An investigation alleges that the death of a girl in China was covered up after she was given an experimental gene therapy

Gene therapy strategies for very rare diseases with severe neurological effects are attracting the attention of many researchers, particularly what is known as ‘n=1’ therapy – that is, therapy targeted at a single patient. Following the successful cure of a baby in the United States in 2025, after receiving a one-off liver treatment for an extremely serious metabolic disorder, China does not want to be left behind. In this case, the original scientific study was made possible thanks to private investment from a Chinese family with an affected daughter, who paid the equivalent of around 800,000 dollars to develop a CRISPR-based therapy using a base editor, delivered intrathecally (into the cerebrospinal fluid) via adeno-associated viruses. Unfortunately, the viral dose was so high that it caused the patient’s death due to an exacerbated inflammatory response.

This case goes beyond human tragedy or bad luck, as there are numerous indications of scientific and bioethical malpractice. Although treatment in mice using a different virus appears to have been effective (though some scientists dispute this), the virus would have had to be changed for use in humans, and trials in four primates showed severe liver and kidney damage, which raises a red flag, as a therapy must, above all, be safe. However, these results were not communicated to the parents before their daughter was treated (thereby limiting their ability to make an informed decision) and have been omitted from the publication; likewise, any reference to the family’s funding, how the mutation was selected, and the fact that a patient who died was treated has also been omitted. Similar to what happened in the case of the girls treated by He Jiankiu in 2018, the rush to be the first to publish an innovative therapy has led researchers to disregard the most basic bioethical principles. Furthermore, with a clear lack of transparency, the Chinese authorities have maintained secrecy and ignored the well-founded complaints of the affected parents. The Chinese government prioritises biomedical innovation, which is handsomely rewarded, and has more lenient regulations for gene therapies compared with those in Europe or the United States. The journal Nature, in which the research on mice was published, is considering whether the publication has adhered to the required bioethical standards.

Gene therapy has always offered hope to many patients with rare genetic diseases, but it carries risks, particularly when viruses are used, as these can trigger an exaggerated immune response. The approach is experimental and, as such, it is not enough to have an excellent technical and scientific approach; one must also adhere impeccably to bioethical principles to ensure every step is sound and to avoid the backlash of mistrust that this generates in society.

This is extremely hard news. The truth is that we are still learning to strike a balance between optimism and caution. These new precision genomics medicines enable us to turn dreams into reality, but our experience in this area is still limited. In any case, in this cutting-edge field of science, maximum transparency is more essential than ever, both to ensure the best possible decisions are made now and to lay solid foundations for the future.

Just over a year ago, we learnt of the case of a boy born with a very serious metabolic disorder (a deficiency in a key enzyme in the urea cycle, called carbamyl phosphate synthetase 1, which is absolutely essential for removing ammonia from the body and detoxifying it) who was treated during his first few months of life with an experimental therapy based on CRISPR base editors and managed to survive, thus avoiding the certain death he was heading towards, just like any other child born with similar mutations in the same gene.

Therapeutic interventions for individual patients do not always end in success. In late 2022, we learnt of the case of a boy in the US with Duchenne muscular dystrophy in an advanced stage who received an experimental treatment, following approval for compassionate use by the regulatory body, the FDA. The therapy was based on the use of a variant of CRISPR tools known as epigenetic editing, to reactivate the mutated dystrophin gene using other regulatory sequences. The treatment was carried out by the company Cure Rare Disease, whose CEO was the brother of the patient with Duchenne, which in itself raised an underlying conflict of interest. He was injected with such a large quantity of viruses carrying the CRISPR tools (based on the doses injected into mice) that his immune system was unable to cope, and he died within a few days, partly due to an unknown predisposition he had which led to immune failure.

This week, through an exclusive report in the journal Science, we have learnt of the case of a Chinese girl with a mild form of a neurological syndrome associated with cognitive impairment and autism. Although this patient did not have a severe form of the condition, she was nonetheless treated with an experimental procedure that had not been sufficiently validated or tested in preclinical stages; this resulted in her death within a few days of receiving the treatment, the substantial cost of which was borne by the family. Once again, a treatment involving a single patient has had fatal consequences. In this case, the patient received a spinal cord infusion containing a huge quantity of viruses carrying a CRISPR base editor designed to correct a single nucleotide in the affected gene expressed in the brain. The researchers had previously published preclinical trials in animals (mice and non-human primates) in which liver and kidney failure had already been observed, leading to recommendations that the procedure be reviewed before proceeding with human subjects. The researchers chose to ignore these preclinical warnings and decided to continue with the experiment, which has ended in the girl’s tragic death.

