Use of the Corus CAD Gene Expression Test in Coronary Artery Disease

The Corus CAD test (CardioDx) assesses whether or not a stable, non-diabetic patient’s symptoms are due to obstructive coronary artery disease (CAD). It has been clinically validated in multiple independent patient cohorts, including two prospective, multicenter U.S. studies, PREDICT¹ and COMPASS^2-3. A retrospective, multicenter chart review study⁴ and the prospective IMPACT^5-6 trial at Vanderbilt University demonstrated that Corus CAD use yields statistically significant and clinically relevant changes in patient management decisions in both primary care and cardiology settings. 

Covered by Medicare, the test is intended for use in non-diabetic stable patients who present with typical or atypical symptoms suggestive of CAD, with no known history of CAD, no prior myocardial infarction or revascularization procedure, and are not currently taking steroids, immunosuppressive agents or chemotherapeutic agents.

Tell us about your practice.

I am in a 12-physician cardiology private practice. Our team of specialists includes interventional cardiologists, an electrophysiologist and a lipidologist. We work at 10 different hospitals, but are predominantly at three larger hospitals in Baton Rouge, Louisiana: Our Lady of the Lake Regional Medical Center and Baton Rouge General Medical Center - Bluebonnet and Baton Rouge General Medical Center - Midcity. I perform coronary and peripheral interventions, and implant devices, such as pacemakers and defibrillators. As a group, we are aggressive with primary prevention with regard to lipid and hypertension management. 

It’s interesting that as an interventional cardiologist, you are also involved with primary care patients. 

That is true. Early prevention is something we advocate. We tend to get a lot of people referred to us much earlier than someone who is having active classic angina. Many people are referred simply for cardiac screenings and hypertension management, which means we have the opportunity to intervene fairly early in regard to lipid management. It’s an opportunity for us to deal more with coronary artery disease prevention than some cardiologists typically do.

Can you tell us about the Corus CAD test?

After investigating thousands of genes and their signals (measured as RNA), researchers developed an algorithm that integrates the expression levels of 23 genes that highly correlate with coronary artery stenosis of 50% or greater as measured by quantitative coronary angiography (QCA), also known as obstructive coronary artery disease. The Corus CAD test is a simple blood test that measures the expression levels of these 23 genes. It is a first-line test for the assessment of obstructive CAD in individuals who are non-diabetic and who have no previous interventions, coronary artery bypass, or documented coronary artery disease. After a blood draw, results are available in approximately three days. 

In the initial PREDICT¹ trial, patients undergoing a heart catheterization also underwent a Corus CAD test prior to the catheterization. The trial showed the higher the Corus CAD score, the more likely it was that the patient had a 50% or greater stenosis in one of the major epicardial vessels as verified by quantitative coronary angiography (QCA).  We know if someone has a high Corus CAD score, we are almost definitely going to find coronary artery disease. What we do not know is the extent of disease, whether it is a 50%+ lesion or diffuse coronary artery disease. A high score in the neighborhood of 28 or 29 means there is a 50% chance of an obstructive lesion  in a major coronary artery. 

If noninvasive testing is negative, yet the patient is symptomatic and has a high Corus CAD score and a strong family history of heart disease at the patient’s current age, I will consider coronary angiography for this patient. I have experience with a fair number of patients with significant disease who had high Corus CAD scores and warranted either surgical or percutaneous coronary intervention. 

COMPASS^2-3, a follow-up trial to PREDICT, compared the performance of Corus CAD to nuclear stress testing or myocardial perfusion imaging (MPI). The Corus CAD test actually outperformed MPI in overall accuracy. In COMPASS, the negative predictive value of the Corus CAD test was 96%, whereas the negative predictive value of MPI was only 88%.  This was a surprising observation and underscored the value of the Corus CAD test. 

Depending on whether I am excluding or confirming the presence of obstructive coronary artery disease,  I can start off with a Corus CAD test and potentially eliminate some noninvasive testing. For instance, in patients with low Corus CAD scores (≤15), the risk of significant coronary artery disease is very low, and it has been found that the major adverse event rate (stroke, heart attack, or death) at one year for low-scoring patients is also extremely low (0.5%).  Corus CAD is another valuable tool in the evaluation of heart disease. 

We have learned that diagnosing women can be challenging, because they tend to have more atypical presentations and more false positives in noninvasive testing.  When we do noninvasive imaging in women, quite often, breast tissue can create attenuation, leading to inaccurate imaging results, which can show up as normal catheterizations in women. 

Importantly, the Corus CAD test is the only sex-specific assessment tool for the assessment of obstructive coronary artery disease, and has been shown to work equally well in both men and women. In the PREDICT study, the Corus CAD test was found to be the only independent indicator of obstructive CAD in both women and men.

