In the latest study, the researchers used 41 embryos that were frozen by fertility clinics. They thawed them and prompted them to grow to the eight-cell stage, then used a tiny glass pipette to remove one or two blastomeres. Most of the donor embryos continued to develop normally for two more days before they were refrozen.
The scientists created four cell lines that were able to grow into all the main tissue types in the body, including neurons and beating heart cells.
Colleagues at UC San Francisco repeated the experiment with two more embryos, but they did not use established stem cells to help the blastomeres grow. They produced a stem cell line that didn't involve any materials derived from dismantled embryos.
There is a potential flaw with the technique: Even the delicate removal of one cell could place an embryo's health at risk.
"How do you demonstrate that the embryo was not harmed?" Landis said. "Right there, you have a significant problem."
Protecting embryo health
The ideal experiment -- in which embryos that have been biopsied to make stem cells are implanted in surrogates so that the babies can be studied for medical problems -- is impossible to conduct for health and ethical reasons.
But Lanza said there is a straightforward solution.
To avoid potential risk to embryos, scientists could piggyback on PGD biopsies that clinics are performing anyway, he said. After the blastomere is removed, it could be nurtured in a dish for about 12 hours until it divides. Then one cell could be used for genetic screening and the other would be available for research.
Landis agreed that stem cells derived in this manner would appear to meet the federal standards.
Lanza said he believed that the method was safe enough to use on spare embryos from fertility clinics and still qualify for federal funding. In his experiments, 80% to 85% of the donor embryos developed to the stage where they could be implanted in a uterus -- a higher rate than at fertility clinics.
"Any clinic would be thrilled with 85%," said Dr. Mark Hughes, director of the Genesis Genetics Institute in Detroit. "Many clinics would be happy with 60%."
Landis said there was no way to know whether the embryos that didn't survive were lost because of the initial thawing or the subsequent biopsy.
Researchers could get some idea by comparing the pregnancy rates of couples that used PGD to those of couples that opted for standard in vitro fertilization.
A study published last year in the New England Journal of Medicine found that among older women who had IVF, those who also used PGD were about 30% less likely to have a baby. But specialists who perform the diagnostic procedure said that the study was flawed and that an alternative reading of the data showed that genetic screening slightly boosted the odds of having a baby.
The entire exercise of trying to satisfy the Bush policy could become moot after the presidential election if a new administration scraps the restrictions. But Lanza said the researchers couldn't wait.
"We need more lines to study," he said. "We can't afford to waste time."