Read our COVID-19 research and news.

He Jiankui (center) during a Q&A after his presentation in Hong Kong, China, today; he was joined by meeting co-organizer Robin Lovell-Badge (left) and pediatric stem cell researcher Matthew Porteus (right).

Imaginechina via AP Images

Researcher who created CRISPR twins defends his work but leaves many questions unanswered

HONG KONG, CHINA—The researcher who set off a global firestorm 2 days ago when he announced the birth of the world’s first gene-edited babies defended his study at a meeting here this morning. He Jiankui of the Southern University of Science and Technology in nearby Shenzhen, China, said he was “proud” of the work, which he said could lead to disease prevention “for millions of children,” and provided some details about the unpublished research that many scientists and bioethicists—as well as a phalanx of journalists—were clamoring to hear. Thousands around the world watched a webcast of his talk.

He dropped another bombshell, acknowledging that a second woman is in the early stages of a pregnancy with a gene-edited embryo.

But He, who used the genome-editing technique CRISPR in his attempt to make twin girls resistant to HIV infection, failed to provide many key details about the trial, was evasive about the ethical debate that had preceded the study, and provided sometimes confusing answers about the balance between risks and purported benefits. And his talk immediately triggered fresh criticism. The work should be “considered irresponsible,” biologist David Baltimore of the California Institute of Technology in Pasadena said after hearing the presentation. “I don’t think it has been a transparent process, we only found out about it after it happened … after the children were born. I personally don’t think it was medically necessary,” Baltimore added, reflecting concerns that were widely voiced even before any details of the study had become available. “I think there has been a failure of self-regulation by the scientific community because of a lack of transparency.”

For this specific case, I feel proud, actually. I feel proudest because [the father] had lost hope for life.

He Jiankui, Southern University of Science and Technology

He’s talk took place at the International Summit on Human Genome Editing here, a global forum to discuss scientific progress and ethical issues in one of the most fraught areas of biomedical research. Earlier this week, He announced in an Associated Press (AP) interview and a series of YouTube videos that his team had engineered the genomes of twin baby girls to cripple a key receptor, CCR5, that HIV uses to infect white blood cells—a modification they may pass onto their descendants.

The girls’ father is infected with HIV, but the purported justification for the study was not to protect the girls from the AIDS virus during pregnancy. The parents used in vitro fertilization and the father’s sperm was washed to remove HIV, which virtually eliminated the risk of the embryos being infected. He’s goal was to protect them later in life, he told AP, even though experts agree there are other safe and effective ways to prevent acquiring HIV.

The news completely dominated discussions at the summit even before He appeared; to accommodate the sudden interest, a special 1-hour time slot was carved out for He today. The hall was filled to capacity; hundreds of videographers, photographers, and journalists crammed shoulder-to-shoulder into one area. “I have the right to just cancel the session if there’s too much noise or interruption,” moderator Robin Lovell-Badge of The Francis Crick Institute in London warned at the outset. When He started his talk, clicking shutters nearly drowned him out, prompting Lovell-Badge to repeat his warning.

He started by apologizing for the way his claims “leaked unexpectedly” earlier this week; he said a paper has been submitted to a peer-reviewed journal. (News of the breakthrough was first reported by MIT Technology Review, after which AP released a story about the work, and He made the YouTube videos and several documents related to the experiment public.)

He also justified his research target, explaining that HIV/AIDS still imposes a heavy disease burden on much of Africa, and those infected often face severe discrimination. Mutations in the CCR5 gene confer HIV resistance, and He said previous experiments have suggested CCR5 was a promising target for editing. But He didn’t explain how gene editing could help the world prevent the disease; nobody thinks it’s feasible to edit the genomes of entire populations.

He outlined how he took his research from work in mice to nonhuman primates to human embryos in vitro—much of which, he pointed out, he had presented at previous talks. Then he added some details about the human studies. A large focus of his research was on optimizing the precision with which CRISPR cuts the genome to minimize the possibility of untargeted genes being altered.

