1998


From: New Scientist

False Security: You Can't Always Believe The Results Of A Genetic Test

Many laboratories offering genetic tests for cystic fibrosis don't test for enough mutations to give reliable results, according to a new study.

Last year, a panel convened by the US National Institutes of Health (NIH) recommended that prospective parents be routinely offered tests to see if they carry mutations that could give their children cystic fibrosis. This was the first time a genetic test had been recommended for the general population (This Week, 26 April 1997, p 12). The new findings, from a team of geneticists led by Wayne Grody of the University of California at Los Angeles, suggest that this recommendation was premature.

Cystic fibrosis, in which the lungs and airways become clogged with mucus and prone to infections, is caused by defects in a single large gene. There are hundreds of these mutations. Some are very rare, and certain mutations are more common in particular ethnic groups.

To be effective, the widespread screening urged by the NIH panel would have to test for many of the most common mutations. Exactly how many is still a matter of debate. "One thing that was conspicuously lacking from the panel's recommendation was how many mutations should be tested for," says Grody, who is co-chair of a committee drawing up guidelines on the issue for the American College of Medical Genetics (ACMG). To identify 90 per cent or more of people carrying cystic fibrosis genes, it may be necessary to test for 20 of the most common mutations, Grody predicts.

Grody also leads a team of geneticists monitoring the testing labs on behalf of the ACMG and the College of American Pathologists. The team's latest survey, published in The American Journal of Human Genetics (vol 62, p 1252), reveals a disturbing picture. The investigators asked 45 labs how many mutations in the cystic fibrosis gene they usually test for. Forty-three responded, reporting numbers ranging from just one to 70. It was most common to test for between 12 and 14 mutations, and only 10 labs tested for more than 20 (see Figure).

Even some of the labs that looked for a respectable number of mutations didn't include some of the more common ones. "We wanted to bring this to people's attention," says Walter Noll, a pathologist at the Dartmouth-Hitchcock Medical Center in Lebanon, New Hampshire, and a member of Grody's team.

Noll stresses that the survey did not distinguish between labs that tested commercially and those that tested for research purposes. Some of those testing for few mutations might be doing so as part of a research project, he points out.

But experts in genetic testing agree that there is a urgent need to set standards for testing labs, given the NIH panel's recommendation and growing public awareness of the availability of genetic tests in other countries.

Maurice Super, a consultant paediatric geneticist at the Royal Manchester Children's Hospital, says he has noticed an upsurge in interest in testing for cystic fibrosis in Britain in the past few years, since commercial labs began advertising their services in consumer magazines. In Britain, as in the US, there are no official guidelines on the number of mutations that should be tested for.




This article comes from Science Blog. Copyright � 2004
http://www.scienceblog.com/community