Spit, genes and the FDA

The 23andMe kit.

Nearly half a million people have put their spit into a little tube in the last six years and sent that tube to a company called 23andMe. The company then genetically analyzes that saliva. The results are supposed to show customers how susceptible they are to more than 250 diseases. But the FDA now says the company hasn't proved the tests are accurate enough, and the agency is worried Americans are relying on the results instead of going to the doctor.

Dov Fox specializes in medical ethics at the University of San Diego School of Law. He says the FDA has given the company an ultimatum.

"This field of personalized medicine is really in its infancy, and its terrifically exciting what we might learn one day, but we're not there yet.  23andMe hasn't shown that their reports about your health from your genes alone are all that useful. They're just not accurate at this time in the way that the FDA requires."

About the author

Lizzie O'Leary is the new host of Marketplace Weekend.
Log in to post2 Comments

Thank you shechter for taking the words out of my mouth. Dr. Fox also somewhat misrepresents the results provided by 23andme stating that "lowers your genetic risk of 12.5% to 12.4% ..... the information you receive back from a company like 23andme doesn't tell you any of this."

The 23andme results do detail your risk and the average population risk.

Direct access to your personal genetic information for a reasonable price is valuable for several reasons beyond disease risk (e.g. they provide carrier status for a variety of inherited diseases- important for young couples hoping to start families; 23andme also provides information on drug susceptibility - important for the use of certain blood thinners and general anesthesia). This wasn't a balanced report, and it will be a shame if 23andme is prohibited from providing a direct-to-consumer service.

I appreciate Dr. Fox's example re: the BRCA gene, but I feel like those statistics mislead the audience and gloss over the potential benefits of this type of testing. Yes, the NCI confirms that the general population risk of developing breast cancer is around 12% ( http://www.cancer.gov/cancertopics/factsheet/Risk/BRCA ). Since the prevalence of the BRCA mutations in the general population is rather low ( http://ww5.komen.org/BreastCancer/InheritedGeneticMutations.html ) it is reasonable that if you don't have the BRCA mutations, your risk is only slightly less (Dr. Fox claims a reduction from 12.5% to 12.4%). However, you completely fail to report that, on the flip side, a person with a true positive test for the BRCA mutation has a much higher risk of developing cancer (45-65%). Not to get deep into Bayesian statistics, but I think a more balanced discussion is warranted.

With Generous Support From...