Last updated: November 14, 2012
Addressing the challenges of using genetic variants in medical care
By Geoff Spencer
Associate Communications Director for Extramural Research
Researchers link a set of genetic variations to high blood pressure. Should the doctor preemptively treat his patient carrying those variations for high blood pressure before the symptoms arise?
A family history shows generation after generation suffering from diabetes, but a genome scan only shows a few genetic variations that moderately predispose for type two diabetes in the patient's DNA. Will that information change the patient's behavior to avoid this sometimes catastrophic illness or should the physician start treatment right away?
Increasingly, as genomic technologies enter medical care, health care providers will confront such questions for which the experts lack good answers. Faced with the hard problem of identifying clear genomic signals on which doctors can rely to make medical decisions, the National Human Genome Research Institute (NHGRI) organized a workshop titled "Characterizing and Displaying Genetic Variants for Clinical Action," on Dec. 1 and 2, 2011.
This key question drew approximately 80 participants and speakers to aday-and-a-half-long workshop from a wide range of disciplines and organizations, including basic and clinical researchers, clinicians, program directors from numerous institutes and centers at the National Institutes of Health, representatives of government agencies such as the Centers for Medicare and Medicaid Services, and the Food and Drug Administration. In addition, there were representatives from private entities such as Blue Cross Blue Shield and Associated Regional and University Pathologists (ARUP) laboratories. The workshop was sponsored by NHGRI and the Wellcome Trust, a global charity based in London that supports biomedical research.
"The time is right to begin to chart a course to using genomic information as part of medical care," said Eric D. Green, M.D., Ph.D., director of NHGRI, who attended the workshop. "This is new territory for us, but we are prepared now to start planting the seeds that will yield results in the coming decades that the community needs."
Researchers already have identified a growing number of genetic variants that may be used in the clinic. For example, doctors treat patients with certain types of cancer by first analyzing genetic variants in an individual's tumor to further categorize the subtype of cancer they're dealing with. In such cases, the information dictates a more appropriate and targeted treatment.
Genetic variants can also shed light on how certain medications are metabolized, a research area known as pharmacogenomics. Each person carries genetic variants that may make a drug more or less effective. The Food and Drug Administration maintains a growing list of more than 100 approved drugs with pharmocogenomic information on the labels recommending tests for genetic variants to help target a proper dose.
Despite the recent advances, researchers and healthcare workers still face many barriers to the widespread use of genomic information in the routine care of patients. For instance, it takes a great deal of time and expertise to determine the clinical validity and utility of a genetic variant. Moreover, there is not a standardized way of making that determination. Two different researchers or clinicians presented with the same genetic variant may come to two different conclusions about whether the information is useful or how to use it in treating a patient. They have to consider numerous clinical factors, as well as the psychological risks and benefits of reporting genetic variants to patients. It's not always an easy call.
NHGRI organized the workshop to develop strategies on how best to solve these thorny problems and how best to move genomic analysis into clinical care. The workshop was also part of the institute's effort to implement NHGRI's strategic plan for the field of genomics. In addition to ongoing basic research, the plan lays out the steps the field must take to move discoveries about the human genome to applications in medical care. One of the imperatives for genomic medicine is "practical systems for clinical genomics informatics." Such systems need to be developed to capture and display the most clinically relevant genetic variants for healthcare providers to use.
"We need to separate the wheat from the chaff, to identify the genetic variants that are really important that should be reported to patients and develop systems for clinicians to use them," said Teri Manolio, M.D., director of the Office of Population Genomics at NHGRI, which organized the workshop.
Workshop attendees made a number of requests and recommendations about the processes and resources needed to translate genetic variants into clinical action. Some of those are:
- Acquire reliable and accessible data to validate the association of genetic variants to diseases that are clinically relevant.
- Develop new databases and resources to combine existing data on genetic variants that may be clinically actionable.
- Engage medical experts to evaluate genetic variant data and come to a consensus as to which variants are actionable.
- Create user-friendly clinical support tools to provide evidence on genetic variants to help clinicians interpret the information for patient care.
- Construct automated feedback loops to evaluate outcomes and curate information on genetic variants between clinics and genetic variation resources.
"I think we got a clear message that some kind of resource is needed and everybody has different needs," Dr. Manolio said. "We can play a role in bringing together disease disciplines with informatics people and epidemiologists who do the research in order to provide the data that says yes, we should report this variant or no we should not report that variant."
NHGRI hopes to fund an effort in the future to implement some of the workshop recommendations, though it is not clear yet what such an effort will look like.
NHGRI also plans to coordinate any activities it funds with organizations such as the Agency for Healthcare Research and Quality (AHRQ), the Office of the National Coordinator for Health Information Technology (ONC), commercial electronic health record (EHR) vendors and others to address data interoperability and approaches to integrate genomic information and actionable variants into a variety of systems.