divided into three domains: shoulder, arm, and hand function. A researcher can either look at the "total measure" or isolate an individual domain.
The total measure is attractive because it captures a wide range of tasks, potentially reflecting meaningful changes in both the "strongest" participants (who may struggle with shoulder tasks) and the "weakest" (who may struggle with hand tasks). Yet, this approach introduces "noise." If a significant portion of the cohort has already lost shoulder function, or if another portion is projected to maintain hand function regardless of treatment, the total data becomes cluttered. One could pivot to focusing only on arm tasks to sharpen statistical significance, but in doing so, do we risk ignoring the meaningful changes occurring in the participants at the extreme ends of the spectrum? These are the dilemmas that researchers—and the families who rely on them—face every single day.
Chronology of a Decade-Long Struggle
The Duchenne community has been in the trenches of clinical trial development for over a decade. We have watched as methodologies evolved, failed, and improved.
- Early Research (2015-2017): The community began pushing for standardized outcome measures, recognizing that without them, comparing results across trials would be impossible.
- Methodological Shift (2018-2021): The adoption of more nuanced metrics, such as the Duchenne Video Assessment, allowed researchers to move beyond traditional, less sensitive testing, capturing subtle improvements in limb function that truly mattered to patients’ daily lives.
- The Phase 3 Era (2022-2025): Capricor initiated its Phase 3 trials for deramiocel. During this period, the focus moved from simple survival to the preservation of functional independence.
- The Advisory Committee Pivot (July 2026): The FDA advisory committee convened to review the data. While the primary endpoint data showed promise, the secondary outcome regarding heart function created enough statistical noise to lead to a negative recommendation.
- The Regulatory Response (August 2026): In a move that signaled hope for the community, the FDA invited Capricor to submit additional analyses and data to clarify the primary outcome. This decision recognized that the "noise" in the secondary outcome should not necessarily invalidate the efficacy seen in the primary function metrics.
Supporting Data: Why "Noise" is Not Failure
The critique of the deramiocel data stems from the difficulty of measuring stabilization in a heterogeneous population. In clinical trials, "statistical significance" is a blunt instrument. When you group patients with varying levels of cardiac health, those who are already stable may not show a "statistically significant" improvement compared to the placebo group. This creates a statistical "drag" on the total data set, masking the genuine benefit felt by those whose disease was progressing more rapidly.
This is the central conflict of modern drug development. If we demand that every subgroup shows a perfect, uniform response, we will almost never reach statistical significance in a rare disease population. The data will always appear "noisy" because the biology of human disease is inherently messy. The question is whether we are willing to apply rigorous, modern statistical methods to "clean" this data, or if we will continue to rely on antiquated, broad-stroke requirements that ignore the individual patient’s reality.
Official Responses and Regulatory Philosophy
The Food and Drug Administration (FDA) is tasked with an unenviable role: they must act as the ultimate arbiter of safety and efficacy while facing immense pressure from two sides. On one side are the proponents of strict, pre-specified scientific rigor—the view that if a trial does not meet its pre-defined statistical goals, it cannot be considered a success. On the other side are the patient advocates who argue that the "unmet need" is so profound that the FDA must be willing to exercise flexibility, particularly when the data shows a clear trend of benefit.
Last week, the FDA’s decision to allow Capricor to submit further analyses on the upper limb data was a profound moment of regulatory maturation. By opting for a dialogue rather than a final rejection, the agency demonstrated a willingness to weigh the clinical reality—the actual lives and function preserved—against the strict adherence to a pre-specified secondary outcome. They recognized that the time required for a completely new trial would cost families years of progress and, in some cases, the window of opportunity for the patient to remain functional.
Implications: The Future of Rare Disease Research
The implications of this decision stretch far beyond Duchenne muscular dystrophy. If the FDA continues to embrace this "art of analysis," it sets a precedent for other rare diseases. It moves the conversation away from "Did you hit the P-value on every single metric?" toward "What is this treatment actually doing for the patient?"
For my family, this is not a theoretical debate. It is the story of my son’s ability to feed himself, to operate his computer, and to pursue his master’s degree. When regulators sit down to navigate the totality of these tiny decisions, they are not just looking at spreadsheets. They are looking at the potential to halt the decline of young men who are fighting for every inch of independence.
My hope for the future is that the guiding principle for the FDA remains the pursuit of the truth, even when that truth is complex. We need a regulatory framework that encourages the use of rigorous, sophisticated methods to understand what is truly happening to patients on treatment. We need to acknowledge that in the world of rare disease, we are all learning as we go.
Regulatory flexibility, when paired with scientific rigor, does not mean lowering the bar. It means acknowledging that the bar must be measured in a way that respects the reality of the disease. My son’s future—and the futures of thousands of others—depends on the agency’s ability to keep that balance, ensuring that in our search for the perfect study, we don’t lose the very people we are trying to save.
