The young girl tugged on her mother’s hand as they pressed through the doors of the hospital in Shanghai. She was 6 years old, bouncing along in a pink jacket and blue pants decorated with cartoon bears. Behind them, her father rolled a large suitcase with everything the child needed for the weeklong stay: stuffed animals, Play-Doh, an iPad loaded with episodes of Peppa Pig.
She told her parents it felt like they were going on vacation. In fact, they brought her here for an experimental gene therapy. The girl was slipping behind her peers in kindergarten. She still spoke in simple sentences and ate with training chopsticks. Underneath it all was a single mutated DNA base, a T that should have been a C.
“Your ‘book’ has a small mistake, which has caused you to have a disease that affects your growth,” read the children’s version of the informed consent form from the hospital. “Over time, it can get more serious.” Doctors hoped to repair that mistake while her brain was still building itself. It would be a clinical trial of one, funded in part by $860,000 the parents had scraped together from their own savings and from relatives.
The parents felt they were in good hands. Xinhua Hospital, which is affiliated with the Shanghai Jiao Tong University School of Medicine, was acclaimed for its pediatrics department. It was the first Chinese institution to perform open heart surgery on infants, and the first in the country to separate conjoined twins. If all went well, the hospital would make history once again. The girl would be the first person in the world to receive a gene-editing therapy directed at the brain. It would rewrite the mutated gene in her neurons, restoring the needed DNA base so she could make a vital protein.
Leading the effort was Zilong Qiu, a neuroscientist at the university’s brain center, the Songjiang Research Institute. At the time, in late March 2025, Qiu was one of several researchers around the world vying to push base editors—a more precise form of the powerful gene editor CRISPR—into custom treatments for children with rare diseases. One month earlier, KJ Muldoon, an infant with a life-threatening metabolic disorder, had quietly received his first intravenous infusion of one such treatment at the Children’s Hospital of Philadelphia.
Although news that base editing saved “Baby KJ” would soon rocket around the world—Science named the feat one of the runners-up for its 2025 Breakthrough of the Year—the story of what happened at Xinhua Hospital has remained hidden. An entry for the study posted to ClinicalTrials.gov has not been updated for more than a year. And when Qiu and his colleagues published proof-of-concept animal studies related to the trial in Nature early this year, they stripped the paper of references to the family and its financial contributions, noting only that “bridging the gap between preclinical research and clinical translation remains a significant challenge.”
That vague language glossed over tragedy: Seven days after the girl’s medical team infused trillions of viruses carrying the recipe for the base editor into her spinal fluid, she died of a severe immune reaction linked to the therapy, Science and Retraction Watch can now reveal.
According to official documents and accounts provided by the girl’s parents, the hospital had allowed Qiu’s experimental treatment to proceed under a regulatory provision that does not require approval from national regulators. After the child’s death, the hospital paid a modest fine to a local health authority but Qiu was not publicly sanctioned.
The troubling death represents a new stumble in China’s push to rival the United States as a biotech power. Three years ago, the government strengthened its biomedical research rules in response to the 2018 scandal around He Jiankui, a biophysicist who violated ethical norms to secretly produce gene-edited babies. The lax oversight of this recent trial and the failure to publicly report the fatality “shows the gap between what is intended and what has been put in place,” says Joy Zhang, a sociologist at the University of Kent who has written about the pervasive culture of secrecy in Chinese scientific institutions.
Seven experts in fields including genetics, virology, and bioethics who reviewed details of the Nature study and the clinical trial for Science and Retraction Watch expressed concern that Qiu and his team downplayed the trial’s risks in describing them to the parents, overlooked safety signals in animal studies, and proceeded even though success was unlikely. “This shouldn’t have gone to trial,” says Steven Gray of the University of Texas Southwestern Medical Center, who develops viruses for gene therapy.
Gray and several of the other experts are calling for a full review of the images and other data in submitted and published versions of the Nature paper and full disclosure of the study’s funding. Some of the issues might warrant a retraction, they say. Neither Qiu nor his university or the hospital responded to multiple requests for comment for this story, and Nature says it was not aware of the issues surrounding the clinical trial before it published the group’s paper.