While the test is valuable to cardiologists, it is also valuable to the primary care physicians. If a patient is having atypical symptoms, many primary care physicians are not sure when to refer a patient for a cardiac evaluation. Using the Corus CAD test in patients with atypical symptoms of coronary artery disease, and who have some risk factors, is a good first step for the primary care physician. If the score comes back very low, then physicians can be very confident that the patient’s risk of obstructive coronary artery disease is low. If the score is intermediate or elevated, then the patient’s risk of obstructive coronary artery disease goes up. This is an appropriate opportunity to refer the patient to a cardiologist for evaluation. 

How long have you been using the Corus CAD test?

I have been using Corus CAD since the latter part of 2010. Early on, I had a few cases where the test made a significant impact on treatment decisions. As a result, I started using the test a fair amount.  At this point, I have used the test in over 160 cases. 

Patients often come in with typical or atypical chest pain, and have undergone some form of stress testing, whether stress echo or MPI.  Some patients come with negative MPI studies, yet their history and family history are both very concerning. 

I began using the Corus CAD test in patients with a family history of CAD who are about the same age as the family member who had a cardiac event, such as a myocardial infarction (MI), coronary artery bypass graft surgery (CABG), or percutaneous coronary intervention (PCI). I found a strong correlation between the extent of coronary artery disease and elevated Corus CAD scores in this population. 

I have experience with patients who actually had very severe coronary artery disease that was missed by standard noninvasive testing.  One case in particular involved a 72-year-old man who presented with vague chest pain symptoms and had a family member diagnosed with heart disease at approximately the same age. He had multiple risk factors, but was very active. I did an MPI on this patient, which came back negative. About 6 months later, I saw him for a follow-up on his lipids and hypertension.  I decided to do a Corus CAD test on this patient, and the patient’s score was found to be significantly elevated.  I then suggested the patient go to cath, despite a negative MPI, 6 months prior. 

When I performed a catheterization on this patient, I found severe left main and right coronary artery disease. There was a 95% right coronary lesion (Figure 2a) and at least a 70% left main lesion with some additional disease (Figure 2b). This patient represents the typical balanced perfusion reduction in MPI that appears normal. Here is someone who was literally a walking time bomb. After I performed the catheterization, everyone’s jaws dropped as they witnessed the blockage. Balanced ischemia is one of the downfalls of MPI and these are the people with the highest risk for a bad outcome. 

Corus CAD is a valuable test for assessing symptomatic patients at risk for coronary artery disease due to increased age, multiple risk factors, and atypical symptom presentations like fatigue, heartburn and shortness of breath. Corus CAD has become a very useful test for me to help confirm the presence of significant coronary artery disease. However, the strength of the test really lies in its high negative predictive value. For example, if a 45-year-old female comes in with chest pain, but has no risk factors, no family history and has a negative stress test, Corus CAD can be used as an alternative test to help clinicians exclude an obstructive CAD diagnosis. If the Corus CAD test is done prior to the stress test and comes back with a very low score, this patient’s risk of significant coronary artery disease is negligible. At this point, we can forego the stress test, reassure the patient, and pursue other causes for her symptoms. 

Corus CAD scores range from 1 to 40.  A score of 15 or below is considered low, and in the COMPASS study, the negative predictive value was 96%. High scores are defined as 27 and above. The intermediate score range is between 16 and 26. The lowest Corus CAD score in which I have documented significant coronary artery disease (95% blockage in the right coronary) was 18.  However, this patient was on nasal steroids, and I wonder if that influenced the lower score.

Do the results tell you anything about the plaque itself?

Patients with high scores almost always have coronary artery disease. What we do not know is whether it will be diffuse or somewhat focal. The Corus CAD test will tell me that there is some degree of plaque burden. Preliminary work looking at plaque composition does seem to show some correlation with softer or more vulnerable plaque, but it is too early to be definitive.

How has the Corus CAD test impacted your practice overall?

It is a decision-making tool. Sometimes the workup is straightforward. Patients present with what seems like classic angina and they just go straight to cath. If a patient presents with symptoms that are somewhat atypical and the patient has multiple risk factors, we are obligated to do some form of stress testing. However, if a patient presents with chest pain that is atypical, has no real risk factors, and no family history, the Corus CAD test becomes a great first option. I use the test in patients with recurrent chest pain who are at low risk of coronary artery disease. The test can help reassure the patient that the likelihood of obstructive coronary artery disease is extremely low. 

