Health and Medicine
PUBLISHED
Last week, journalists investigating a trial of a gene-editing medication revealed that the experimental drug caused the death of its lone recipient, a six-year-old girl. Her parents are worried that the doctors and researchers who ran the trial have not been held to account, and that failure to publicly acknowledge the death could lead to tragedies for other families in the future.
After many centuries of progress against infectious disease, the advent of genetic editing technologies has brought hope that inherited genetic conditions are also coming within science's capacity to treat.
But as described in heart-rending detail in a report by Science and Retraction Watch, speculative and experimental treatments like this one must be subject to thorough testing, and their potential dangers ethically and honestly disclosed, before they are used in humans.
A single misspelled gene
The girl at the center of this case, who was given the pseudonym Mei by her parents, had shown slower-than-normal intellectual development, particularly in speech and fine motor control.
Doctors at Shanghai Children's Medical Center traced the issue to a mutation on the CHD3 gene, where one letter (or base) of her genetic code was a thymine rather than a cytosine, hindering the expression of other genes and ultimately preventing the production of proteins needed for proper brain development.
The condition, known as Snijders Blok-Campeau syndrome, is extremely rare, but with recent improvements in a kind of gene editing technology called base editing, Dr Zilong Qiu at Shanghai Jiao Tong University thought a breakthrough was possible.
Qiu had previously worked on using base editing - which is a more precise variant of CRISPR gene editing - to treat children with other rare diseases.
Where CRISPR breaks both strands of DNA and relies on the body's own machinery to paste over the damage with correct code, base editing breaks only one strand and chemically converts the base at that point, leading to a more precise fix.
Mei's parents were concerned her condition would hinder her throughout her life and prevent her from being independent as an adult, and the pair initially contacted Qiu in 2023 after seeing his work discussed in a social media group for parents of children with developmental conditions.
Over the following two years, they contributed a sum of $860,000 to help fund Qiu's work, and in March 2025, Mei became the first person to receive gene-editing therapy directed at the brain, despite concerning results from a trial on monkeys conducted just a month before.
Tragically, the infusion triggered an immune reaction leading to kidney failure, and Mei died just a few days later at Xinhua Hospital, which had run the trial.
Ethical questions
Besides the decision to apply the treatment in a human before the results of preclinical trials had been reviewed and published, Science reports multiple other instances of questionable ethics in the run-up to Mei receiving the base-editing treatment.
For one, because it was the family, rather than a commercial sponsor, that paid for much of the work, responsibility for approving the trial lay with the hospital rather than a national regulator. This means some regulatory processes that might have identified the dangers of the treatment were bypassed.
In addition, although Mei’s parents were warned initially that the price would be high, they were only informed of additional costs after already making some initial payments. These subsequent payments, they said, were made directly into the bank account of one of Qiu's collaborators. In a recording at a meeting to discuss these payments, obtained by Science, Mei’s father can be heard saying "I'm feeling a bit of pressure."
Before any trial, patients or their guardians must sign an informed consent form, a document setting out what the treatment involves and what could go wrong, so that anyone agreeing to take part knows the risks they are accepting.
The form Mei’s parents signed a month before the procedure warned of the kinds of side effects Mei experienced, but it didn’t say she could die. Neither, the pair told Science, did Qiu or any of the other doctors. "Death should always be mentioned in a first-in-human trial," Stanford University bioethicist Hank Greely told Science.
Possibly the most disturbing part of the story is that neither Mei's death nor her parents' involvement in the research were included in the report of the animal research when it was finally published, potentially leaving others to make the same fatal mistake.
Science and Retraction Watch invited seven experts across to review the paper in light of what their investigation had uncovered. The responses were damning. Professor Steven Gray at the University of Texas Southwestern Medical Center, who works on using viruses for gene therapy, told Science "This shouldn't have gone to trial."
Qiu did not respond to repeated requests for comment from Science and Retraction Watch, and nor did his university, the hospital, or the doctor who oversaw the clinical side of the trial.
Most alarming of all is that the tragedy has only come to light because the parents contacted Science and could provide documentation to prove their claims, including recordings of their meetings with Qiu.
It took the ensuing publicity for the tragedy to be taken seriously. A follow-up report in Science included a statement from Shanghai Jiao Tong University that it: "Has taken very seriously the recent online reports concerning medical research papers published by a researcher at our institution surnamed Qiu, as well as a clinical study conducted at our affiliated Xinhua Hospital."
The university has now launched an investigation into the trial.
Without a culture of a free press, it's unlikely most Chinese parents would have done what Mei’s parents did in contacting Science. That raises the question of whether similar cases have occurred without coming to light.
"Learning the reality of these missing safeguards has fundamentally changed how we now view the entire project," Mei's father told Science. "We did not realize how unusual and dangerous many of the arrangements were."
"After the child's death, the hospital paid a modest fine to a local health authority, but Qiu was not publicly sanctioned," Science reports.
Failure to disclose
It's not as though the risks of gene therapy would have taken Qiu and colleagues by surprise. In 1999, experimental gene therapy to treat 18-year-old Jesse Gelsinger's incapacity to metabolize ammonia caused his death, leading to global caution.
China had its own cautionary moment in 2018 with He Jiankui's use of CRISPR to edit the genes of babies without consultation. Qiu condemned He's work at the time, but it seems the lesson didn't stick, as Mei's parents say he misled them regarding the risks involved in his own project.
For Qiu's base editor to be delivered into Mei's brain, it had to be packaged inside two adeno-associated viruses. These viruses have been used for gene therapy and in certain vaccine technologies for some time, but they can cause an inflammatory response, particularly when received in high doses.
With both viruses needing to access a high proportion of brain cells for the therapy to work, the dose Mei received was orders of magnitude larger than would be expected for someone receiving a virus-based vaccine. This carried with it the risk of damage to her liver and kidneys.
In trials in monkeys, the results of which were known to Qiu over a month before Mei's procedure, but which hadn't been reviewed by the hospital or published in a scientific journal, all the animals experienced liver damage, and in one case kidney damage, that would normally have halted progression to human trials.
In addition, experts consulted by Science expressed doubts about microscopy evidence for the effectiveness of the treatment, which also came from the monkey trial. When the paper was initially submitted, these experiments lacked any controls, making their results "completely unconvincing," according to Purdue University's David Sanders.
While the paper Qiu finally published in Nature included these controls, one expert told Science that the control experiments appeared to have been performed at a different time with different microscope settings and so were difficult to compare.
For parents whose children have a fatal condition, such risky gene therapy might be the only hope. Yet while Snijders Blok-Campeau syndrome can be extremely debilitating in severe cases, Mei had a milder case and had made progress with the help of speech and occupational therapists.
Had her parents been honestly informed of the risks, it is entirely possible they would not have gone ahead with the treatment.
A similar gene modification approach appears to have cured "Baby KJ" of a metabolic disease a month before Mei died, which Science awarded runner-up status for its 2025 Breakthrough of the Year.
Notably, however, Baby KJ was given a low dose to test his response, which was increased after the reaction was moderate.
Few would advocate scrapping the idea of medical gene editing entirely. However, this tragic case serves as a reminder that such treatments are still in their early stages, which is why ethics regulations need to be closely enforced, and patients honestly informed about the potential risks.
[H/T: Science]