The girl’s parents, who requested that Science use pseudonyms for them and their daughter for privacy reasons, have decided to tell her story now because they are angry about what they feel is a lack of accountability by the researchers and the institutions. “Learning the reality of these missing safeguards has fundamentally changed how we now view the entire project,” says the father, a software engineer. He asked that he be called Jason, his wife Linda, and their daughter Mei (Chinese for “beautiful”). “We did not realize how unusual and dangerous many of the arrangements were.”
As with Jesse Gelsinger, the 18-year-old whose death during an early U.S. gene therapy trial in 1999 chilled research in the field for nearly a decade, Mei’s story raises questions about whether such risky treatments should be initially tested on patients with nonfatal conditions. “I don’t think these kinds of experiments should stop, but we need to make sure we are appropriately careful about them,” says bioethicist Hank Greely, director of the Center for Law and the Biosciences at Stanford University. “It’s so easy to be blinded by hope—whether it’s hope for your kid, hope for your research, or hope for your company.”
Jason and Linda had been trying to have a child for 4 years before Mei was born in early 2019. They marveled at every milestone: her first words, her first steps. She did seem a little clumsier than other toddlers. But within a few years, she was attending Parkour classes and bouncing on a neighborhood trampoline.
When Mei was 4, one of her kindergarten teachers pulled Linda aside: Mei didn’t draw or write as well as the other kids and her language skills weren’t developing normally. Her mother might want to get her evaluated, the teacher said. In March 2023, Mei was diagnosed with global developmental delay, a broad label with many causes. Specialists explained that some of Mei’s behaviors—the funny sounds she liked to make, for instance—were associated with autism.
The family followed up with a battery of tests at the Shanghai Children’s Medical Center. A brain MRI, chromosome analysis, and a test panel for genes known to influence neurodevelopment all came back normal. Detailed sequencing of more of Mei’s DNA finally pointed toward the delay’s root cause: a defect in a gene, called CHD3, that influenced how strands of DNA were wound into chromosomes inside Mei’s cells.
This packing affects how well other genes are expressed, including those involved in the early development of the brain. By altering gene activity, CHD3 mutations produce a condition called Snijders Blok-Campeau syndrome that is known to afflict just 237 individuals around the world.
Philippe Campeau, a medical geneticist at the University of Montréal who helped identify the syndrome’s genetic basis in 2018, says people with the mutation often have a normal life expectancy, but their symptoms vary widely. Most have slightly larger than normal heads, and about two-thirds have intellectual deficits. Moderate to severe cases may be nonverbal, suffer from seizures and heart problems, and have fluid-filled voids in their heads. For those patients, gene-editing therapies could be revolutionary, according to Campeau. “For somebody who is just mildly affected … that’s a bit more debatable,” he says.
Mei sat on the milder end of the gradient, but Linda soon quit her job to devote her energy to their only child. She and Jason took Mei to speech and occupational therapists and got her special education support. It worked, to a degree. After they transferred Mei to another kindergarten to repeat a year, most of the other parents weren’t even aware of her condition, they said.
Whereas Linda was an optimist by nature, Jason feared the worst. The condition is not degenerative, but the gap between Mei and her peers in school would likely widen as courses grew harder. She was also skinnier than other children and had poor muscle tone. She might not ever be able to live independently. “We were worried about her future,” Jason says.
After Mei’s diagnosis in 2023, the family joined a private WeChat group in which parents of children with autism and similar disorders trade advice. It was there that they first heard about Qiu.
An expert in neurodevelopmental disorders, he did his Ph.D. at the Shanghai Institute of Biochemistry and Cell Biology and, like many promising Chinese students of his generation, went to the U.S. for a postdoctoral position. In 2003, he joined the laboratory of Anirvan Ghosh, a neuroscientist then at the University of California (UC) San Diego, who found Qiu to be “talented and motivated.”
In 2009, as China was ramping up efforts to lure its scientists back home, Qiu took a post at his undergraduate alma mater. He published in all the right places—Nature, Neuron, and the Proceedings of the National Academy of Sciences. He also strove to demystify genetics for the Chinese public with a book, The Gene Enlightenment, and through public lectures, including a TEDx talk titled “Defying the destiny dictated by our genes.”
