The email was the opposite of scary. Subject: “Your Genetic Profile is Ready at 23andMe!” Six weeks earlier, I had mailed the genetic testing company a tube of my own spit. Now it was time for me to “start exploring” the results. I was terrified.
I read a lot about genetics, so I knew that this sort of test isn’t usually very useful, health-wise. 23andMe estimates your risk of getting common diseases — heart disease, diabetes, asthma, and many others — based on large studies of people who carry the same DNA-letter blips as you do. Think of a chromosome as a long string of DNA letters. According to one study, people who have a ‘T’ on one spot of chromosome 8 have a 4-fold higher risk of stomach cancer than do people who carry a ‘C’ on the very same spot. That may sound awful for T people, but in real numbers, it means C people have a really tiny risk, and T people have a slightly-less-tiny, but still tiny risk. In other words, the vast majority of people who carry the “high-risk” T will never get stomach cancer.
I understood all of this and was quite familiar with the oodles of scientific commentaries and popular press articles and books about why these risk tests probably aren’t worth fussing over. Their interpretation is so fuzzy that some doctors think they shouldn’t be sold to consumers without expert supervision. Like a rebellious teenager, I loathe this kind of paternalism, and bought the test partly in defiance. If I waited too long, regulators might deny me the opportunity. I’ll show them!
So after all of that huffing and puffing and spitting and waiting, I finally had access to my precious genetic profile. And I was scared of the damn website.
Sometimes, genetic tests spur life-changing decisions. A good friend of mine had just found out — by taking a different test, which analyzes a patented part of the genome — that she carries a glitch that puts her at a very high risk of breast and ovarian cancer. She’s seriously considering a preventative double mastectomy when she turns 30 (that’s when the risk really kicks in), and a hysterectomy after she has children.
The email and its exclamation points sat in my inbox for 18 days before I mustered the courage to click on the link to my account. And when I did, it wasn’t as bad as I thought it would be — it was worse.
After checking a few boxes under paragraphs of legalese, I reached the screen outlining my disease risk. The top reads Elevated Risk, followed by a long table. Each row lists a different disease, my risk, the average risk, and yellow stars signifying the company’s confidence in the estimate. The diseases with the highest risk and confidence are listed first. The one I was most worried about is right there in the third line: melanoma. My risk: 2.4%. Avg. risk: 1.7%. Four stars. A lot of thoughts raced around my head, which I remember as: This makes total sense; all of dad’s family has skin cancer. One of my freckles exploded in blood last week. I’m going to die of skin cancer. I don’t want to read anymore. I closed the window.
I went back to the site many times over the next few days, gradually reading the rest of my health profile — including the more palatable Decreased Risk and Typical Risk tables further down on the page — and the company’s well-written technical reports about my melanoma-associated genes. As it turns out, in addition to three risky DNA variants, I carry another blip that has been tentatively (three stars) linked to a decreased risk of melanoma. I also learned that melanoma has a heritability of about 20 percent, which means that 80 percent of your risk is controlled by environmental factors — including, of course, sun exposure, which I can do something about.
I’m still working through what these numbers mean for my life, and I can’t say that I’m entirely comfortable with my 23andMe page yet. But I’m happy that I came to the service — was allowed to buy the service — on my own. Like it or not, the genetic revolution is here, and it’s really confusing. But I gained a much more sophisticated understanding of its complexity, and its power, by thinking about it in the context of my own health.
**
The picture, from Wellcome Images, shows a blood vessel in a melanoma cell.
Hi, Virginia! FWIW, my 23andMe results told me I was at decreased risk of melanoma, but that was based on just two genes, and there are surely many more genes than that involved in melanoma risk. Given that my mother had melanoma two years ago — and given that I’m a fair-skinned European who spent her childhood burning herself to a crisp — I’m certainly not going to use my 23andMe results as an excuse to stop using sunscreen, anymore than you should look at yours as a reason to panic. Context is everything. Sounds like you already know that!
I had the exact same reaction. And I got more freaked out when ‘cousins’ started contacting me. I thought I had hidden all my information–and to see people finding me and wanting family information made me viscerally angry.
I have calmed down to some extent. But I definitely don’t share the unbridled enthusiasm that I see from others.
Great point, Mary — genetics is not destiny. And EVERYBODY, WEAR SUNSCREEN! 🙂
Since this post went up, a lot of people have been emailing me asking whether I would recommend the test. The answer is, it depends. You have to be willing to dig into this stuff, and you have to be emotionally resilient. Even though I realized going into the test that genetics is not destiny, I still had a really strong emotional reaction to my results. But now that that’s passed, I’m totally fascinated by them.
Thanks for the post, Ginny! I’m curious whether that absolute estimated melanoma risk–2.4%–struck you as high? Perhaps it’s just my particular form of the Irish combination of pale skin and an expectation of tragedy talking, but even that elevated risk seems acceptably modest. Your results spurred me to look at overall risks for the first time, and I’m finding it all to be a bit of a relief, as in 2.4% is a heck of a lot better than 50%+. Or of the 100% probability of dying of some damn thing, for that matter. Still and all, I’m glad you looked. And that you took it all with an appropriately probabilistic grain of salt.
So where do you get the 50% risk estimates? Link?
On its own, 2.4% doesn’t seem high. But then when it told me that’s 40% higher than average…well, I panicked.
But that’s sort of the point of this whole post. It’s not a very big risk, and I knew, rationally, that it wasn’t very big. And I still freaked out.
50% wasn’t a statistic! It was as estimate of the uninformed irrational melanoma risk fear *I* usually walk around with. (20% heritability X 10^6% sunburns ~= 50% risk, no?) Having gotten the point of your post, I was attempting to agree with you and relate to the whole rational/irrational thing. And I guess I failed miserably 🙂
haha!! oh Tom, I’m sorry that I can’t read! To be fair, I was on Amtrak after a long lab visit, so my mind was unable to comprehend humor. 🙂
I agree that the tests should be freely available. It’s my DNA, and there’s little need to *require* a doctor to get in the middle. Maybe it makes sense for people to see doctors to analyze these results and create a comprehensive health plan. But are the people who will pay for their own data the ones who we need to encourage to see doctors? Or are the uninterested also the unhealthy?
Thanks for sharing the personal aspects of looking at real numbers: however large the margin of error is, there is more fact on the page than in trying to remember your great-uncle’s liver disease.