It has the ring of airtight logic: Screen embryos for chromosomal abnormalities and implant only those that appear healthy. But a new study suggests that this strategy, currently used by hundreds of fertility clinics worldwide, is worse than useless, making it even harder for a couple struggling with infertility to have a child. The study of several hundred Dutch women is the largest of its kind to test preimplantation genetic screening (PGS), but fertility specialists are fighting back, saying it has fatal flaws.
PGS grew out of an identical approach used for a different purpose. In the early 1990s, doctors began screening embryos from couples who knew they carried a specific disease gene. That way, only embryos that didn't harbor two copies of the defective gene would be implanted into the mother, promising a baby without a dreaded disease.
Fertility specialists soon began wondering if this approach could be applied to a much larger population: women, particularly those over 35, who have relatively few healthy eggs and are struggling to conceive a child. Embryos with chromosome abnormalities could be discarded, whereas those that had the correct chromosome number and appeared healthy could be implanted. Ostensibly, this should increase the odds of getting pregnant. PGS quickly caught on. A 2006 survey of 186 U.S. fertility clinics found that two-thirds offered the technique to infertile couples. But does it work? Only one other PGS trial with a control group has been published, in 2004, and the Belgian researchers conducting it found that PGS didn't increase the odds of getting pregnant.
Reproductive biologist Sjoerd Repping of the University of Amsterdam and his colleagues decided to launch another trial that focused on ongoing pregnancies. They randomly assigned 408 women, age 35 to 41, to receive several cycles of in vitro fertilization (IVF) alone or IVF along with PGS. As is fairly standard, to perform PGS, Repping's group extracted one cell from an embryo 3 days after fertilization and examined eight chromosomes for abnormalities that would doom an embryo. Only those free of defects were implanted.
"Everybody, including us, was hoping that this method would increase the chances of pregnancy in these women," says Repping, "but it doesn't." In the PGS group, 24% of the women gave birth to a live baby, compared with 35% in the IVF-only group, the researchers report 5 July in the New England Journal of Medicine. They also present their findings today at the European Society of Human Reproduction and Embryology meeting in Lyon, France. Repping says it's possible the biopsy process harms the embryo, making it less likely to develop once in utero. In addition, some of the embryos may be "mosaic," says Repping, meaning that the chromosome numbers seen in the biopsied cell are not the same as the numbers in the other, unexamined cell of the same embryo. This might give some embryos a false bill of health when in fact the odds are stacked against them.
Critics of the new paper include Santiago Munné, director of Reprogenetics, a preimplantation genetic diagnosis facility in West Orange, New Jersey. Munné dismisses the findings, arguing that the authors are not sufficiently experienced at biopsying embryos. Twenty percent of embryos biopsied did not yield a conclusive result, he points out, compared to 5% in his facility. Some of those embryos may have contained genetic defects, and because they were implanted, that might have led to fewer births in the PGS group, Munné says. He is preparing a rebuttal with nine other fertility specialists to dispute the findings of the Dutch study. "It's very damaging" to the reputation of PGS, he says.
But many others in the fertility world believe that, although not perfect, the study shows that PGS for infertility should be curtailed. John Collins, a reproductive medicine specialist and professor emeritus at McMaster University in Hamilton, Ontario, says the study suggests that PGS doesn't help women in this age group. Collins sees the findings as a stark reminder that procedures such as PGS should be tested in clinical trials before they become widespread.