Login to Your Account

FDA advisory committee grapples with egg cell modification’s scientific side


By Anette Breindl
Science Editor

As an FDA advisory committee met over the past two days to discuss what it would take for clinical trials of “oocyte modification in assisted reproduction for the prevention of transmission of mitochondrial disease or treatment of infertility” to go forward, the committee was trying walk the line between sticking to their part of the issue – the science that would inform such trials – and acknowledging that the science was not the only, and not even the most important, issue about whether such trials would be a good idea.

Medically, the goal of such a procedure would be to allow women with mitochondrial disease to have biological children who do not inherit that disease. There are dozens of such diseases, with symptoms ranging from mild to fatal and even for different individuals with the same mitochondrial mutation.

The concept of oocyte modification is to create an egg cell with genetic material from two people. The nuclear DNA would come from a woman with mitochondrial disease, which would be transplanted into the cytoplasm of an egg without mitochondrial disease. That egg would then be fertilized in vitro by paternal sperm.

Public discussion of the procedure quickly zeroed in on “three-parent babies.” And the procedure would indeed create babies with genetic material from three parents – a set of nuclear chromosomes each from the mother and father, and about 16,000 base pairs, containing 37 genes, of mitochondrial DNA from a second mother. They would also endow those babies with genetic modifications that, while not germline in the most technical sense, would be passed on to the offspring of the females created by the method.

The procedure carries with it formidable ethical issues. Germline engineering, as was repeatedly noted in the public comments section of the meeting, is prohibited in more than 40 countries. It is not prohibited in the U.S., but the sorts of experiments that will be necessary to give a solid basis for clinical trials could not be funded by the National Institutes of Health under the Dickey-Wicker Amendment, and for now, the sorts of procedures that were discussed at the adcom are banned by the FDA.

FDA officials speaking at the meeting pointed out that while such issues may be what ultimately decides whether such trials are acceptable from a societal standpoint, the agency does not necessarily have the luxury of waiting to see what comes of public debates on the issue. If someone proposes such a trial, in the absence of legislation making it illegal, the agency will need to make its decision based on the scientific merits of the proposed trial, and the preclinical data supporting it.

If that seems daunting to agency officials, the discussions of the advisory committee would indicate that for now, they are in luck. The sort of preclinical data that would make approval of such a trial a realistic possibility does not exist, at least in the public domain.

In fact, even which species is most informative for preclinical studies is far from clear. Mice, monkeys, cows and even carp have been used in such studies, with each species having unique strengths and weaknesses.

Mitochondrial and nuclear DNA need to cooperate in making several proteins, and it is possible that when both sets come from a different parent, certain combinations would not work well together.

Last year, the laboratory of Shoukhrat Mitalipov, who testified at the advisory committee meeting, showed that it was possible to combine the cytoplasm of one human egg with the nuclear DNA from another, fertilize it and grow the resulting embryos to the blastocyst stage. (See BioWorld Today, Oct. 25, 2012.)

In primates, using that same method and having the embryos carried to term by a surrogate has resulted in apparently healthy and definitely cute baby monkeys.

But so far, there are only a handful of animals that have resulted from the procedure – far too few to say whether combining two eggs in this way is generally safe. To top it off, none of those monkeys have had offspring yet (though mouse work has been carried out across several generations).

Chinese studies done in carp have suggested that there can be compatibility issues between nuclear and mitochondrial DNA, but they were fairly subtle, and it took nearly 500 animals to see them. That kind of quantity of data does not yet exist in any other species.

Studies on bos taurus have shown another preclinical issue that will need to be addressed before oocyte manipulation is ready for prime time: scientists are simply not good yet at predicting which scrambled eggs will go to term and produce healthy offspring.

Larry Couture from the Beckman Research Institute said that the studies that exist so far are “seminal work,” but there is not really yet “a preponderance of evidence that this is safe and reproducible.”

Or as John Gearhart of the University of Pennsylvania put it, “this is one of the things where you need numbers, numbers, numbers.”