Single-patient experimental therapies often make big headlines in the media and feature on the news. When they work. These are risky therapeutic procedures, not yet formally approved, in which the safety and efficacy of proposed new treatments are assessed in a single patient, in the absence of any other cure. However, if something goes wrong, we do not usually find out about it, unless the consequences are fatal for the patient and the story makes the headlines again due to the harm caused.

These single-patient therapies leave very little room for interpretation. We can only rejoice when they work as expected. But when they fail to meet expectations, they generate more doubts than certainties, as there is no control case against which to compare them, nor any possibility of comparing the results of different doses or different routes of administration. And, when they result in the death of the treated patient, they raise even more questions and cause distress amongst the patient’s family and the scientific and medical community.

Ethically, they also raise a debate that cannot be ignored, given the decision to devote vast resources to saving the life of a single patient, whilst other patients with the same condition are excluded from the potential benefits of this therapeutic approach. Objectively speaking, it will always be better to organise a clinical trial that meets all the necessary requirements, with well-established phases and formalities, in order to assess the safety and efficacy of an experimental medicine or therapy in a cohort of patients, rather than staking everything on the possible success of a single patient. Nor can we overlook or fail to take into account the problems that arise during the preclinical phases of a therapy’s development, in animal models, which must precisely be included in order to anticipate potential major problems in patients. These preclinical phases are indispensable and must be undertaken and interpreted appropriately. There are no shortcuts in the development of therapies.

Mei’s tragic story is one of abuse of authority, trust and hope. Mei suffered from Snijders Blok-Campeau syndrome, an ultra-rare condition caused by a mutation in the CHD3 gene, which affects neurodevelopment in paediatric patients. Mei’s parents, concerned for her future, searched hopefully for solutions to their only daughter’s condition. In their search, filled with trust and optimism, they decided to place Mei’s life in the hands of Zilong Qiu, an expert in neurodevelopmental disorders with an impeccable CV, brimming with prestigious titles and brilliant achievements (such as publications in elite scientific journals and positions at top-tier institutions). However, Qiu was blinded by his ambition and desire for glory.

Mei’s parents, Jason and Linda, paid hundreds of thousands of dollars to fund the development of a gene therapy for their daughter. On occasion, certain payments were made through informal arrangements and in the form of personal gifts to Qiu, thereby blurring the line between the personal and the professional—a line which Qiu should have had the good judgement and moral clarity not to cross. Eventually, following exorbitant sums of money and excessive promises made by Qiu to Jason and Linda, Mei received an injection of the gene ‘therapy’ into her spinal fluid, at a dose exceeding the range supported by the available preclinical data. Less than a week after the injection, having developed a fever and signs of acute kidney failure, Mei died of thrombotic microangiopathy, a serious adverse event caused by the experimental treatment. In shock, Jason and Linda asked Qiu not to publish the study detailing the development of the experimental treatment that had caused their daughter’s death.

To this day, the article in question—which was published in the prestigious journal Nature on 18 February 2026—remains available, accumulating views and citations, and thus presenting an incomplete picture of the experimental treatment’s safety. Furthermore, Qiu decided to remove any mention of Mei and her parents from the study, including the funding they had provided. Moreover, having achieved what he wanted—a high-profile article in Nature—Qiu stopped responding to Jason and Linda. Unfortunately, it is still unclear what the consequences for Qiu will be. However, it is clear that Mei’s case deserves more attention than it has received. Likewise, stricter oversight and safeguards must be demanded for the development of gene therapies in China, to prevent further researchers, blinded by their ambition, from exploiting the hopes of parents like Jason and Linda.

To mark Mei’s first birthday since her death, whilst Jason was at work, Linda bought a chocolate cupcake and sat on her daughter’s bed, imagining she was sharing the cupcake with her.