Patients Should Get DNA Information, Report Recommends

Fourteen genetics experts, with the backing of the American College of Medical Genetics and Genomics (ACMG), are proposing a radical shift in how and what patients learn about what's in their DNA. They argue that anyone whose genome is sequenced for any medical reason should automatically learn whether 57 of their genes put them at risk of certain cancers, potentially fatal heart conditions, and other serious health problems. The information would be provided whether patients want it—and often when they're seeking care from a doctor for something else entirely—because, the experts say, knowing the makeup of this DNA could save an individual's life. The recommendations apply to sequencing children's DNA as well, even if there's no preventive care available until adulthood. The college's guidelines on a range of issues are usually written by influential geneticists and physicians and carry significant weight, although they are not binding. Today's report includes the first recommendations ever given to labs and doctors about how to handle unexpected findings when the genome or its protein-coding "exome" is sequenced.

The ACMG recommendations, released this morning, come as DNA sequencing is about to take off in doctor's offices and hospitals. Sequencing an entire genome soon won't cost much more than sequencing a single gene, says Robert Green, a neurologist and medical geneticist at Brigham and Women's Hospital in Boston who co-chaired the ACMG working group. That means that if a child has an undiagnosed heart disorder that may merit sequencing of a gene, the parents could, for practical and cost reasons, agree to have their son or daughter's full genome sequenced. "You can't undo the sequence" once you've got it, Green says. "To somehow mask or ignore it doesn't seem right either."

Under the new recommendations, the parents would learn not only if their child's heart condition is due to a particular gene mutation—they'd also learn whether she is at high risk of certain breast cancers, colon cancers, cardiomyopathies, aneurysms, and other diseases. (Testing negative for these, however, would not guarantee that she's in the clear, because not every mutation for a given gene is on the list.) Many of the genetically driven conditions can strike children, but others rarely show up before adulthood.

Arriving at today's recommendations was an arduous task. Over the course of a year, a committee led by Green and Leslie Biesecker, chief of the Genetic Disease Research Branch at the National Human Genome Research Institute in Bethesda, Maryland, has been weighing how to handle "incidental findings" that turn up when a genome or exome is sequenced for some other medical reason. The effort is part of a much broader debate that spans research and clinical care over which genetic results should be returned—an ethical, legal, and practical minefield.

And the ACMG committee waded right in. Some of those not involved in it were most startled by the recommendation that patients should have no choice but to learn whether they carry dozens of mutations. "[T]he Working Group did not favor offering the patient a preference as to whether or not to receive the minimum list of incidental findings described in these recommendations," the group wrote. The "fiduciary duty to prevent harm by warning patients and their families," the authors argued, "supersedes concerns about autonomy"—the right not to know certain information. Labs, they suggested, should "seek and report" mutations in dozens of genes, many of them linked to cancer syndromes. The doctor who ordered the test would be responsible for communicating the results to patients—something the working group admitted some physicians might be uncomfortable with, especially if, say, the doctor is a cardiologist and the sequencing turns up a cancer gene.

Several experts who did not participate in the report voiced practical and ethical concerns. "The assumption that it's always benign to give that information, and potentially helpful—I don't think that's true," said Mark Rothstein, a law professor who studies bioethics and genetics at the University of Louisville Brandeis School of Law and School of Medicine in Kentucky. "Information is often very damaging to people, that's why they don't want it." Rothstein notes that patients might begin undergoing "test after test and worrying themselves sick, and they'll never be affected" by the condition in question.

The ACMG working group limited its list to genes that carry a high risk of disease—for example, mutations in BRCA1 and BRCA2, which significantly raise the risk of breast and ovarian cancer, and for which extra screening and prophylactic surgery can reduce risk. But even for these genes, the chance of disease varies from individual to individual. Ashkenazi Jewish women with a family history of breast cancer may have a disease risk as high as 90% or more if they carry mutations in BRCA1; another woman with no family history may have a much lower risk. In some cases, the conditions highlighted by ACMG are rare and have been studied in just a few families in which many members have the disease. It's not known how those risks translate to someone without any affected family members. The working group estimates that about 1% of the population would have a positive result for at least one of the variants on the list. The list was limited to mutations for diseases where the chance of getting sick can be mitigated—for example, with aggressive screening for those at high risk.

"In many perspectives, they have made an important and brave contribution," writes Isaac Kohane, a pediatric endocrinologist and chair of the informatics program at Boston Children's Hospital, in an e-mail. Still, while he thinks we have the "sequencing capacity" to deal with the demands, "[I'm] not sure we have the cognitive capacity." And our understanding is constantly changing, even for the diseases and mutations already on the list. The Working Group treads carefully here, urging that the list be regularly revisited and that a voluntary registry be established to follow those given these findings, documenting "the benefits, harms and costs that may result."

Many questions remain: Are doctors equipped to return this information to patients? Can labs handle the additional sequencing? Who will pay for it? Do these recommendations establish a new legal benchmark for the standard of care, something Rothstein thinks is a possibility?

The working group emphasizes that its 30-page document is a beginning, not an end. "We took so long and worked so hard on this," Green says. "We took time to imagine how the future is going to be different."