Some doctors dream of diagnosing diseases—or at least predicting disease risk—with a simple DNA scan. But others have said the practice, which could soon be the foundation of preventative medicine, isn’t worth the economic or emotional cost. Now, a new pair of studies puts numbers to the debate, and one is the first ever randomized clinical trial evaluating whole genome sequencing in healthy people. Together, they suggest that sequencing the genomes of otherwise healthy adults can for about one in five people turn up risk markers for rare diseases or genetic mutations associated with cancers.
What that means for those people and any health care system considering genome screening remains uncertain, but some watching for these studies welcomed the results nonetheless. “It's terrific that we are studying implementation of this new technology rather than ringing our hands and fretting about it without evidence,” says Barbara Biesecker, a social and behavioral researcher at the National Human Genome Research Institute in Bethesda, Maryland.
The first genome screening study looked at 100 healthy adults who initially reported their family history to their own primary care physician. Then half were randomly assigned to undergo an additional full genomic workup, which cost about $5000 each and examined some 5 million subtle DNA sequence changes, known as single-nucleotide variants, across 4600 genes—such genome screening goes far beyond that currently recommended by the American College of Medical Genetics and Genomics (ACMG), which suggests informing people of results for just 59 genes known or strongly expected to cause disease.
Of the 50 participants whose genomes were sequenced, 11 had alterations in at least one letter of DNA suspected to cause—usually rare—diseases, researchers report today in The Annals of Internal Medicine. But only two exhibited clear symptoms. One was a patient with extreme sensitivity to the sun. Their DNA revealed a skin condition called variegate porphyria. “Now that patient knows they will be much less likely to get bad sunburns or rashes if they avoid the sun and certain medications,” says Jason Vassy, a primary care clinician-investigator at Veteran Affairs Boston Healthcare System and lead author of the study.
The team also found that every sequenced patient carried at least one recessive mutation linked to a disease—a single copy of a mutant gene that could cause an illness if two copies are present. That knowledge can be used to make reproductive decisions—a partner may get tested to see if they have a matching mutation—and prompt family members to test themselves for carrier status. And in what Vassy calls a “slightly more controversial result,” the team examined participants’ chances of developing eight polygenic diseases, conditions that are rarely attributed to a single genetic mutation. Here, they compiled the collective effects of multiple genes—up to 70 for type II diabetes and 60 for coronary heart disease—to predict a patient’s relative risk of developing the disease.
Just 16% of study volunteers who only reported their family history were referred to genetic counselors or got follow-up laboratory tests. In the genome sequencing group, the number was 34%.
Some researchers have expressed concern that such whole genome screening will skyrocket medical costs or cause undue psychological harm. Aside from the initial cost of sequencing (which was covered by the study), patients who underwent the genomic screen paid an average of $350 additional in healthcare costs over the next 6 months, Vassy and colleagues reported. But contrary to fears of emotional trauma, neither the sequencing group nor the control group showed any changes in anxiety or depression 6 months after the study.
Vassy stresses that their study was small and needs follow-up, but it still impressed Christa Martin, a geneticist at Geisinger Health System, in Danville, Pennsylvania, who worked on the ACMG’s recommendations for genome sequencing. “I almost feel like the authors undersold themselves,” she says. “Many of their patients are making health behavioral changes, so they are using the information in a positive way.”
“The study was extremely well designed and very appropriately run,” adds Barbara Koenig, a medical anthropologist who directs the University of California San Francisco Bioethics Program. But she still questions the assumption by many physicians, ethicists, and patient advocates that more information is always beneficial. “It is just hard to know how all this information is going to be brought together in our pretty dysfunctional healthcare system.”
Another paper published last week on the preprint server bioRxiv, which has not yet undergone peer review, yields similar results. Using whole-exome sequencing, which looks only at the protein-coding regions of the genome, Michael Snyder, director of the Stanford Center for Genomics and Personalized Medicine in Palo Alto, California, and colleagues found that 12 out of 70 healthy adults, or 17%, unknowingly had one or more DNA mutations that increased the risk for genetic diseases for which there are treatment or preventative options.
Both studies suggest that physicians should look at genes beyond the ACMG’s 59 top priorities, Snyder says. He argues that whole-genome sequencing should be “automatically” incorporated into primary care. “You may have some super-worriers, but I would argue that the information is still useful for a physician to have.” Vassy, however, says that there isn’t yet enough evidence to ask insurance companies to reimburse whole genome sequencing of healthy patients.
“We like a quick fix and the gene is an important cultural icon right now, so we probably give it more power than it really has,” Koenig says. “But these are still really early days for these technologies to be useful in the clinic.”