One in five apparently healthy primary care patients carry a genetic variant expected to cause a rare disease, according to results of a pilot study published online June 27 in the Annals of Internal Medicine. However, for most patients, this information yields little change in their care, and may boost costs, researchers suggest.
As the cost for sequencing all 3 billion base pairs in the human genome has fallen to about $1000, many have speculated that genomic information could become a powerful tool in primary care. Already, genomic sequencing has become a useful tool in oncology, prenatal screening, and helping aid diagnosis of patients with rare conditions.
However, concerns have been raised about whether primary care physicians would be able to interpret genomic information and appropriately apply it to patient care, according to Jason Vassy, MD, MPH, a primary care physician at the Veterans Affairs Boston Healthcare System in Massachusetts, who conducted the study at Brigham and Women's Hospital. In addition, concerns have been raised about whether genomic testing might drive up health costs or lead to patient harm.
To begin to assess the utility of genomics in primary care, Dr Vassy and colleagues randomly assigned 100 generally healthy primary care patients in their 50s and 60s to receive a detailed report on their family history and its implications for their care or to receive a detailed family history plus a report on the results of whole-genome sequencing of their DNA. Patients' clinical outcomes and healthcare usage were assessed on the basis of 6 months of medical records and recordings of their discussions about the reports with their physician.
The study revealed that 11 (22%) of the 50 patients who underwent whole-genome sequencing carried genetic variants linked to rare disease, but only 2 had any physical manifestation of the disease. Therefore, the clinical utility of whole-genome sequencing in the study population was very limited, the authors say.
The pilot study did, however, uncover a few silver linings. Primary care physicians in the study were able to manage the often-uncertain results appropriately in most cases, and patients did not experience increased anxiety.
Holly Tabor, MD, associate professor and associate director at the Stanford Center for Bioethics, told Medscape Medical News the study was well designed but had important limitations and may not be widely generalizable. She noted, for example, that the study was small and enrolled healthy, predominantly white individuals who were treated at an academic center.
"It is an interesting and important set of findings," Dr Tabor said.
Also, she noted that the control group, which used an electronic family history tool, might not be representative of patients in a typical practice. The family history tool itself may have provided important genetic information.
Dr Vassy told Medscape Medical News it was surprising that so many healthy middle-aged patients in the cohort had genetic variants linked to rare diseases, and that so few had any symptoms. This suggests more research is needed to understand the role of genetic variants in disease
"The quality of the scientific evidence laboratories use to interpret genetic variants' clinical significance is variable," he said. "That doesn't mean the gene association isn't real, but we do need to understand why that particular patient doesn't have the disease."
But Dr Tabor said that finding uncertain results in healthy individuals is not surprising. This might make interpreting such data harder for primary care physicians.
"We do know a lot about some things, and some of that info might be very [clinically] useful, but coming along with that is a lot of information we are uncertain about," she said.
Physicians and Experts Will Be Needed
By and large, the primary care physicians were able to accurately interpret the results and recommended appropriate care, the study found. A panel of 11 clinician-geneticists analyzed recorded patient interactions and medical records to assess this outcome. For 8 of the 11 patients with a disease-associated variant, the geneticists rated the primary care physicians' actions appropriate. Physicians recommended new interventions for 34% of those who underwent genetic testing compared with 16% of those who received only a family history.
"This provides some reassurance that physicians will be able to handle genomic information," Dr Vassy told Medscape Medical News. He explained that physicians are already accustomed to managing uncertain information in practice and are likely applying this skill to managing uncertain genomic information.
Patients who received genomic results did not report excess anxiety about the findings. They were, however, more likely to report changing their health behaviors than those who received only family history–related results (41% vs 30%).
However, Dr Tabor does not think the study had enough statistical power to detect potential differences in patients experiencing psychological distress. She notes that many patients were referred to specialists and might experience distress after referral if they receive information that causes worry or sets off a medical odyssey.
Also, in her view, the high rate of referral also suggests that primary care physicians will need expert help, which could burden limited expert resources as more healthy people are referred to geneticists or genetic counsellors.
"The message isn't that primary care physicians could do it on their own," she said.
Additional testing to assess uncertain results may have driven up costs for patients in the sequencing group, who paid on average $350 more than the family history group for their care during the follow-up period. However, larger studies would be needed to find a statistical difference between the two groups on cost.
"We didn't have enough statistical power to say if it was different," Dr Vassy said. But 6-month health costs were twice as high in a small subgroup of patients in the sequencing group who had disease-linked variants compared with those who did not carry a concerning variant ($2526 vs $1198).
A larger study with longer follow-up is needed to better understand the utility of genomics in primary care, Dr Vassy told Medscape Medical News.
Until such results are available, he said, it is too soon to begin implementing whole-genome sequencing in primary care.
"In 2017, we can't say our results argue in favor of routine use of whole-genome sequencing in the primary care of otherwise healthy people," he said.
Study authors report consulting for Vermont Oxford Network, Gerson Lehrman Group, ClearView Healthcare Partners, and Human Longevity. They also report receiving grants and/or fees from EarlySense, S.E.A. Medical Systems, QPID Health, Center for Digital Innovation (Negev), Enelgy, Valera Health, Intensix, MDClone, InVitae, Merck, MedSeq, and Regeneron; and equity from S.E.A. Medical Systems, QPID Health, Enelgy, Valera Health, Intensix, and MDClone. One coauthor has a patent for Medicalis with royalties paid to Brigham and Women's Hospital and another coauthor has a patent for gene testing in cardiomyopathies with royalties paid to Partners HealthCare. Dr Tabor reported no relevant conflicts. Full study disclosures available on the journal's website.
Ann Intern Med. Published online June 27, 2017. Abstract