He recruited couples that wanted to start a family in which the man was infected with HIV and the woman was not. There was a two-step informed consent process with a team member first visiting the couple at their home, and then a discussion of an hour or more at He’s lab in Shenzhen. In the case of the couple that eventually produced the twins, He collected eggs from the woman, injected them each with a sperm cell from her husband, and soon after used CRISPR in attempts to modify four of the resulting embryos. In one, both copies of the CCR5 gene were disrupted; in another, CRISPR only crippled one of the copies, leaving one copy of the gene intact. The couple agreed to have both of these embryos implanted and both resulted in live births.

In the wide-ranging Q&A that followed, Lovell-Badge noted that the CCR5 gene has multiple functions in the immune system, that its absence leaves individuals more at risk for West Nile virus and potentially other diseases. Did He “really know enough about CCR5 and its role in the immune system to choose that for therapy?” Lovell-Badge asked. He responded that his team chose CCR5 because it is a well-understood single gene target appropriate for a first trial before moving to multiple-gene diseases. But he didn’t address the question of potential inadvertent side effects.

“I just don’t see an unmet medical need for these girls,” said David Liu, a biochemist at the Broad Institute in Cambridge, Massachusetts. The technology could benefit millions of children potentially exposed to HIV from their parents, He said, without explaining this further. Speaking about the babies, he added, somewhat cryptically: “For this specific case, I feel proud. I feel proudest, because Mark, [the babies’ HIV-infected father], had lost hope for life. But with this protection, he sent a message saying he will work hard, earn money, and take care of his two daughters and his wife for this life.” He did not address the fact that safe and effective ways to prevent HIV transmission exist, or why gene editing might be preferable.

I think there has been a failure of self-regulation by the scientific community because of a lack of transparency.

David Baltimore, California Institute of Technology

This set a pattern for the rest of the Q&A: He answered some questions but left many partially unanswered—or didn’t give an answer at all. He went off on a tangent when asked directly why he had chosen to conduct a study that international scientific consensus said shouldn’t be done. He was vague about whether the study had been approved by an institutional review board. He said he is personally paying for the medical expenses of the study participants, but didn’t say how much or whether he can continue that support long-term. It’s still unclear where he conducted the lab work and which hospital was involved.

Questions from the floor indicated widespread dissatisfaction with his answers on his informed consent procedures. He said four people had reviewed the consent form before it was given to the couples; in the United States, “hundreds of people would review a consent form” before it would be put into use, says Matthew Porteus, a pediatric stem cell researcher at Stanford University in Palo Alto, California. Although He’s outline of his work overall indicated a “reasonable approach,” in going from mice to nonhuman primates to humans, Porteus adds, “there are still gaps” in the information provided, such as details on how many eggs were used, the success rate in the gene modification step, whether there had been any miscarriages, and the date of the births.

In addition, “There are questions about the secrecy,” Porteus says. For instance, He claimed that he had discussed his work at several conferences. “But I was at those conferences and He didn’t indicate he was moving in this direction.” Unless He produces more information and better answers, “he risks becoming a pariah,” Porteus says.

The fact that it’s possible that the first instance of human germline editing came forward as a misstep, should in no way lead us to stick our heads in the sand.

George Daley, Harvard Medical School

Later in the day, the meeting's organizers said He had withdrawn from a planned breakout session on embryo and germline editing to take place Thursday.

With pending investigations by national and local authorities and his university, He’s saga is set to roil the scientific and biomedical community for months or even years to come. What impact his go-it-alone approach may have on the field remains to be seen. Appearing at the summit in the morning session, before He’s presentation, George Daley of Harvard Medical School in Boston expressed concern but urged researchers to have faith in the self-regulation of the research community while acknowledging that this requires transparency and external oversight.

“The fact that it’s possible that the first instance of human germline editing came forward as a misstep, should in no way lead us to stick our heads in the sand,” Daley said. Rather than go on the defensive, he proposed “that it’s time for us to at least consider a responsible pathway for clinical translation” of germline editing.

With reporting by Jon Cohen.

*Update, 28 November, 2 p.m.: This story has been updated with additional information.