Less than two years after producing an exhaustive report on human genome editing, the U.S. National Academy of Medicine and the National Academy of Sciences are planning an international commission on the most controversial use of that technology — creating “CRISPR babies,” medicine academy president Dr. Victor Dzau announced on Thursday at the World Economic Forum in Davos, Switzerland.
The recent birth in China of twin girls whose genes had been edited while they were embryos highlighted the shortcomings of the existing guidelines, he said. The 2017 academy report and others on embryo editing “were not clear enough” on when it would be scientifically and ethically acceptable to permanently alter the DNA of an embryo, sperm, or egg in such a way as to make it heritable by all of that individual’s children, Dzau said, and too open to interpretation.
For instance, criteria in the 2017 report include that any such germline editing be carried out only in the “absence of reasonable alternatives,” address a “serious disease,” and make safety paramount. He Jiankui, the Chinese scientist who used CRISPR-Cas9 to alter embryos’ DNA in a way intended to prevent HIV infection, has claimed he met all of those criteria.
Dzau called that a “rationalization,” adding, “All of these things need to be revisited, and that’s what we plan to do.”
The commission is still in the planning stages, but Britain’s Royal Society, the Chinese Academy of Sciences, and other scientific and medical groups outside the U.S. have signed on, he said.
A key question, virtually unaddressed in the many CRISPR discussions and reports and white papers so far, is what specific scientific tests must be done to determine whether edits made in a human embryo are “safe.” In He’s experiment, CRISPR did not change the embryos’ HIV-related gene (CCR5) into a form that’s carried by about 10 percent of Europeans and that protects against infection by the AIDS-causing virus. Instead, he got genome edits that experts have called a chaotic mess with unknown consequences for the health of the two girls born from the embryos last November.
In a case like that, Dzau said, perhaps a scientist should be required to conduct “multigenerational observations in [lab] animals” to determine the effects of such never-seen-before edits.
At the International Summit on Human Genome Editing in Hong Kong where He dropped his bombshell two months ago, organizers called for the creation of an international body to address such technical issues. The planned Academies commission would lay the groundwork for that, legal scholar Alta Charo of the University of Wisconsin told STAT. She co-chaired the academy’s 2017 genome editing report and has been involved in informal conversations about how to set up the new project.
Among the other scientific questions that should be addressed, she said, are what kind of experiments would be required to test the safety of a genome edit, including whether those in human cells growing in lab dishes are sufficient, or whether tests in lab animals, perhaps followed for years, would be required.
“I do think the He case shows that clarification [on germline editing guidelines] would be useful,” legal scholar and bioethicist Hank Greely of Stanford University told STAT. “He’s efforts to bring himself under the 2017 conditions [in the National Academies report] were almost laughable — but not quite. Clarification could make that kind of rationalization more difficult.”
Experts say criteria for germline editing need to address how thoroughly to scrutinize a genome for unintended edits (“off-target effects”), for intended edits (that is, at the correct site in the genome) that were more extensive than expected, and for possible cancer-causing genes, and, if so, what their consequences might be.
An even more basic question should be included, Dzau argued at Davos: Is is safer and more effective to edit eggs, or sperm, or the very early human embryo? In He’s experiment, the embryos became two cells faster than CRISPR could work, with the result that some of the girls’ cells (descendants of the un-CRISPR’ed cell) do not contain the supposed HIV-protecting edit. Editing sperm or eggs would presumably prevent such “mosaicism.”
The commission Dzau described would also address more value-laden questions. Presumably an inherited, lethal disease would qualify for germline editing, he said. But if existing reproductive technologies, such as genetic analysis of IVF embryos and implantation only of one without the disease-causing mutation, can give parents a good chance at having a healthy baby, he asked, should it be mandatory to try those first?
Similarly, criteria such as “unmet” or “serious” medical need are highly subjective. In justifying his experiment, He argued that HIV/AIDS carried such stigma in China that protecting babies from ever acquiring the virus cleared that bar. “What’s an unmet medical need may vary by country,” Charo said.