HuidaGene’s first-in-human HG302 study found gene editing and dystrophin in muscle, but its four participants cannot show whether the therapy helps, and one boy died after a severe immune reaction.
A targeted edit for some patients
Duchenne muscular dystrophy is caused by mutations in the DMD gene that prevent the body from making enough dystrophin, a protein that helps protect muscle fibers. HG302 targets a subset of boys whose mutations may be addressed by skipping exon 51, a section of the gene’s instructions. The edit is intended to restore the gene’s reading frame so cells can make a shorter dystrophin protein. It is not designed for every Duchenne mutation or to replace the whole gene.
The treatment was intended as a one-time change to a patient’s DNA, rather than repeated doses of a drug that encourages exon skipping. HG302 uses an AAV9 virus to carry its editing system through the bloodstream toward muscles throughout the body.
What the small study could show
HuidaGene reported encouraging results in humanized mice and non-human primates before beginning the first-in-human study in 2024. The open-label dose-escalation study enrolled four ambulatory boys aged 4 to 8, treated at one of two levels: 5.0×10¹³ or 1.0×10¹⁴ vector genomes per kilogram. With no control group and so few participants, it could assess whether the therapy reached muscle and identify obvious safety concerns, but not establish whether it worked.
Muscle biopsies from the three surviving participants showed editing at the intended DMD site, at measured frequencies of 1.02% to 4.31%, and detectable dystrophin. Two participants who completed 12 months also recorded gains in walking and motor ability. Those results cannot show that HG302 caused the gains, and the biopsies cannot establish that enough muscle fibers were changed for lasting benefit.
The serious safety concern
The fourth and final participant received the higher dose and died in August 2025. The published account describes rapid illness after infusion, including acute respiratory distress syndrome, marked activation of complement, a part of the immune system, and a systemic cytokine response. HuidaGene’s investigation and the Cell paper link the event to high-dose systemic AAV administration. They point to an immune reaction associated with delivery, not a reported off-target edit.
The Cell paper reports editing and manageable safety in primates, but says the toxicology work was too limited to predict uncommon or host-dependent harms. Reaching muscles throughout the body may require a substantial viral dose, while immune responses can vary between people.
Questions about reporting and review
HuidaGene said the death was reported through the hospital’s ethics and study-oversight processes within the applicable timeframe, and that its full investigation was submitted for peer review in January 2026. STAT reported that the company’s public announcement came after its investigation and repeated questions, about a year after the death.
The trial was investigator-initiated, and the public ClinicalTrials.gov researcher record says no data-monitoring committee was listed. That does not by itself establish a breach of rules, but it raises questions about independent review of safety information during dose escalation and how quickly families and researchers learned what happened.
What happens next
The stated next step is long-term safety follow-up of the three surviving participants; dosing in the four-person protocol is complete. The paper does not announce another trial or a restart. Any future development would have to show both that a dose can produce enough editing to matter and that it has an acceptable safety margin.
One death among four children cannot reliably show how often such a reaction would occur. But the event makes clear that further work, if any, will need stronger evidence on immune risk as well as dystrophin and function.