Exon-Skipping Gene Therapy in DMD: A Surrogate Endpoint, a Contested Functional Benefit
A 7-year-old boy, genetically eligible for a therapy built for his exact mutation. The disagreement isn't about the mechanism — it's about whether the confirmatory trials required by the drug's own accelerated approval have actually shown it helps.
The father of J.M., a 7-year-old boy, has spent the past year watching his son's North Star Ambulatory Assessment score slide downward at each clinic visit — a number that used to feel abstract until it started matching what he sees at home, in how long it takes J.M. to climb the school bus steps now compared with a year ago. Genetic testing confirmed a dystrophin gene deletion amenable to exon 51 skipping, and J.M.'s father has spent the months since reading everything he can find about eteplirsen, the therapy built specifically for boys with his son's exact mutation. J.M. remains ambulatory, on stable deflazacort therapy with normal growth and cardiac function, and insurance prior authorization for the new drug is already in progress.
What eteplirsen's own approval history actually shows is more complicated than "a drug for his exact mutation." The FDA granted accelerated approval based on dystrophin production — the protein DMD's mutation prevents the body from making — increasing to only a small percentage of normal levels on muscle biopsy, a surrogate marker rather than a direct measure of function. PROMOVI, the phase 3 confirmatory study required as a condition of that accelerated approval and reported by McDonald and colleagues in 2021, was open-label, and its own designated untreated comparison cohort was judged an inappropriate control because those boys carried different mutations with a different expected trajectory. Its favorable conclusions therefore rest on post-hoc comparison against external natural-history controls rather than on the randomized confirmation the accelerated-approval pathway was meant to deliver — leaving the functional outcomes families care about, timed function tests, ambulation, North Star scores, genuinely unsettled. J.M.'s father doesn't know any of that yet; he only knows the drug exists and his son is eligible.
A surrogate endpoint against a family's hope
I'd offer eteplirsen. J.M. is genetically eligible, the mechanism is biologically sound — restoring the reading frame to produce a truncated but partially functional dystrophin protein — and open-label extension data show sustained dystrophin production over years of use. This is a fatal, progressive disease with very few disease-modifying options, and the drug's acute toxicity profile is low. Given that combination, I think it's reasonable to offer a plausible-mechanism therapy even with real uncertainty about the size of its functional benefit.
I want to separate two claims that are easy to run together: that the mechanism is biologically plausible, which I agree it is, and that the drug has been shown to functionally help boys like J.M., which the confirmatory trial evidence doesn't clearly establish. Accelerated approval rested on dystrophin production reaching only a small percentage of normal — PROMOVI, the confirmatory trial required specifically to test whether that translates into functional benefit, was open-label and lost its own control cohort to a genotype mismatch, so what came back was a post-hoc external comparison rather than a positive randomized confirmatory finding.
This isn't a low-stakes trial-and-see decision either — the therapy runs into the hundreds of thousands of dollars annually, and treating "the mechanism makes sense" as sufficient justification skips past exactly the question accelerated approval's confirmatory-trial requirement exists to answer.
I don't think this needs to resolve as accept-or-decline. I'd start eteplirsen, but set up an explicit, pre-specified monitoring plan with J.M.'s father from day one — North Star Ambulatory Assessment and timed function tests at fixed intervals, with a real conversation now about what a meaningful change, or the absence of one, would actually mean for continuing.
That gives his father honest information either way, rather than either withholding a mechanistically reasonable option or letting an unproven surrogate stand in unexamined for years of ongoing treatment.
Agreed: eteplirsen started, with North Star Ambulatory Assessment and timed function testing at six-month intervals, and an explicit conversation with J.M.'s father about what those numbers will and won't tell the family.
Not agreed: the Clinical Pharmacologist's broader concern about accelerated-approval therapies being adopted on biological plausibility ahead of confirmed functional benefit remains open, unresolved by this individual family's decision.