An undisclosed death following a personalised gene-editing treatment in China has intensified scrutiny of safety testing, informed consent and research transparency.
Can an experimental gene-editing treatment safely correct a rare brain disorder in a child?
A case in China has shown how dangerous the answer can become when scientific ambition moves faster than safety safeguards. A six-year-old girl died seven days after receiving a personalised gene-editing therapy designed to correct a mutation associated with a rare neurodevelopmental disorder.
The girl, identified by the pseudonym “Mei,” received the experimental treatment at Xinhua Hospital in Shanghai in March 2025. Her death was not publicly disclosed until a joint investigation by Science and Retraction Watch revealed the case more than a year later.
The investigation has raised serious questions about preclinical safety testing, informed consent, financial conflicts, regulatory oversight and the reporting of fatal adverse events.
Advertisement
What Is Snijders Blok-Campeau Syndrome?
Mei had Snijders Blok-Campeau syndrome, an ultra-rare genetic condition caused by changes in the CHD3 gene.(1✔ ✔Trusted Source
Snijders Blok-Campeau syndrome
).
The CHD3 gene provides instructions for making a protein involved in controlling how DNA is packaged within cells. This process helps regulate when different genes are switched on or off and plays an important role in early brain development.
People with the condition may experience developmental delays, speech difficulties, intellectual disability, low muscle tone, distinctive facial features and structural differences in the brain.
The severity of the disorder varies considerably. Some affected individuals may also experience seizures or heart problems, but most are believed to have a normal life expectancy.
Mei reportedly had a mild form of the condition. She spoke in simple sentences, attended kindergarten and was receiving speech and occupational therapy.
Advertisement
Parents Paid $860,000 to Develop a Personalised Treatment
Mei’s parents reportedly contributed approximately $860,000 toward developing, testing and administering a treatment created specifically for their daughter’s mutation.
Such treatments are sometimes called “n=1” therapies because they are designed for a single patient rather than tested across a large group of participants.
The therapy was developed by a research team led by neuroscientist Zilong Qiu. Its aim was to correct a single-letter mutation in Mei’s CHD3 gene using a CRISPR-based technique known as base editing.
Unlike conventional CRISPR-Cas9 editing, which can cut both strands of DNA, base editing is designed to change one DNA letter into another without producing a double-strand break.
Although the technique offers considerable promise, its safety depends not only on the gene editor but also on how that editor is delivered into the body.
Advertisement
High Viral Dose and Animal Toxicity Raised Safety Concerns
On March 24, 2025, Mei received a spinal infusion containing hundreds of trillions of AAV9 particles designed to carry a base-editing system into her brain cells.
Although AAV vectors are widely used in gene therapy, very high doses can trigger severe immune reactions. Despite receiving a steroid, Mei developed fever, stopped producing urine and experienced a sharp fall in platelet levels. She was admitted to intensive care but died seven days after treatment.
Earlier animal studies had reportedly shown warning signs: four monkeys developed moderate-to-severe liver damage, while one also experienced kidney damage. However, Mei’s procedure was allegedly approved without a complete review of these findings, and her parents said the possibility of death was not clearly explained.
Because Mei’s condition was not life-threatening, experts questioned whether the available safety evidence justified the experiment. “This shouldn’t have gone to trial,” said gene-therapy researcher Steven Gray.
Child’s Death Was Omitted From Later Nature Publication
In February 2026, the researchers published related preclinical findings in the journal Nature.
The study described how a base-editing system corrected a CHD3 mutation and improved behavioural abnormalities in mice. It also reported that spinal delivery of AAV9 vectors achieved broad distribution in the brains of nonhuman primates.
However, the published paper did not mention Mei’s treatment, her death or her family’s financial contribution to the development of the therapy.
Nature said it was unaware of the circumstances surrounding the human treatment before publishing the paper and was examining the matter.
Mei’s parents reportedly asked the researchers to withdraw the paper because they feared other families might pursue the same experimental treatment without understanding its risks.
Several independent experts have also called for a complete review of the data and circumstances surrounding the publication.
Why Personalised Gene-Editing Therapies Need Greater Oversight
Personalised gene-editing treatments may offer hope to patients with mutations so rare that conventional drug development is commercially difficult.
However, single-patient therapies leave little room for comparing doses, identifying uncommon complications or determining whether a treatment is genuinely effective. (2✔ ✔Trusted Source
Death of girl in Chinese gene-editing trial was never made public
The risks become even greater when the treatment involves a child who cannot provide independent consent, a nonfatal condition, an exceptionally high viral dose and limited preclinical safety evidence.
Experts stress that unsuccessful treatments and serious adverse events must be disclosed as openly as successful cases. Without complete reporting, other researchers cannot accurately evaluate risks or prevent similar harm.
Gene-Editing Promise Cannot Come Before Patient Safety
Gene-editing technologies have already produced encouraging results in inherited blood disorders and some life-threatening genetic diseases. However, success in one organ or condition does not guarantee that the same approach will be safe in the brain.
Mei’s death does not mean that personalised gene editing should be abandoned. It shows why experimental treatments must be supported by rigorous safety evidence, independent ethical review, fully informed consent and complete transparency.
When a therapy is designed for only one patient, every safety decision carries even greater weight. Scientific innovation can move quickly, but the responsibility to protect patients must move first.
Precision medicine cannot fulfil its promise unless precision in safety, ethics and transparency comes first.
References:
- Snijders Blok-Campeau syndrome – (https://medlineplus.gov/genetics/condition/snijders-blok-campeau-syndrome/)
- Death of girl in Chinese gene-editing trial was never made public – (https://www.science.org/content/article/exclusive-death-girl-chinese-gene-editing-trial-was-never-made-public)
Source-Medindia