Background

Heart Disease

Heart disease is the leading cause of death in the United States, accounting for approximately one-third of all deaths in people over age 35.^1, The death rate is higher in men compared with women, and in blacks compared with whites but lower in Hispanic populations compared with blacks and whites. The most common form of heart disease is ischemic heart disease, also known as coronary artery disease (CAD).

Angina is the first symptom of CAD in approximately 50% of patients. However, women and the elderly are more likely to present with atypical symptoms such as nausea, vomiting, gastric discomfort, or atypical chest pain, which makes diagnosis more challenging.^2,

Diagnosis

Patients with signs and symptoms of obstructive CAD may be evaluated with a variety of tests according to prior risk. Coronary angiography is the criterion standard for diagnosing obstructive CAD but it is invasive and associated with a low but finite risk of harm. Coronary angiography also has a relatively low yield. In a study of nearly 400000 patients without known CAD undergoing elective coronary angiography, approximately 38% were positive for obstructive CAD (using the CAD definition, ³50% stenosis of the diameter of the left main coronary artery or ³70% stenosis of the diameter of a major epicardial or branch vessel >2.0 mm in diameter) and 41% if using the broader definition (³50% stenosis in any coronary vessel).^3, Thus, methods of improving patient risk prediction before invasive coronary angiography are needed.

In an initial proof-of-principle study of the Corus CAD score in patients referred for invasive coronary angiography, Wingrove et al (2008) evaluated 27 cases (96% symptomatic) with and 14 controls without angiographically defined CAD for expression of genes that differed significantly between the 2 groups, selecting 50 genes.^4, To that authors added 56 genes selected from relevant literature reports and evaluated the expression of these 106 genes in an independent set of 63 cases and 32 controls, resulting in the selection of 14 genes that independently and significantly discriminated between groups in multivariable analysis. The significance of 11 of these 14 genes was replicated in the third set of 86 cases and 21 controls. Expression of the 14 genes was proportional to maximal coronary artery stenosis in the combined cohort of 215 patients.

Elashoff et al (2011) described the final Corus CAD score development.^5, Investigators conducted two successive case-control gene expression discovery studies using samples from independent cohorts. Cases were angiographically defined as 75% or greater maximum stenosis in 1 major vessel, or 50% or greater in 2 vessels, and controls defined as less than 25% stenosis in all major vessels. Of clinical factors, diabetes had the most significant effect on gene expression; in the first case-control study in symptomatic patients (CATHeterization GENetics; n=195), expression of 42 genes in nondiabetic patients and 12 genes in diabetic patients were found to (p<0.05) discriminate significantly between cases and controls with no overlap. As a result, the second case-control study, in a subset of 198 patients from the prospective Personalized Risk Evaluation and Diagnosis In the Coronary Tree study, and final development of the assay was limited to nondiabetic patients (62% symptomatic). The participants were 76% male and 89% white. Final variable selection comprised the expression of 20 CAD-associated genes, 3 normalization genes, and terms for age and sex. The majority of the selected genes were immune and inflammatory-related. All terms were incorporated into an algorithm that resulted in an obstructive CAD score ranging from 1 to 40.

Regulatory Status

Clinical laboratories may develop and validate tests in-house and market them as a laboratory service. Laboratory-developed tests must meet the general regulatory standards of the Clinical Laboratory Improvement Amendments. The Corus® CAD test (CardioDx, Palo Alto, CA) is available under the auspices of the Clinical Laboratory Improvement Amendments. Laboratories that offer laboratory-developed tests must be licensed by the Clinical Laboratory Improvement Amendments for high-complexity testing. To date, the U.S. Food and Drug Administration has chosen not to require any regulatory review of this test.

Related Policies

• KIF6 Genotyping for Predicting Cardiovascular Risk and/or Effectiveness of Statin Therapy (Policy #052 in the Pathology Section)

1. The Corus® CAD test is considered not reasonable and necessary for patients who are currently taking steroids, immunosuppressive agents, or chemotherapeutic agents or for patients with:
o acute or previous myocardial infarction;
o high-risk unstable angina;
o a history of obstructive CAD;
o a previous revascularization procedure;
o a history of a previous invasive procedure to open a blocked or narrow artery;
o systemic infectious or systemic inflammatory conditions; or
o diabetes.