Qiu was among more than 100 Chinese scientists who condemned He’s rogue embryo gene editing in an open letter. “The project completely ignored the principles of biomedical ethics, conducting experiments on humans without proving it’s safe,” he told the South China Morning Post.
Qiu still saw potential in tackling autismlike conditions that have a clear genetic link, such as Rett syndrome. “He was taken by Rett syndrome and wanted to make a difference,” says Monica Coenraads, CEO of the Rett Syndrome Research Trust, which funded some of Qiu’s research in the U.S.
For parents in the WeChat group, Qiu was someone to watch. His gene therapy for Rett syndrome had recently been patented in China, and he founded a company, Lanqi Xintu Gene Technology, which would soon bring it into the clinic for testing. In July 2023, he gave a lecture about this work and other efforts. A parent from the group attended, and posted slides.
The following week, Jason sent Qiu an email describing Mei’s condition. “We understand that gene therapy requires complex experiments and can cost tens of millions [Chinese yuan]. We are willing and able to bear these costs,” he wrote. “Watching our child suffer every day is the most painful ordeal for us as parents.” He attached Mei’s genetic results and hit send.
Thirteen minutes later, Qiu replied with interest. He soon invited Jason and Linda to meet at the institute. Seated in his office with a leafy window view, the parents told him they weren’t interested in funding research for the sake of research. “We are really coming at this with the goal of an actual treatment,” Jason said.
Qiu was encouraging. “To be honest, if you had come to me last year, I wouldn’t have dared to say we could do this,” he said during the meeting. “We’ve reached a tipping point now.” (The parents recorded many of their conversations with Qiu and the other parties related to Mei’s trial because they were overwhelmed by the technical information. They shared the recordings with Science and Retraction Watch.)
Editing the brain, Qiu explained, would require delivering viruses into the fluid in the base of Mei’s spine. Because her condition was not life-threatening, however, the bar would be high to get the trial approved by the hospital’s ethics board, he added. It would require up to 2 years of work to develop the treatment and test it in mice and monkeys, with no guarantee it would be effective. But a top-tier institution such as Xinhua Hospital, Qiu said in one recording, could ensure there would be no safety catastrophes.
- Jason
- Mei’s father
- Yilong Qiu
- Songjiang Research Institute
The parents had heard about serious side effects, including deaths, caused by other gene therapies, and knew the greatest risk would be Mei’s immune response to the massive dose of virus. Qiu said getting the dose right was critical, but infusing the viruses directly into Mei’s spinal fluid, rather than the blood, would minimize the threat of a reaction because it would bypass the kidneys and liver.
The parents were starting to feel at ease. “The psychological cost of doing this is very high for me,” Linda told Qiu. “But I am willing to try this because you have proven to be very trustworthy.”
Jason and Linda say they soon wired their first payments to a foundation affiliated with another university. This payment would support one of Qiu’s former students who was designing the base editor. More of their money, according to documents reviewed by Science and Retraction Watch, flowed to a Chinese affiliate of a U.S.-based firm that would start to make the viruses, and to PriMed, a contract research organization in Sichuan province that would run the monkey studies.
Jeremy Sugarman, a medical doctor and bioethicist at Johns Hopkins University, says it’s not unusual for a family to bear the costs of developing a personalized treatment. But, according to text messages shared by Jason and Linda, Qiu also asked the couple to pay other members of the research team directly, through informal arrangements they found increasingly troubling.
In April 2024, at a Starbucks café near the university, Jason met with Kan Yang, the researcher overseeing the lab work and a co-inventor on Qiu’s Rett Syndrome patent, to discuss a fee schedule. Jason told Yang he was at a financial breaking point after learning it would cost more than $250,000 to produce clinical-grade viruses. “I’m feeling a bit of pressure,” Jason said, according to a recording.
Yang outlined why he needed even more money than Qiu had initially suggested. “The thing is, if you finish those viral injections and don’t have a proper person to analyze the results, it’s all for nothing,” Yang replied. He said the same would be true when it came time for the monkey studies. “It requires extremely meticulous work to get results.” (Yang did not respond to multiple requests for comment.)
