Two Children Died in Hidden Chinese Gene-Editing Trials—What Went Wrong

What happened. On March 24, 2025, a doctor injected a CRISPR base editor — delivered via a virus — into the fluid at the base of Mei's spine . Within days, she developed fever and kidney damage. Seven days after the injection, she died from thrombotic microangiopathy, a condition that causes deadly microscopic blood clots . The hospital's own ethics board later concluded the death was "definitely related" to the treatment .

The cover‑up. The researchers never updated ClinicalTrials.gov or any other registry. They subsequently published a paper in Nature in early 2026 describing their preclinical animal studies — without mentioning that a human patient had already died . The omission also included the fact that the family had funded the therapy themselves . A joint investigation by Science and Retraction Watch, published July 23, 2026, was the first to reveal the death to the public .

Regulatory gaps exploited. The trial was an "investigator‑initiated trial" (IIT), a category that faces significantly weaker oversight than company‑sponsored studies . Experts have noted that China's regulatory framework has "overlapping responsibilities of multiple governing agencies and limited involvement of patient groups and the public in the legislative process," creating gaps that allow such trials to proceed with minimal transparency . Seven independent experts who reviewed the case for Science expressed concern that safety signals in animal studies were overlooked and that the trial should not have proceeded .

The Second Death: A Child with Duchenne Muscular Dystrophy, HuidaGene HG302 Trial — August 2025

The patient and the therapy. A child (reported as a boy in multiple sources) with Duchenne muscular dystrophy (DMD) was the fourth and final patient enrolled in HuidaGene's HG302-01 investigator‑initiated trial at Shanghai Children's Medical Center . The trial, which launched in December 2024, targeted boys aged 4 to 8 who were still able to walk . HG302 used a CRISPR‑based DNA‑editing therapy delivered via a high dose of adeno‑associated virus (AAV) .

What happened. The child died in August 2025 from acute respiratory distress syndrome (ARDS) caused by a severe immune reaction to the high‑dose AAV therapy . HuidaGene confirmed this in a statement on August 5, 2026, attributing the death to a "fatal serious adverse event" .

The cover‑up. The death was not reported for approximately one year. In February 2026, the clinical trial registry was updated to show the study was "complete," with no mention of a fatality . Just weeks before the death was finally disclosed, in July 2026, HuidaGene had presented data at the CRISPR Medicine conference claiming that HG302 had demonstrated "strong safety with no SAEs, DLT, or major toxicities" . The company's own former CEO, Alvin Luk, had previously acknowledged that high‑dose AAV microdystrophin trials from other developers had caused deaths at high doses . The death was first revealed by STAT News on August 5, 2026, after a months‑long investigation .

Key departures and silence. After the conference, HuidaGene went dark — issuing no press releases for 15 months, during which time both Luk and Chief Technology Officer TJ Cradick left the company .

Systemic Failures Across Both Cases

Both cases share a striking set of common problems:

Fallout and Ongoing Investigations

In late July 2026, Shanghai Jiao Tong University School of Medicine announced it had formed a task force to investigate the Xinhua Hospital trial and the related Nature publication . The university said it would take "serious measures" depending on the investigation's outcome . Calls have emerged for a full institutional review of the Nature paper and potential retraction or correction .

Mei's parents told investigators that doctors did not warn them that their daughter could die, and that the hospital and researchers hid the death . The family's decision to share their story was driven by a hope that other families would be spared the same fate .

Broader Implications

The two cases have reignited a global debate about the oversight of gene‑editing clinical trials, especially investigator‑initiated studies that can bypass the more stringent monitoring applied to industry‑sponsored trials. A 2025 academic review of China's genome editing regulatory framework had already identified ongoing shortcomings, including overlapping agency responsibilities and limited public input . Scientists commenting on the cases described them as revealing "serious failures in risk assessment, transparency and institutional oversight" .

Regulatory gaps similar to those exploited in these trials have been previously documented in the wake of the 2018 He Jiankui gene‑edited babies scandal, which also involved violations of ethical and regulatory norms in China .