Twenty-three skidoo
One of the more interesting and well-reported stories this week came from Kira Peikoff, who had multiple DNA tests for the New York Times and compared the results. The upshot: they were all different. Peikoff's experience indicates that the US Food and Drug Administration had a point in late November when it told the personal genetics company 23andMe to stop marketing its saliva test product, calling the company's product a diagnostic device (and therefore subject to regulation) and requiring the company to prove its claims. The company has pulled its health-related test; it still offers ancestry-related tests and raw genetic data.
As the FDA explained in the letter it sent 23andMe CEO Ann Wojcicki, the risk is that people will make serious decisions about their health based on the information they receive. After many interactions with the company, FDA deputy director James woods wrote, "we still do not have any assurance that the firm has analytically or clinically validated the PGS for its intended uses, which have expanded from the uses that the firm identified in its submissions."
Wired, predictably, seems to find the whole thing pointless as an intervention in the unstoppable flow of progress toward cheap, accurate genetic testing. In an op-ed that ran on New Year's Day, Larry Downes and Paul Nunes concluded, "Regulators and medical practitioners must focus their attention not on raising temporary obstacles, but on figuring out how they can make the best use of this inevitable tidal wave of information."
This statement reminds me of the things people defending homeopathy say: that the medical establishment opposes any treatment they can't profit from. Sure, doctors might have both bad and good reasons for balking when patients demand that they interpret the results of genetic tests. But the bottom line with medical treatments, devices, and tests has to be evidence that they work as advertised - which in the case of personal genomics is massively incomplete. As a golden rule, independently conducted tests should produce the same results, the same thing we expect of any real science. This is why the New York Times piece is so powerful; it clearly shows the flaws in the state of the art. There is, as Peikoff points out, "a lack of industry standards for weighing risk factors and defining terminology". In that case, even smart people will indeed make hasty and dangerous mistakes.
The other argument, that this is the future so it cannot be stopped, is specious. On Dave Farber's mailing list, John Bosley compared the Silicon Valley habit of speaking of the future as an unstoppable force to the passive bureaucratic tense beloved of Donald Rumsfeld and President Richard Nixon. As per this last link, Bosley credits Language Log, but I believe the use of the passive voice to abdicate responsibility was first highlighted in Richard Mitchell's Underground Grammarian. Mitchell called it "the Divine Passive and likened it to a worm in the brain.
In more immediately relevant terms, say you're living at the tail end of the horse-and-buggy era, and a buggy manufacturer points out that these new-fangled motorcars have a tendency to explode. Do you say, "Look who's talking; automobiles are the future; accept your fate"? Or do you say, "Automobiles are the future, but not this design at this time"? We have the FDA and analogues in every country because people are terribly vulnerable easy prey for charlatans when they're sick and scared. Fake medicines are where "snake oil" comes from.
Myriad failed first-mover companies in a variety of fields will tell you that being early to market is not always an advantage. Sometimes your product really just wasn't ready, and while you had the right vision of the future it's someone else who's going to be the one to make it really happen. Wojcicki herself sounds more grown-up about the whole thing than her company's fans.
Peikoff's story is unsurprising to anyone who read Danish author and neuroscientist Lone Frank's 2011 book My Beautiful Genome, which began with a personal genetics test and ended with a survey of the many cowboys the burgeoning field was already attracting. As Frank discovers, genes are malleable, and their individual workings are not so neatly predictable. Among the absurd misapplications, she also finds outfits offering to test you and your potential mates for "genetic compatibility" - not whether your children will be born with a horrid disease, but personal compatibility. They can probably spot the gene for gullible-enough-to-buy-swamp-land-in-Florida, too.
In a personal appearance in London to promote her book, Frank argued that companies like this - and like the ones she found that purported to use genetic testing to identify how you should raise your child - should not exist. She called the current generation of personal genomics version 1.0 and compared it to the Commodore PET: the earliest stage of something that yes, someday, will be in everyone's hands. But not yet, and not in its present form.
Yes, some form of what 23andMe is doing is part of the future. That doesn't mean we can't make choices about it.
Wendy M. Grossman Wendy M. Grossman is the 2013 winner of the Enigma Award. Her Web site has an extensive archive of her books, articles, and music, and an archive of earlier columns in this series. Stories about the border wars between cyberspace and real life are posted occasionally during the week at the net.wars Pinboard - or follow on Twitter.