They haggled over terms, according to the recording. The family ultimately sent $130,000 directly to Yang’s personal account over the course of the project, according to bank records. In the U.S., similar payments by desperate parents seeking a treatment may cross a line, according to Sugarman, who says he would be concerned about any “undue inducement” to provide excessive gifts or money.
Qiu never asked for anything himself. Every few months, however, Jason visited him at his office, a nearby restaurant, or his home a couple hours away. Jason always picked up the tab and gave Qiu multiple iPhones, an iPad, and a case of Maotai, a Chinese liquor made of sorghum and wheat, according to family records.
- Jason
- Mei’s father, responding to a request for payment
- Kan Yang
- Songjiang Research Institute
By the end of 2024, the lab’s efforts were paying off. Qiu’s team had engineered mice to have a human version of the CHD3 gene with their daughter’s mutation, R1025W, which results in a protein with the amino acid tryptophan where there should be an arginine. The mutant pups developed autismlike traits and didn’t squeak as much as normal mice when separated from their mothers. When the researchers repaired that mutation, the pups developed normally.
Such a surgical edit was only possible because of the base-editing techniques pioneered nearly a decade ago by David Liu’s lab at Harvard University. The original gene editor, CRISPR, which had been used by He Jiankui and others, often damaged the DNA it was meant to fix. “Chaos” is how Qiu described its effects to the family during their first meeting. To make an edit, CRISPR breaks both strands of DNA’s double helix, which can result in large insertions or deletions in the genome.
Base editors, which use a modified version of CRISPR’s DNA-cutting enzyme, only nick a single strand. With the help of an RNA guide, they unwind the DNA at the right spot and chemically convert the base on the intact strand into another base. The cell then repairs the nicked strand with the complementary base. For Mei’s mutation, the base editor would transform a specific adenine (A) base in the mutant gene to a guanine (G), which would cause the complementary thymine (T) to change into the correct base, cytosine (C).
Base editing wasn’t foolproof or easy to do, though. “Far fewer people have mastered this technique than gene therapy,” Qiu said in one recording. He was second only to Liu, he told the family, when it came to taking it to new frontiers.
Unlike the U.S. trial for Baby KJ, which delivered Liu’s own base editor into the infant’s liver via tiny fatty capsules, the Chinese researchers needed to reach Mei’s brain. They chose to pack genes for the editor’s enzyme and guide RNA into adeno-associated viruses (AAVs), which have long been used in gene therapy but can trigger inflammatory responses at the doses required. Because only a small fraction of these viruses enter their target cells and spur the production of the base-editing components, fixing enough brain cells would require hundreds of trillions of viruses—thousands of times more than a person receiving a virus-based vaccine gets.
Qiu’s team faced a second challenge. Because of the size of the base editor, they needed to split its genetic instructions in half and pack the DNA sequences into two different viruses that would be injected at the same time. Success required both viruses to be expressed inside the same cells throughout the brain. By comparison, Otarmeni, a U.S. Food and Drug Administration (FDA)-approved therapy for hearing loss that uses two vectors infused directly into the tiny, fluid-filled cavity of the inner ear, alters many fewer cells. “I’m skeptical that a dual vector approach would achieve high enough editing” to treat Mei’s condition, Gray says.
The experimental therapy had apparently worked in the mice. But the virus Qiu and colleagues used in those experiments, AAV-PHP, can be injected into a mouse vein and crosses the rodent’s blood-brain barrier. Because of species differences in certain cellular receptors, that virus would not cross the barrier as easily in humans or other primates. So Qiu’s team shifted to another common gene therapy “vector,” AAV9, which would be injected directly into the spinal canal, bypassing the barrier. Before Mei would be treated, however, the researchers intended to show the method could safely deliver the editing instructions to the brain cells of monkeys.
In January 2025, Qiu shared some good news with Jason and Linda. In a WeChat message reviewed by Science and Retraction Watch, he sent them a manuscript he had recently submitted to Nature. It compiled all his lab’s preclinical work on Mei’s mutation over the previous 18 months. The first figure in the paper illustrated the gene sequences of Jason, Linda, and Mei, along with a laboratory study demonstrating that the R1025W mutation damages the protein encoded by CHD3.
