News & AnalysisAssisted Reproduction

FDA Considers Trials of 'Three-Parent Embryos'

See allHide authors and affiliations

Science  21 Feb 2014:
Vol. 343, Issue 6173, pp. 827-828
DOI: 10.1126/science.343.6173.827

An experimental technique that manipulates a woman's DNA could spare her from passing on a potentially deadly disease to her children. But the technique breaks new and ethically fraught ground: It would create a child that has DNA from "three parents"—the mother, the father, and an egg donor. And any daughter could in turn pass on the new DNA mix to future generations. Until now, procedures that produce inheritable gene alterations have been ethically taboo.

Now, regulators on both sides of the Atlantic are grappling with whether to allow the first human trials of the technique, called mitochondrial DNA replacement therapy, to go forward. In the United Kingdom, the government has given the technique a cautious endorsement. And at a meeting on 25 and 26 February, an advisory committee to the U.S. Food and Drug Administration (FDA) will consider the topic. It is expected to draft its recommendations in the coming months, which the agency will draw on to develop regulations.

Controversial therapy.

Mitochondrial DNA replacement could help carriers of severe disease have healthy children.


FDA's Cellular, Tissue, and Gene Therapies Advisory Committee, which includes doctors, researchers, and representatives from industry and patient groups, will weigh whether the technique is effective and safe enough to try in humans. Animal models are imperfect, and cell-based studies give only limited clues about possible long-term side effects.

Next week's meeting will focus on safety, but there are equally knotty ethical issues for regulators to consider: whether the benefit of averting severe disease overrides societal objections to changing the human germ line. Another is how the ethical calculus is affected by another potential use of mitochondrial DNA replacement: to treat age-related infertility, which would widen the potential pool of patients to include millions of women. "Once it's used, it will be used in all sorts of ways for all sorts of people. That's the reality of this kind of medicine," says bioethicist Jeffrey Kahn of Johns Hopkins University in Baltimore, Maryland.

Pushing to go forward are patients with mitochondrial diseases and the doctors who care for them. "I have nothing to offer these people, and that's heartbreaking," says Douglas Wallace, an expert in mitochondrial genetics at the University of Pennsylvania. The diseases arise when something goes wrong in the mitochondria, the organelles that provide energy to cells. Mitochondria carry their own set of genes, called mitochondrial DNA (mtDNA), and mutations in those genes cause many of the syndromes. But because each cell contains many mitochondria, and each mitochondrion has up to 10 copies of its genome, some cells and tissues can acquire more faulty genes than others. This randomness means that apparently healthy women can be carriers of mitochondrial disease, not discovering the problem until they give birth to children who are very ill.

Because brain, muscles, and heart require so much energy, symptoms often appear in those tissues first. Some cases are diagnosed at birth and are soon fatal. Others don't appear until adulthood. There are no treatments; doctors attempt to alleviate symptoms with antiseizure medications or physical therapy.

Although sperm carry mitochondria, they are usually degraded shortly after fertilization, so mitochondrial diseases are passed down through the mother. To avoid having a desperately ill child, women who carry severe mitochondrial mutations can adopt, undergo in vitro fertilization (IVF) with a donor egg cell, or allow embryos or fetuses to be tested for mitochondrial defects—although such tests aren't very accurate. Several research groups say the best way for carriers to have a healthy, genetically related child could be to remove the nuclear genes from an egg with mutated mitochondria and place them into a donor egg with healthy mitochondria (see diagram). Using different approaches, these groups have made enough progress in animals and with human cells that they would like to have permission to try the techniques in patients. In the United Kingdom, current law prohibits IVF techniques that alter the DNA of an embryo. But since 2011, both ethical and scientific review panels have said mtDNA replacement research should be allowed to proceed for women who are carriers of severe disease, and the government has agreed. The Department of Health is expected to release a proposed law this spring, and Parliament could vote on a final version later this year.

So far so good.

Mito and Tracker, the first rhesus monkeys born after mtDNA replacement, seem healthy so far. They will reach adulthood, and breeding age, when they turn 5 later this year.


In the United States, FDA has said it has the power to regulate any transfer of mitochondrial DNA in embryos, because it is a form of gene therapy. At least one researcher, Shoukhrat Mitalipov of Oregon Health & Science University in Beaverton, has asked the agency for guidance on what a clinical trial would require. (Mitalipov's lab has produced seven monkeys born after mtDNA replacement.)

Kahn, the ethicist, puts the key issue facing the panel this way. By mixing new DNA into the germ line, "we're not treating humans. We're creating humans. There's not a model for that." Then there is the thorny issue of whether the technique should be used to treat infertility. Some reproductive biologists think that faulty mitochondria—or perhaps other factors in the egg cytoplasm that are exchanged in mtDNA replacement techniques—might be one of the key reasons fertility falls in women in their late 30s. They say that the procedure could help such women conceive, but critics say there is little animal data to support the idea.

"No studies clearly indicate the efficacy of treatment" with the technique, Mitalipov says. But clinical trials "would be the way to test it." He would like to go forward with trials in both disease carriers and infertile women. "Mitochondrial diseases are rare, but age-related infertility is a huge problem."

Others say existing safety data are insufficient to green light any clinical trial. One worry is that a mismatch between nuclear and mtDNA could cause health problems in children conceived with the technique (Science, 20 September 2013, p. 1345). Many nuclear and mitochondrial genes work together and depend on each other. In some animal studies, says Klaus Reinhardt of the University of Tübingen in Germany, mismatched nuclear and mtDNA caused a range of problems, including infertility, especially in male offspring.

Wallace thinks there is "some possibility of incompatibility." To minimize the risk, he would recommend finding egg donors who have a similar mitochondrial DNA type as the potential mother.

But Robin Lovell-Badge, a developmental biologist at the MRC National Institute for Medical Research in London who led a U.K. scientific review of the technique, says the studies that found developmental and health problems were in inbred strains of animals and are unlikely to apply to humans. Other studies have produced apparently healthy animals, he says.

Mitalipov says no problems have appeared in the rhesus monkeys his lab produced, even though the egg donor and the nuclear donor were from different macaque subspecies. "We monitor them very closely, but so far the monkeys were pretty normal," he says. Whether relatively subtle effects like cognitive defects would show up in monkeys is unclear, however. Nor would fertility problems be apparent in the monkeys so far.

Some children have in fact been born with mitochondria from unrelated donors. In the late 1990s, a fertility clinic in New Jersey treated the unfertilized eggs of infertile women by injecting them with cell cytoplasm from eggs from fertile donors. Along with proteins and other cellular material, mitochondria from the donor were injected, resulting in children who carried a mix of mitochondria from their mother and from the egg donor.

More than a dozen children were born after the procedure, but several had developmental problems. One was missing one copy of the X chromosome, and another developed pervasive development disorder, a spectrum of conditions that includes autism. The treatments were not part of a controlled trial, however, so it is impossible to say whether the cytoplasm injection actually helped the women become pregnant or whether it was the cause of the developmental problems. The clinic stopped the practice in 2001 when FDA said it had the authority to regulate the practice as germline gene therapy.

Many mtDNA disease carriers, who otherwise face a virtual certainty of having a severely ill child, would be eager to accept a low risk of more minor problems, Wallace says. But, he adds, the risk-benefit ratio for society at large is different. "It's going to be hard to find what a fair balance is."

View Abstract

Stay Connected to Science

Navigate This Article