NEW YORK – Researchers affiliated with the VA Boston Healthcare System have launched a randomized, controlled clinical trial to better establish whether polygenic risk scores (PRS) for select diseases provide measurable clinical utility.
The Genomic Medicine at Veterans Affairs (GenoVA) study, funded by a $1.5 million grant from the National Institutes of Health, will assess whether the inclusion of PRS in screening exams for atrial fibrillation, coronary artery disease, type 2 diabetes, colorectal cancer, breast cancer, and prostate cancer shortens time to diagnosis and impacts what other tests physicians order.
These specific diseases were chosen because they are frequently seen in primary care settings and there are established guidelines for their prevention, screening, and diagnosis.
The study will additionally explore how to counter the potential for PRS to exacerbate healthcare disparities.
Experts lack broad consensus on how to define and measure the clinical utility of PRS and many studies into the topic rely on mathematical modeling of data collected retrospectively.
The GenoVA study has enrolled 966 participants between the ages of 50 and 70, without diagnoses of any of the target illnesses. All participants underwent complete genotyping and those with at least one actionable monogenic disease result, as defined by the American College of Medicine Genetics and Genomics (ACMG), receive both monogenic and PRS results from a genetic counselor. The remaining participants are randomized to receive interventions that may be based on their PRS report at their physician's discretion or usual care without prior knowledge of their PRS results.
The clinical utility of PRS is an unsettled research question.
While some studies have shown that PRS only marginally improve risk prediction, if at all, others have shown enough evidence of potential utility that they are already being offered as part of larger tests, such as is the case with Myriad Genetics' RiskScore, and in clinics such as the Preventive Genomics Clinic at Massachusetts General Hospital, which offers them alongside monogenic testing for heart disease.
For the purposes of the GenoVA study, clinical utility is defined as the ability of a PRS to result in the earlier diagnosis of clinically significant cases of the study's target diseases, as compared to participants receiving standard care without PRS testing.
As of May 10, clinicians associated with the study had received interpreted PRS reports for 840 participants. Thirteen of these had at least one actionable gene result according to current ACMG guidelines.
Jonathan Mosley, a professor of medicine and biomedical informatics at Vanderbilt University, commended GenoVA for looking at both the outcomes of time to diagnosis and also healthcare utilization, saying that these constituted key study strengths.
"It seems very likely that this paper will drive up utilization," he said, "so it would be important to determine whether these costs offset the benefits in outcomes."
Mosley cautioned however, that screening studies like GenoVA can be prone to lead-time bias, wherein a disease that is detected by a screening or surveillance test earlier than it would have been otherwise can make outcomes like survival time appear to improve without necessarily affecting overall survival.
"A new screening biomarker for cancer will allow for detection of more cancers," Mosley said. "Even if you do nothing to treat the cancer, detecting it earlier will make it look like survival is longer. So thinking about their findings in this context would be important."
Mosley also cautioned that watching for adverse outcomes will be critical to fully assessing the benefits of clinical use of PRS.
"For instance," he said, "if the person undergoes a prostate biopsy as a result of this, that would count as utilization. But if they develop an infection from the biopsy, that is potentially a harm."
Although adverse outcome monitoring was not directly addressed in the GenoVA publication, Jason Vassy, a professor of medicine at Harvard Medical School and the study's lead investigator, said that this had been taken into consideration.
"We are monitoring the study for safety, including any adverse events considered related to study participation," he said. "None have occurred to date."
Mosley also commented that with the validity of PRS-based screening still to be demonstrated, there remains a question of how the outcomes of individuals deemed high-risk but with mid-range PRS values compare to those with PRS results in the upper range for risk.
"If the outcomes were the same as those in the high-risk group who also had high PRS values, it would suggest that the PRS may not be adding much," he said.
Vassy acknowledged that PRS is a continuous measure of risk and that while setting cutoffs for risk is somewhat arbitrary, they were still necessary in order to observe differences between groups for the purposes of the study.
"We created cut-offs for high risk to give the primary care providers some kind of guidepost for which patients might need some additional action in their health care," Vassy said. "Of course risk is on a continuum, but the end of the day, providers want to know at what point they should treat a given patient differently than they would have without the new information."
Although it is too early in the study to report results related to diagnoses and healthcare utilization, 307 participants have had at least one high-risk PRS result, and 54 of these have had two or more such results.
"If our PRS intervention accelerates the time to diagnosing new disease, next steps would include an evaluation of barriers and facilitators to the more widespread adoption of this new technology," Vassy said. "Some of these we are identifying and addressing in the GenoVA study, but more implementation science work remains."
Although the GenoVA study is geared toward establishing whether or not PRS may provide meaningful clinical benefits, it doesn't necessarily mean that meeting its endpoints would usher PRS-based screening into clinics.
"I think the use of PRS in routine screening is still some time off," he said. "But it is clearly an active area of research."
The VA has been actively studying the possible the use of PRS in prostate cancer screening through its Million Veterans Program and recently funded a large clinical trial of PRS-based prostate cancer screening that will begin enrolling soon.