I’m sick of reading about the dangers of the genome. There are lots of popular articles I could point to, but let’s start with a recent series in Time that included eight online features and the Dec. 13 cover story, ominously titled “The DNA Dilemma.”
The series, written by Bonnie Rochman, is thoroughly reported, balanced, and full of fascinating personal stories about children whose genomes have been sequenced. It’s also timely: The primary question Rochman raises—how much information is too much information?—has been dominating commentaries about genetic testing in the medical literature.
But this is the wrong question, or at least one that’s becoming increasingly irrelevant. The personal genomics horse has bolted, and yet many paternalistic members of the medical community are still trying to shut the barn door. In doing so, they’re fostering a culture of DNA fear when what we really need is a realistic and nuanced genetics education.
There are many kinds of genetic tests, but most of the hoopla revolves around whole-genome sequences—the impossibly long, letter-by-letter readouts of the DNA inside the nucleus of each of your cells. In 2003, the first human genome was fully sequenced for just shy of $3 billion. Today a doctor can order yours for around $10,000.
Though dropping every day, the cost is still prohibitive enough that most people who get their genome sequenced are part of a medical research study. But the technology is beginning to seep into everyday clinical settings, especially for children with rare diseases. In either situation, the doctor or researcher might inadvertently discover genomic information—known as “incidental findings” in the scientific literature and “dark DNA secrets” in one of the Time articles—that has nothing to do with the child’s sickness or the study at hand. Hence the big dilemma: How much do patients want to know? How much do they need to know?
This so-called "return of results" is one of the hottest issues in genetics and will only get more pressing as genetic testing becomes cheaper and more routine. Last year, the National Human Genome Research Institute set aside $40 million for studies about how genome sequences can be applied in a medical setting, including $5.5 million for projects on returning research results to participants. No fewer than 28 scientific reports on the subject appeared in 2012 alone, as did an editorial in the journal Nature.
The public discussion about DNA testing tends to focus on ethical dilemmas: What if doctors find that a person’s father isn’t really? Should they tell a patient about a DNA glitch if it's only occasionally linked to disease? What if, while looking for mutations that could explain a known sickness, they stumble on others that might predict late-life dementia or indicate the presence of HIV? Would adding this data to someone's medical record affect health insurance rates? What if—gasp—we end up with a real-life Gattaca?
These questions are worth talking about. But the genetics community and popular press spend too much time debating when and how the medical establishment should "protect" people from their children’s or their own DNA.
For example, many bioethicists argue that DNA glitches shouldn’t be disclosed if they’re ambiguous or linked to untreatable conditions. Doing so "may create unwanted psychosocial burdens on parents," according to a commentary on newborn sequencing in the Journal of the American Medical Association.
Here’s a real-life example of how that philosophy can play out. As described in one of the Time stories, researchers at the Children’s Hospital of Philadelphia screened the genome of a sick baby and discovered a genetic variant that is linked to dementia around age 40. Because the researchers apparently hadn’t asked the parents ahead of time what extra genetic information (unrelated to the illness) they would want to know, the researchers were in a tough spot. They ultimately decided not to tell the parents about their child’s dementia risk. “We came around to the realization that we could not divulge that information,” one of the geneticists told Time. “One of the basic principles of medicine is to do no harm.”
Though I’m sure the researchers meant well, their implicit message—that a nonexpert couldn’t possibly understand the concept of an uncertain risk and would therefore be harmed by the knowledge—is not only patronizing but probably false.
I learned a lot about uncertain risk thanks to 23andMe, a private company that for $99 screened a tube of my spit for about 1 million SNPs, or common genetic variants that are associated with small increases or decreases in the risk of various diseases. I received my results (all of them) on an easy-to-use website. I found out that I have an elevated risk of heart disease and melanoma, among other ailments. My immediate reaction was panic, I’ll admit, but after chewing on the information I realized it wasn’t all that scary.
My response is pretty typical. Last year, the New England Journal of Medicine published a study of how more than 2,000 people reacted to getting SNP results from another direct-to-consumer genetic testing company. The bottom line: Nobody freaked out. Six months after testing, people who received this information didn’t show any differences in anxiety compared with those who didn’t receive it.
Part of the reason for the ho-hum reactions, mine included, is that SNPs are notoriously ambiguous: Sometimes they point to disease and other times they don’t. Whole-genome sequences, in contrast, sometimes finger rare mutations that are far more predictive—and more frightening.