2. The Corus® CAD test is considered not reasonable and necessary when used for any of the following:
o to be used to screen for stenosis among patients who are asymptomatic and not considered at high-risk for CAD;
o to predict or detect response to therapy, or
o to help select the optimal therapy for patients.

For members enrolled in Medicaid and NJ FamilyCare plans, Horizon BCBSNJ applies the above medical policy.

FIDE SNP:

For members enrolled in a Fully Integrated Dual Eligible Special Needs Plan (FIDE-SNP): (1) to the extent the service is covered under the Medicare portion of the member’s benefit package, the above Medicare Coverage statement applies; and (2) to the extent the service is not covered under the Medicare portion of the member’s benefit package, the above Medicaid Coverage statement applies.

[RATIONALE: This policy was created in 2011 and has been updated regularly with searches of the MEDLINE database. The most recent literature update was performed through January 30, 2019.

Evidence reviews assess whether a medical test is clinically useful. A useful test provides information to make a clinical management decision that improves the net health outcome. That is, the balance of benefits and harms is better when the test is used to manage the condition than when another test or no test is used to manage the condition.

The first step in assessing a medical test is to formulate the clinical context and purpose of the test. The test must be technically reliable, clinically valid, and clinically useful for that purpose. Evidence reviews assess the evidence on whether a test is clinically valid and clinically useful. Technical reliability is outside the scope of these reviews, and credible information on technical reliability is available from other sources.

Gene Expression Testing for Suspected Stable Ischemic Heart Disease

Clinical Context and Test Purpose

The joint guidelines by the American College of Cardiology Foundation (2012) and 6 other medical associations on the diagnosis of stable ischemic heart disease provides details on the diagnostic pathway for evaluation and treatment of heart disease. The pathway is summarized in Figure 1 and in the following paragraphs. When patients present with signs and symptoms of obstructive coronary artery disease (CAD), the estimated risk (or pretest probability) of obstructive CAD is estimated using clinical characteristics such as age, sex, type of angina symptoms, smoking, and other comorbidities (eg, diabetes, hyperlipidemia).^2,6, The guidelines provide a table of pretest probabilities of CAD by age, sex, and type of angina adapted from the Diamond-Forrester tool.^2, For example, a woman age 30 to 39 with nonanginal chest pain has a 4% pretest probability of CAD and a man age 60 to 69 with typical anginal chest pain has a 94% pretest probability of CAD.

For patients initially assessed at low-risk (<10% pretest probability of obstructive CAD), no further testing is generally needed, and the patient can be observed and treated with medical therapy.^2, Patients at high-risk of obstructive CAD may proceed to coronary angiography if the symptoms or findings suggest a high-risk lesion.

The classification of intermediate risk varies in the literature but is frequently defined as a pretest probability between 10% and 90%. In patients with an intermediate pretest probability of obstructive CAD, noninvasive diagnostic methods, such as exercise or pharmacologic stress tests with or without imaging methods such as myocardial perfusion imaging (MPI), or coronary computed tomographic angiography may be recommended. The noninvasive testing used depends on patient characteristics such as the ability to exercise, electrocardiographic results, and other comorbidities as well as local expertise, availability of the testing modality, and patient preference. Some noninvasive imaging methods have potential risks of exposure to radiation and contrast material. After noninvasive testing, patients initially classified as having an intermediate pretest probability of obstructive CAD are further risk-stratified based on the estimated risk of coronary event or death using clinical data and results of noninvasive testing. The American College Cardiology Foundation (2012) joint guidelines also provide risk stratification following noninvasive testing.^2,For example, severe stress-induced left ventricular dysfunction (peak exercise left ventricular ejection fraction <45% or drop in left ventricular ejection fraction with stress ≥10%) indicates a high (>3%) annual risk of death or myocardial infarction; a 1-mm ST-segment depression occurring with exertional symptoms indicates an intermediate (1% to 3%) annual risk of death or myocardial infarction; and a normal stress or no change of limited resting wall motion abnormalities during stress indicates a low-risk (<1%) annual risk of death or myocardial infarction. Patients at high-risk of coronary event or death following noninvasive testing may proceed to coronary angiography.

