Field of Science

Pages

Spitting in a Tube, and Other Risky Behaviors



In my office, someone occasionally leaves a copy of the Tribune on the kitchen table for people to peruse throughout the day. One day last month, I glanced at that morning's paper on my way to make a cup of tea. A second later, halfway across the kitchen, I stopped in my tracks and said "Oh no." Seriously, it was like a cartoon. I'm glad no one was watching.

Here's the headline that caused my double-take: "Walgreens to Sell Gene Testing Kits."

For $20 or $30, the paper said, you'd soon be able to go to Walgreens and buy a little tube and an envelope. After spitting in the tube, you could pay another $79 to $249 to find out about your disease risks and other vague-but-valuable-sounding information.

You might think this sounds pretty OK, or you might think it sounds like a disaster. More on that in a moment. Someone at the FDA must have also said "Oh no," because the next day's headline was: "Walgreens Halts Sale of Genetic Test." The FDA has since cracked down on five companies involved in at-home genetic testing, including Pathway, the maker of the over-the-counter test Walgreens was planning on selling. (The other companies offer their products online; some of them are much more expensive.) All of the companies got letters from the FDA last week saying that they must either submit their products for approval, or explain why they aren't medical devices and therefore don't need to be approved.

What kinds of information are these companies actually offering? They don't send you a giant "ATTGCCCCAGTTCA..." printout like in Gattaca. (By the way, those scrolls of paper were totally unrealistic. In 12-point Times New Roman, I can fit about 2400 capital letters on a page. At that rate, a printout of 3 billion base pairs would take up 1.25 million pages. I'll leave it to someone else to figure out how large that scroll of paper would be.) Instead, they test you for certain genetic markers. For example, if you request the "Health" report from the company 23andme, they'll look at about 150 of your genes. Within each gene, they'll check to see which version of a common variation you have: is this position a C or a T? An A or a G? The variants they're testing for have been associated with an increased or decreased risk of some disease or trait. The report you get back will say things like: Your variant of this gene gives you a 50% increased risk of developing this condition, compared to the population as a whole.

So: Pretty OK? Or problematic?

You could certainly get some useful information from having your genes tested. If you find out you're at increased risk for breast cancer, you can make sure to get regular screenings. If you carry a gene for cystic fibrosis, you might decide to have your partner's genes tested, too, before having any children. There's also a lot of non-useful information you'll get: for example, 23andme will tell you about your earwax type (sticky or dry) and your eye color (probably something you've picked up on without help).

You might also receive information you don't want. You could find out that your father is not really your father; or that you're at increased risk for a brain aneurysm, and there's nothing you can do about it.

How accurate are these tests? It's hard to say, without anyone regulating the companies that perform them. You might take a genetic test at three different companies and get three somewhat different results. Recently, 23andme got mixed up and sent 96 customers the wrong data. Even if the company is absolutely certain that they've analyzed your genes accurately, the best they can offer is usually a statistical correlation. They might know that a certain gene variant is correlated with a 20% increase in the risk of a disease, but have no idea what that gene has to do with the disease itself--if anything. Your destiny may be written in your genes, but researchers are still struggling with the decoder ring.

Assuming you get accurate information, how valuable is it? Let's say you're a man, and your testing company tells you that your gene variant increases your odds of bladder cancer by 20%. Do you start Googling bladder cancer symptoms? Do you worry that your frequent urination means you have an advanced tumor, or that you've unfairly passed on a dangerous gene to your kids? Well, the average bladder cancer risk for men is 4%, so yours is now 4.8%. In the population at large, that gene variant is meaningful. But for you personally, maybe it's not.

In other cases, your genes themselves might be insignificant compared to environmental factors, like how much you smoke or eat or drink. For example, these tests will tell you your genetic risk of Type II diabetes, a disease that's 75% determined by environmental factors. If an inactive, obese person finds out she has a genetically low risk for Type II diabetes, and therefore assumes she's safe from it, that test has done her a disservice.

The testing companies argue that information about people's bodies should be available to everyone who wants it, and it's paternalistic to say otherwise. But they're not only offering information; they're also interpreting it, and it might be hard for customers to see the line between one and the other. What seems like information can really be a false impression of security, or of risk, or of control. (How crushing must it be for the "previvor"--the woman who finds out she has a risky gene variant and opts for a preventive double mastectomy or hysterectomy--who later develops heart disease or colon cancer?)

Of course we all want to know more about ourselves. Instead of arguing with human nature, I'll just leave you with one more caveat: At some point in the future, you may be forced to tell insurers about your genetic testing. And while those tiny statistical differences may not be meaningful to you, your health insurance provider is going to love them.

No comments:

Post a Comment

Markup Key:
- <b>bold</b> = bold
- <i>italic</i> = italic
- <a href="http://www.fieldofscience.com/">FoS</a> = FoS

Note: Only a member of this blog may post a comment.