The submitted paper also included striking images that Yang and colleagues had captured from brain slices from two macaques that had received Mei’s potential therapy via AAV9. Yang had stained the neurons green, and the gene editor red to estimate what fraction of neurons it had reached. Under fluorescence imaging, a zoomed-in view of the cerebellum appeared almost entirely red, suggesting the editor had made it into nearly every neuron in the region where CHD3 is most strongly expressed.
This was welcome proof that the clinical trial should go ahead, Qiu told the family over WeChat. Indeed, that result apparently helped convince the hospital ethics committee to approve the family’s clinical trial on 2 January 2025.
But four of the experts Science asked to assess this figure, including one of the original reviewers of the Nature manuscript, expressed concerns about how well the primate proof-of-concept study was conducted. The submitted paper had no control samples to show the stain was selectively highlighting the gene editor. “It’s completely unconvincing,” says David Sanders, a biochemist at Purdue University who has studied gene therapy. “One can’t have the confidence that one isn’t mostly looking at background staining.”
An accompanying graph suggested AAV9 only altered 30% of targeted neurons, but that was several times more than achieved with similar vectors in other primate studies. A fifth expert, a neuroscientist who asked not to be named, said the criticisms raised may be valid but there was no “obvious data manipulation or image doctoring.”
Another potential concern, however, emerged about 1 month before Mei was treated: the final conclusions of a primate toxicology study conducted at PriMed. The company’s report, dated 17 February 2025, concluded all four monkeys that received the therapy, whether at low or high doses, developed moderate to severe liver damage.
That data alone “should have triggered additional studies at lower doses to determine the maximally tolerated dose,” says James Wilson, the former University of Pennsylvania researcher who led the ill-fated Gelsinger trial and who is now president of Gemma Biotherapeutics, a gene therapy company.
One monkey, which received a high dose of the viruses, also showed kidney damage. This could have been a result of a well-known reaction to AAV gene therapy: thrombotic microangiopathy, a pattern of microscopic blood clots that can be fatal. Because the team at PriMed did not collect data during the key window in the first month after the monkey’s infusion, it is impossible to be certain of the cause.
Under China’s dual-track biomedical regulatory system, noncommercial, investigator-initiated trials at major hospitals can test novel gene therapies without the kind of rigorous review from national drug regulators that is required in the U.S. The rules have varied over time, but investigators typically register the study in a national clinical trial database and undergo a scientific and ethical review from the local institution, which can vary in their attention to detail.
The hospital ethics committee did not review the final report from the primate safety study before it approved Mei’s trial, according to documents from the Shanghai Yangpu District Health Commission. Jason and Linda say Qiu described the findings to them but seemed unconcerned about their significance.
The trial would move forward. “The contract signing with Xinhua Hospital is basically complete,” Qiu wrote Jason in a WeChat message that January. “Next we’ll have the kickoff meeting, sign the informed consent, and then administer the drug within 30 days.”
On 19 February, the family arrived at Xinhua Hospital to sign the consent document and run some tests on Mei’s blood. The form described thrombotic microangiopathy as a possible outcome that would require “prompt treatment.” The children’s version—the one with the “book” metaphor—asked Mei herself to check a box marked yes or no. “If you don’t want to write,” it told her, “you can draw a smiley face instead of your name.”
According to the parents’ recording, Yongguo Yu, the medical doctor overseeing the clinical side of the study, told them he had selected a doctor who performed 3000 spinal injections a year to deliver the gene therapy. Two days earlier, as part of a clinical trial testing Qiu’s Rett syndrome gene therapy, this doctor had performed the same procedure in a child without incident.
Yu also indicated the biggest risk for Mei was the possibility that she had antibodies to the viral vector. “Other than that,” he said, “the data shows it’s relatively safe.” (Yu did not respond to multiple requests for comment.)
A test before the trial started found Mei had no antibodies. Still, to blunt the body’s immune response to the viral infusion, she would get a short course of prednisone, a steroid, according to the informed consent document. Although some AAV-based gene therapies approved by FDA are still delivered with prednisone alone, gene therapists are increasingly erring on the side of caution and incorporating more powerful immunosuppressant drugs, such as rapamycin and rituximab.