While the positive predictive value is not as strong, I will use the test in patients with high risk of coronary artery disease. I typically perform a stress test on these patients. However, if the test comes back negative, I am left wondering if there is balanced ischemia in this patient. At this point, the Corus CAD test becomes complimentary and helpful in my clinical decision-making. If the Corus CAD score is elevated, then I worry if the patient may have significant coronary artery disease. If the Corus CAD comes back low, I am reassured. 

Despite the overwhelming clinical benefits, I have patients who are not interested in taking statins for their dyslipidemia, mainly because of the side effects. For a stress test to be abnormal, there has to be high-grade disease; however, with the Corus CAD test, an intermediate to high score range will indicate some plaque burden. It may not be high-grade stenosis, but it tells me plaque development is occurring. The Corus CAD test is the objective test I use to convince my patients that statins are indicated.

I can’t emphasize the strength of the Corus CAD test is in its high negative predictive value. The positive predictive value is not as impressive, but still very useful.  From the primary care perspective, the test’s negative predictive value helps clinicians decide whether or not to send a patient to a cardiologist for evaluation.  As a cardiologist, the Corus CAD test helps me decide whether or not to do a cath despite a recent negative MPI, where balanced ischemia is always a possibility. 

Any final thoughts?

Many of the tests we use, such as nuclear stress testing (for example, MPI), coronary angiography, and computed tomography angiography (CTA), involve some degree of radiation. For example, the radiation exposure from an angiogram is equivalent to the total radiation exposure from 400-800 chest x-rays combined. In addition, noninvasive testing is costly, as are coronary angiograms. With its very high negative predictive value, the Corus CAD test may prove to be a cost-effective tool for excluding the diagnosis of obstructive CAD early in the assessment pathway, thus helping clinicians make better patient management decisions which can lead to improved overall healthcare resource utilization. 

If a patient comes back with recurrent typical or atypical chest pain and I am convinced it is not cardiac in origin, yet the patient thinks it is, a low Corus CAD test score can be used to reassure patients that their symptoms are not due to obstructive coronary artery disease. The flip side occurs when I am concerned with a patient’s risk factor profile, and their cardiac imaging result is negative. If the Corus CAD test score comes back elevated, there is good justification to proceed with an invasive approach at this point. Usually these are patients with atypical symptom presentations and who are generally overlooked. When used appropriately, the Corus CAD can be a very helpful decision-making tool for both primary care physicians and cardiologists. n

Dr. Daniel Fontenot can be contacted at (225) 769-0933 or at info@brcardiology.com.

References

1. Rosenberg S, Elashoff MR, Beineke P, et al. Multicenter validation of the diagnostic accuracy of a blood-based gene expression test for assessing obstructive coronary artery disease in nondiabetic patients. Ann Intern Med. 2010;153:425-434.
2. Blood-based gene expression test beats MPI for ruling out obstructive CAD. Diagnostic and Interventional Cardiology. December 14, 2011. Available online at https://www.dicardiology.com/article/blood-based-gene-expression-test-beats-mpi-ruling-out-obstructive-cad. Accessed January 9, 2013.
3. Voros S, Kraus W, Budoff MJ, Elashoff M, Wingrove J, Thomas G, Rosenberg S. Coronary artery plaque burden and stenosis by cardiovascular CT correlate with peripheral gene expression in 614 patients: results from the PREDICT and COMPASS studies. Presented November 5, 2012, at the American Heart Association Scientific Session 2012, Los Angeles. Available online at https://www.cardiodx.com/clinician-resources/publications-and-abstracts. Accessed January 9, 2013.
4. Conlin MF, Herman LE, Wilson, L, et al. The Use of a Personalized Gene Expression Test to Improve Decision Making in the Evaluation of Patients With Suspected Coronary Artery Disease. J Gen Intern Med. 2012;27:S540-S541.
5. ClinicalTrials.gov. Primary care providers use of a gene expression test in coronary artery disease diagnosis (IMPACT-PCP). Available online at https://www.clinicaltrials.gov/ct2/show/NCT01594411?term=Corus+IMPACT&rank=2. Accessed January 9, 2013.
6. McPherson J, Davis K, Yau M, Beineke P, Rosenberg S, Monane M, Fredi JL. Abstract 115: Improved diagnostic work-up of patients presenting to the cardiologist with symptoms of suspected obstructive coronary artery disease: results from the IMPACT (Investigation of a Molecular Personalized Coronary Gene Expression Test on Cardiology Practice Pattern) trial. Circ Cardiovasc Qual Outcomes. 2012; 5: A115. Available online at https://circoutcomes.ahajournals.org/cgi/content/short/5/3_MeetingAbstracts2012/A115?rss=1. Accessed January 9, 2013.