Discovery Yields Possible Blood Test for Coronary Artery Disease
Scientists have identified a handful of genes in circulating blood associated with the presence and severity of coronary artery disease (CAD). If the findings are validated in a larger trial currently under way, it could lead to a molecular diagnostic requiring a simple blood sample from the patient to diagnose CAD, the leading cause of death in the United States.
The findings, appearing in the inaugural issue of Circulation: Cardiovascular Genetics, stem from research at Duke University Medical Center and CardioDx, a molecular diagnostics company in California.
"We have identified 14 genes in circulating blood cells that are exquisitely sensitive to the inflammatory changes that occur when plaque begins to accumulate in coronary arteries," says Dr. William E. Kraus, a member of the Duke Heart Center and the senior author of the study. "What's really exciting is that the expression of these genes is also related to the degree of stenosis, or blockage of the arteries. This means that a blood test based on these genes could tell us not only if someone has CAD, but also how bad the blockage in their arteries really is."
Coronary artery disease affects an estimated 15 million people in the U.S. The hallmark of the disorder is arterial stenosis, or a narrowing of the coronary arteries through a build up of plaque. Coronary arteries supply blood to the heart muscle, and any interruption in that supply can lead to a heart attack, sudden cardiac arrest, or death.
Diagnosing CAD can be a long and complex process, often involving multiple echocardiograms, stress tests and imaging techniques that expose the patient to radiation. But even if these tests indicate narrowing of the arteries, the full extent of the problem isn't known until a patient undergoes angiography, or coronary artery catheterization. Angiography is an invasive and fairly costly procedure that involves inserting a catheter into the coronary arteries to determine the extent of stenosis.
"A blood-based test to diagnose CAD would be less invasive and risky and would prevent patients from radiation exposure," says Kraus. "Patients who receive a positive test result might be able to short-circuit additional preliminary evaluations and head directly to coronary catheterization, where blockages could be treated."
Kraus worked with colleagues at the Helios Heart Center in Germany and CardioDx in identifying 41 patients who had undergone coronary catheterization. Twenty-seven of the patients had blockages in their arteries; 14 did not. Using whole genome microarray analysis, scientists identified 526 genes that behaved very differently between the two groups.
Further testing highlighted a subset of 14 genes that did the best job discriminating between patients with CAD and those who did not have it. Researchers tested those genes in blood taken from 215 catheterization patients at Duke
Investigators say that a gene expression "score" derived from adding the expression levels of the set of 14 genes is proportional to the degree of stenosis.
"We need to stress that we are not talking about a cause and effect relationship here," says Kraus. "We do not know if theses genes cause CAD in any way or if their altered expression is in response to the disease. What we do know, however, is that their collective ‘signature' is clearly associated with the presence of CAD."
Because the study involved a disproportionate number of men, researchers say the addition of more women and a larger study sample may yield even more genes that may be related to CAD in specific subsets of patients or that may narrow the list of the strongest predictors.
Scientists at Duke and CardioDx are continuing to evaluate the 14 genes in an ongoing prospective clinical trial, dubbed PREDICT (Personalized Risk Evaluation and Diagnosis in the Coronary Tree). The trial began in 2007 and is open at 28 sites throughout the United States.
"We are committed to developing a test that will enable more informed therapeutic decisions by allowing cardiologists to fine-tune their approach to diagnosing CAD while minimizing unneeded interventions," says Steven Rosenberg, PhD, chief scientific officer at CardioDx and a co-author of the study.
The research was funded by CardioDx.
Additional scientists from the company who contributed to the study include lead author James Wingrove, Susan Daniels, Amy Sehnert, Whittemore Tingley and Michael Elashoff. Additional co-authors include Lutz Buellesfeld and Eberhard Grube, from the Helios Heart Center; and Kristin Newby and Geoffrey Ginsburg from Duke.
Ginsburg holds equity in CardioDx.
# # #