Genetic screening, in other words, is getting more useful. And as it gets more accurate, yes, it is going to sometimes predict devastating diseases, and yes, cause some anxiety. And?
While bioethicists agonize over how to prevent patient distress, genetic technology is popping up outside of the protective confines of a scientific study or medical clinic. For all you garage do-it-yourselfers, $600 will get you a kit to build a rudimentary gene sequencer using nothing more than two screwdrivers and a pair of pliers. Last September, 23andMe sequenced the exome—the millions of DNA bases in the genome that code for proteins—of a pilot group of customers for $999 a pop. The company plans to eventually offer whole-genome sequences, consisting of 3 billion pairs of DNA bases.
As no-doctor-necessary consumer sequencing becomes more popular, the private genome interpretation industry will also boom. “It’s really not too long before this is no more complicated than setting up an Orange Julius franchise at the mall,” says geneticist Misha Angrist of Duke University.
Competent DNA interpretation is complicated, of course, and I’d rather consult a doctor than go to the skeevy guy at the mall kiosk. But that’s not the choice I have, not yet anyway. While wasting time debating ethical dilemmas, the medical community has neglected to talk about more pressing logistical problems: 1) How to ask people ahead of time what, precisely, they want to know (and don't want to know); and 2) How to improve the medical system so doctors can follow through on those wishes.
The first problem boils down to the concept of “informed consent,” which usually means page after page of consent forms outlining the rights of a patient or research volunteer. (Whenever I talk to researchers about informed consent, they invariably compare it to the iTunes user agreement, where everybody checks the little box without reading the text.) Genome sequencing is so new that informed consent doesn’t always happen, resulting in doctors ordering tests without asking people ahead of time about what to do with the results.
Asking people what they want to know is tricky because you don’t know what will be relevant before you look at the data, notes Amy McGuire, director of the Center for Medical Ethics and Health Policy at Baylor College of Medicine. “It’s impossible at the front end to go through every possible piece of information they’d get back and ask them how they’d feel about that.”
Still, people can be informed about the general categories of concern—and new technology can guide them through the process without much cost. A website called my46, for example, allows research volunteers to set preferences about what types of information they want to know and offers pros and cons for each decision. I can easily imagine a future in which the process of finding out about your genome doesn't require extra trips to your doctor or daunting paperwork. It could be more like filling out an online dating profile.
That’s not to say that doctors shouldn’t be part of the process. A (good) doctor knows you, your habits, the drugs you’re taking, and your family medical history. And doctors are legally bound to keep your information private. I would love to discuss some of my 23andMe results about skin cancer with my dermatologist, for example.
The trouble is, the average doc isn’t likely to be of much help. A survey published in May found that 74 percent of internists from academic medical centers rate their knowledge of genetics as “very/somewhat poor.” A 2011 report of Canadian oncologists, cardiologists, and family doctors showed that 92 percent had no formal undergraduate training in genetic testing, and 89 percent had no graduate training. Only 29 percent said they are able to interpret the results of genetic tests, and just 30 percent said they are comfortable discussing test results with patients.
These studies may even underestimate the problem. “The only thing more crazy than what doctors don’t know about genetics is their lack of understanding about what they don’t know about genetics,” says Laura Hercher, a genetic counselor at Sarah Lawrence College in New York.
Medical schools are beginning to add genetics to their curricula, and some academic medical centers are offering one-off genetics courses for doctors, but they’re far from sufficient. We need to come up with better ways to keep doctors informed about the ever-changing insights of genetics. One way is by asking the advice of genetic counselors. There aren’t nearly enough of them to go around, but the supply might change if doctors were asking for them. A cheaper option is gadgetry. Just as many doctors depend on a phone app called Epocrates to look up details on drug dosing and side effects, they could consult an e-genome cheat sheet.
Even if every doctor in the country were fully educated on genetics, we’d still have the massive problem of cost and access. Although the demand will certainly be there, both for medical utility and plain old curiosity, nobody’s talking about setting up efficient and economical ways to help people understand their genomes on a wide scale. “A lot of this discussion takes place in a total fantasy realm where costs are never considered,” Hercher says. “We’ve got to try to figure out a fair, humane system that will work in the real clinical world.”
There’s no way around the fact that genomes are complicated and, at least for now, difficult to interpret. And sometimes, yes, they will lead to unpleasant information. But they’re not going to suddenly fade away. More and more people are getting their DNA sequenced and eager to find out what secrets, if any, it holds. So let’s figure out how to tell them.