That decision, and others, might have gotten more scrutiny if the trial had been reviewed as a commercially sponsored study by the National Medical Products Administration—China’s version of FDA—rather than an investigator-initiated trial.
In either type of trial, charging for an unproven therapy is forbidden under Chinese law. The informed consent noted that “Participation in this study will not incur any additional costs to you. All expenses will be borne by the sponsor, Lanqi Xintu Gene Technology.”
But the family knew better. Yu seemed to recognize the contradiction during his conversation with them. “On the surface, it’s not money you’re paying, but actually it’s all being paid by the company,” he said at one point. “If you’ve contributed, you have stake in it.”
On 23 March 2025, the day before the procedure, the family settled into a shared room inside a crowded pediatric ward. When the other girl in the room broke into tears, Mei handed her one of her favorite stuffed animals: a blue bird. “She liked to give things to other kids,” Jason said.
The next day Mei put on a brave face as she was taken to the procedure room. Yu’s handpicked doctor inserted a needle between two vertebrae on her lower back and withdrew just over a teaspoon of spinal fluid to create space for the drug. Then, he swapped syringes and gently pressed down on the plunger, filling her spinal canal with viruses.
After 3 days, Mei developed a fever, a common reaction to AAVs that was expected to last for a couple days. Qiu swung by as her fever began to wane. He assured Jason and Linda that their daughter’s condition would soon improve. As they walked him to the elevator, he announced—excitedly—that the Nature paper had been accepted with revisions. Then he added that Mei’s treatment would make an even bigger splash: the first brain-directed gene-editing therapy.
Mei did not improve. Because she hadn’t urinated since the infusion—a sign that her kidneys had been seriously injured—her doctors had limited her fluid intake, and she became desperately thirsty. The platelets in her blood also dropped to dangerous levels. It was the exact sequence of symptoms that the consent form had warned the family about.
But nowhere did the form explicitly indicate that any of this could end in death, nor did that come up during any conversations with Qiu or the other doctors, Jason and Linda say. “Death should always be mentioned in a first-in-human trial,” says Greely, the bioethicist.
Doctors took Mei for a full-body CT scan and then rushed her to the intensive care unit. Her brave face had begun to crack. “Mom, I want to go home,” she told her mother, before vanishing into the ward.
Over the next 24 hours, Jason and Linda sat in the waiting room with their elderly parents. Qiu, who had returned upon hearing of Mei’s problems, seemed unable to process it. He was still with the family the next morning when a doctor emerged to tell them Mei had died. Qiu was in shock, Jason recalled.
So was the family. They stood by Mei’s body and caressed her one last time.
Two days later, the hospital’s ethics board convened an emergency meeting and concluded Mei’s death was “definitely related” to the treatment, according to its report. The cause of death: thrombotic microangiopathy.
Over the next few weeks, Linda and Jason say Qiu expressed his sadness to the family, but he was unwilling to speculate on what went wrong. The parents say they asked him to withdraw the Nature paper out of concern that it could mislead other families with a similarly affected child, as well as researchers. Qiu initially agreed. Eventually, they say, he stopped responding to their queries. Yang returned the full payment he received for the work.
In September 2025, the district health department fined the hospital approximately $3600 for failing to properly oversee the trial and not registering it as commercially sponsored research on the national database. As the responsible physician, Yu was named in the report, and the hospital required him to receive verbal counseling, according to the family.
Although Qiu’s company had taken out insurance for the trial, the family said they were offered no compensation. They had not asked for it, they said; instead, they blamed themselves for not asking more questions and for trusting Qiu’s authority so completely.
The muted response to Mei’s death sharply contrasts with the Gelsinger tragedy decades ago. His death drew widespread headlines and prompted a major self-examination by the gene therapy field. The university where the trial was conducted paid a fine of more than a half-million dollars and had further gene therapies suspended. FDA barred Wilson, the principal investigator, from human research studies for 5 years for violations that included failing to disclose the adverse reactions observed in the primate toxicology studies on Gelsinger’s informed consent form. “Public disclosure … is essential in safely moving these fields forward,” Wilson says.
