A new mass spectrometry technique could pave the way for prenatal diagnosis by a simple and safe test on a pregnant woman's blood, according to an author of a paper describing the technique published in PNAS this week.

Y.M. Dennis Lo, professor of chemical pathology at the Chinese University of Hong Kong, developed the technique, which is based on single-allele base extension reactions and mass spectrometry. It allows for the reliable detection of fetal-specific alleles, including point mutations and single nucleotide polymorphisms, in maternal plasma, according to the authors. The result is that prenatal diagnosis of a genetic disease could be achieved by a simple blood test on the mother-to-be.

Lo's team built on the 1997 discovery that 3 to 5% of the DNA in the cell-free plasma of the mother comes from the fetus. Until that time, people attempting prenatal diagnosis on circulating fetal cells met with limited success, as these are present only in about one in a million cells or less, Lo told The Scientist. "People wouldn't have suspected that DNA would be present in the cell-free fraction, because everybody thinks that DNA should be inside cells," he said.

The approach was applied to diagnose the fetal inheritance of the four most common Southeast Asian beta-thalassemia mutations at 7 to 21 weeks' gestation. "This is quite a common condition: approximately 2.5% of Chinese in Hong Kong carry it," said Lo. Fetal genotypes were correctly predicted in all cases studied.

To be able to diagnose common genetic disorders prenatally, the method has to be extremely sensitive, Lo said. "You are talking about a much smaller degree of difference between mother and baby [with these disorders], sometimes even as little as a single base pair," he said. Detection or exclusion of the paternally inherited mutant allele in maternal plasma as a means of diagnosing these disorders has until now only been possible if the maternal and paternal mutations were different. By examining the polymorphisms linked to a particular mutation instead, the problem is solved, according to Lo. "So we overcome yet another of the perceived limitations of this field. We cracked two things: first the single base limitation, and second the same maternal–paternal mutation," he said.

"It really does indicate that mass spectrometry is a very powerful technique," said Neil Avent, research director at the Centre for Research in Biomedicine, Bristol Genomics Research Institute. Previous attempts to use real-time polymerase chain reaction were abandoned because of the problem of mispriming with an allele-specific primer, said Avent, who wasn't involved in the study.

But Avent said that with mass spectrometry, the result you get is real. "If that allele is there, you get a mass which is precisely what has been amplified," he said. "So we know that were dealing with the right thing—so clearly, to me, it's a very convincing piece of work."

"[The technique] improves sensitivity and specificity—but at a cost," said Diana Bianchi, chief of medical genetics and professor of pediatrics and obstetrics and gynecology at Tufts New England Medical Center. "However, I think the way [the technique] might be useful in widespread clinical use is that it does provide a means of assuring that you do have fetal DNA present in the sample—one of the big concerns that we've had working in this field."

Lo suggested that the technique could also be used for sex selection. But he insisted that use of the technology for prenatal sex selection would be unacceptable to him. "At the moment, we envisage that this technology would only be justified in the case of a serious disease which will threaten the fetal well being—like the thalassemias, like sex-linked disease," Lo told The Scientist.

Still, Philippa Taylor, associate director of the Centre for Bioethics and Public Policy, worried that the technology, like any, could be abused. "Who is actually going to decide which diseases or disability we allow tests to take place for?" she asked. The question, she said, was where the limits would be drawn for which tests, and what society then allows abortion for. "What's driving it all, this question of a totally healthy society? We do need to step back a bit and see what's driving this quest for the perfect prenatal diagnosis," said Taylor, who was not involved in the study.