‘Consumer’ DNA analysis of limited use

Rosie Redfield ponied up $99 to the consumer DNA analysis firm 23andMe and spit into their test tube. What she got back was, well, boring. The test could have revealed any one of a handful of highly predictive markers for inherited disease or cancer. But 23andMe looked at one million locations on Redfield's genome for mutations and found nothing interesting at all, aside from a very slightly elevated risk of Parkinson's Disease.

"I think that has been the case with most people," said Redfield, a microbiologist at the University of B.C. who teaches an online course called Useful Genetics. "Had they found something, I would have taken it to my doctor."

Whether your doctor will be able to do anything meaningful with the results of a consumergrade genetic report is doubtful. The resolution of such tests is far lower than a full genome sequence and American authorities have recently cast doubt on the value and accuracy of clinical interpretations provided by 23andMe.

The California-based firm was ordered late last year to stop providing health reports based on DNA analysis by the U.S. Food and Drug Administration. The FDA said 23andMe was providing a test intended to diagnose disease without proper approval and that the results may cause people to seek improper medical treatment.

The FDA was specifically concerned that a false result for something like warfarin sensitivity might induce people to change their dosage or abandon therapy. They also suggested that a false positive for a cancer-related mutation could trigger a wave of unnecessary testing when people bring the results to their personal physicians, while a false negative result could lead people to abandon proper medical screening.

'Lack of standards' Because Redfield bought her medical analysis well before the FDA shut down that aspect of the business, she retains access online to her data, which is regularly updated as the science on the impact of the genetic mutations in her data improves.

Several groups of researchers have attempted to assess the reliability of consumer-level genetic testing. They reported that the medical interpretation provided from identical samples analyzed by different companies showed significant variability and even outright disagreement, with one firm reporting elevated risk of a disease, while the other reported lowered risk.

In most cases, the absolute risk of having a disease associated with a mutation detected by the test was much less than one per cent.

Researchers writing in the North Carolina Medical Journal concluded "although customers might find their risk profiles interesting, this information provides no guidance for physicians trying to make informed clinical decisions."

"Part of the problem with direct-to-consumer is the lack of standards in that industry," said Brad Popovich, chief scientific officer of Genome British Columbia.

See the rest here:
'Consumer' DNA analysis of limited use