Within days of their daughter’s death, Jason and Linda moved into a hotel and, later, a new apartment. Locked away in the old apartment was all their furniture and their daughter’s possessions. They also cut off contact with some of their closest friends. “I don’t know how to reply to them, so I pretend not to see their messages,” Linda says. “I am not so brave to face everything.”
Jason and Linda were still mourning the loss of Mei when they learned in February that a version of the Nature study had been published after a review process that had stretched over a year. The family’s genetic data had been removed and a sentence thanking them for their “participation and support” had been eliminated.
The paper still described a base-editing therapy directed at their daughter’s mutation, but it featured entirely new data on its ability to work in monkey brains. “I remain very skeptical of the data in this figure,” Gray says. Qiu’s team had added control samples at the request of peer reviewers. But the controls appear to have been processed at a different time from the treated samples, using different microscope settings, according to Mike Rossner, an image integrity expert. That makes any comparison challenging.
All the original mouse experiments, which were funded by the family, also appear to have been redone. Campeau, who reviewed the original manuscript for Nature, said he hadn’t noticed the extent of the changes when the revised version landed in his inbox, and he had approved it. Notably, the paper also did not mention any of the kidney and liver troubles from the primate safety study of the therapy.
- Hank Greely
- Stanford University
The paper had an enthusiastic reception. “These promising results might pave the way for the development of an effective clinical treatment,” Kevin Bender, a neuroscientist at UC San Francisco, wrote in an accompanying commentary. At the time, Bender had no idea that a girl had received it and was already dead. Meanwhile, Chinese state media, CCTV, called the work “the first ray of hope” for “countless families suffering such diseases.” Some families in Jason and Linda’s WeChat group said they planned to reach out to Qiu about their own children.
When Jason and Linda saw that, they decided to act. They submitted a complaint to the dean of Qiu’s university, demanding an investigation and a retraction of the Nature paper “to prevent more children from suffering and our blood and tears flowing in vain.”
Gray, Sanders, and several others who reviewed the paper for this story believe Nature should request and analyze the raw data and image files from the authors. They also say the study’s sources of funding should be fully disclosed. “To not acknowledge one of the sources of funding should be enough for retraction,” Sanders says.
Greely observes that it would be unreasonable for the family to have veto power over the publication, but he would like to know whether Qiu had told Nature’s editors about the outcome of the clinical trial and why the family was removed from the paper. In a June response to an email from the parents noting their child’s death and other concerns, Nature’s deputy editor, Victoria Aranda, wrote that the ethical issues they raised “fall outside our purview in terms of data integrity” and that the matter should be handled by the university.
In a more recent statement responding to queries from Science and Retraction Watch, the journal said it was not informed of these issues or the sanctions the hospital had received before it published the paper. “Had information relevant to the editorial assessment of the manuscript been disclosed during consideration of the paper, it would have been carefully evaluated as part of our decision-making process,” noted Francesca Cesari, Nature’s chief biological, clinical, and social sciences editor.
On 30 April, Jason and Linda received an emailed reply from the university, followed by a phone call: The dean’s office was taking no action against Qiu. The payments made to Yang were simply “research service fees” for the clinical trial, though he was given a formal admonitory talk for his behavior. The Nature paper, the university wrote, “bears no relation to the experiments you funded.”
During a Zoom call last month, framed by the bare, white walls of their home, Jason and Linda described how each had coped with Mei’s death. At times during the conversation, Linda became overwhelmed and turned away from the camera, as Jason pressed on. They recalled that when Mei’s birthday came around earlier this year, they said nothing to each other, though they both assumed it was on the other’s mind.
“We do not discuss that,” Linda said.
During the call, Linda finally told her husband how she had spent that day. While he was at work, she finally returned to the old apartment and sat down on Mei’s bed surrounded by all her things. She had purchased a chocolate cupcake on the way and now picked at it as she imagined that her daughter was still at her side.
By the time she finished telling her story, Jason was welling up. Linda glanced over at him tenderly. “My husband didn’t know this,” she said. “I cannot cry every day, but that day, I think, I can cry in the room.”