Naked Science Forum
General Science => General Science => Topic started by: cheryl j on 26/11/2013 05:59:26
-
According to the CBC, "The U.S. Food and Drug Administration is ordering genetic test maker 23andMe to halt sales of its personalized DNA test kits, saying the company has failed to show that the technology is backed by science.In a warning letter posted online, FDA regulators say the Silicon Valley company is violating federal law because its products claim to identify health risks for more than 250 diseases and health conditions."
If they don't advertize themselves as a regulated diagnostic medical laboratory, what responsibility or liability do they have for the accuracy of their results, or what people do with the information?
On the one hand, it seems reasonable to require 23 and Me to be clear to consumers about their credentials, quality control, error rate, etc. On the other, I can't imagine a woman getting a positive result for a BRCA gene from this lab would actually have a breast removed with out a repeat testing and confirmation from a regulated diagnostic laboratory.
-
People have done more stupid things with less evidence.
Next thing you know, the FDA will ban religion on the grounds of lack of evidence and significant proven harm.
-
In diagnostics labs, when genetic information is used to make irreversible, life altering decisions, they repeat the test 3 different times by different technologists. Granted, there's only one specimen, which could get mislabeled, but the label on the original specimen is verified by the patient, the doctor, and an assistant. So there are different protocols in medical labs.
Nevertheless, the science behind the genetic testing in labs like 23 and Me is probably valid. I don't know about 23 and Me in particular, but I have read that similar labs use genealogy related testing as a way to generate income for other research, eg haplogroups, migration or contact between different groups of people over time, etc. They get permission to use your DNA and you get some information about your origins, which seems to me like a win-win situation. So the science behind it must be valid enough to use for research publication. That doesn't mean a research lab should be liable for damages caused by giving the wrong person the wrong results, the way a medical lab would be. At most, one should only be allowed to ask for his money back.
-
Nevertheless, the science behind the genetic testing in labs like 23 and Me is probably valid. I don't know about 23 and Me in particular
They seem to provide the sample collection device and software to analyse the results. It appears the testing technology is Illumina HumanOmniExpress-24 format chip (FDA approved?), I didn't see who actually does the testing.
Certainly they spell it out in the legal section:https://www.23andme.com/about/tos/ (https://www.23andme.com/about/tos/)
Typically medical tests are ordered by trained physicians that provide an interface to the patient. Would an average consumer that uses Amazon to purchase a test understand "for research, informational, and educational use only"? They encourage talking with the doctor about results, it seems a bit like a screening test, you may or may not have ____, go ask your doctor. So if it gets more people involved in their health that would be a good outcome.
FDA approval is substantial for a diagnostic device, it can be expensive but with $50MM in funding it makes me wonder. Are they trying to get approval or convince the FDA this outside their purview? If it is the latter, that could have some very large collateral effects.
Disclaimer: I have no direct knowledge of the people, products or companies mentioned herein, no financial interests, make no claims as to validity of any statements made, this posting is offered for entertainment purposes only.
-
The FDA regulates foods & medicines (which people consume), medical tools and appliances (which are used on human bodies) and diagnostic tests (eg CAT scanners, gamma ray imaging, etc which could potentially damage human bodies).
When the test is conducted on a saliva sample, there is no direct risk from taking the measurement.
Even if there is no direct risk from the measurement, you would expect that a diagnostic test should be able to prove accuracy before it is used to guide treatment or operations on human bodies - which poses an indirect risk from a diagnostic test. In extreme cases, I guess depression and suicide are potentially indirect side-effects of a DNA test!
However, this service seems to offer indicative links to recent research, rather than proven results from large field trials. I think that there is a useful role for services like this, provided that they clearly state the limitations of their methodology - which they appear to describe fairly well.
The Terms & Conditions linked above describe the services as for "research, informational, and educational use only... We do not provide medical advice". Surely this is no different than Google providing lots of hyperlinks into research papers on genetic analysis?
Given another 10 years, it should be possible to get a full genome scan done by "through the mail" services like this. But interpreting the results in a clear and understandable manner for individual subscribers will still probably be the weak link.