CardioDx, the manufacturer of the gene expression score (GES; Corus CAD), has stated the test “complements and improves the current noninvasive assessment” of suspected obstructive coronary artery disease.^7, The manufacturer-supported registry collects data in the primary care setting and a decision impact study using registry data has suggested that the test may be used to identify stable, nonacute outpatients presenting with symptoms suggestive of obstructive CAD who can safely forgo referral to cardiology or advanced cardiac testing.^8, Other studies have been performed in patients who have been referred for invasive angiography and MPI.

The question addressed in this policy is: Does gene expression testing in patients with stable ischemic heart disease improve the net health outcome compared with standard clinical evaluation?

The following PICOTS were used to select literature to inform this review.

Figure 1. Diagnostic Pathway

Patients

The intended population are patients with suspected ischemic heart disease with stable angina. The manufacturer states that appropriate patients are those who do not have diabetes, without systemic infectious or systemic inflammatory conditions, and who are not currently taking steroids, immunosuppressive agents, or chemotherapeutic agents. The intended use population might be all such patients or a subset of them identified by risk stratification, depending on exactly how the test fits into the diagnostic pathway.

Interventions

A GES classifier (Corus CAD) has been developed based on expression levels derived from the previously described studies, in whole blood samples, of 23 genes plus patient age and sex. This information is used in an algorithm to produce a score from 1 to 40, with higher values associated with a higher likelihood of obstructive CAD. A score of less than 15 has been used to indicate a low risk of obstructive CAD.

Blood for the test is collected using a routine blood draw and stored between 2° and 10°C for up to 1 day before shipping to the CardioDx Commercial Laboratory, which is certified by Clinical Laboratory Improvement Amendments and accredited by the College of American Pathologists. The results are available within a few days.

The intervention of interest for assessing validity would be Corus CAD score added to current risk prediction models..

Comparators

The comparator would be clinical risk prediction models alone that estimate the pretest probability of obstructive CAD (eg, Diamond-Forrester). Noninvasive testing would be a comparator for determining whether a patient would be referred for coronary angiography..

The reference standard for diagnosing obstructive CAD is coronary angiography with obstructive CAD defined as any stenosis 50% or greater in the left main coronary artery or 70% or greater in any other coronary artery according to joint guidelines from the American College of Cardiology Foundation, the American Heart Association, and the Society for Cardiovascular Angiography and Interventions.^9,  However, this is also an imperfect reference standard for the outcome of a cardiac event.

Outcomes

Beneficial outcomes resulting from a true-negative test result are avoiding unnecessary subsequent testing. Harmful outcomes resulting from a false-positive test result are unnecessary noninvasive and invasive testing or receiving unnecessary treatment. Harmful outcomes resulting from a false-negative test result are increased risk of cardiovascular events and death.

In Figure 1,

(ie, a triage “rule-out” test), the test would need to identify precisely a group of patients that could safely forgo additional noninvasive testing; therefore, the sensitivity, negative predictive value (NPV) and negative likelihood ratio are key test performance characteristics.

Timing

The time period of interest for measuring the diagnostic performance is the time to obstructive CAD diagnosis. For assessing cardiovascular outcomes, 2.5 years is consistent with the PROspective Multicenter Imaging Study for Evaluation of chest pain (PROMISE) trial, which compared diagnostic strategies for CAD.^10,

Setting

The test has been evaluated in the outpatient and primary care settings as well as in patients referred for advanced cardiac testing.

Review of Evidence

There are three core characteristics for assessing a medical test. Whether imaging, laboratory, or other, all medical tests must be:

• Technically reliable
• Clinically valid
• Clinically useful.

Assessment of technical reliability focuses on specific tests and operators and requires review of unpublished and often proprietary information. Review of specific tests, operators, and unpublished data are outside the scope of this policy, and alternative sources exist. This policy focuses on the clinical validity and clinical utility.

Clinically Valid

A test must detect the presence or absence of a condition, the risk of developing a condition in the future, or treatment response (beneficial or adverse).

Characteristics and results of clinical validity studies evaluating the performance of the Corus CAD score for diagnosing obstructive CAD are shown in Tables 1 and 2. Four studies reported the performance characteristics for Corus CAD for diagnosing obstructive CAD.  Voora et al (2017), (PROMISE) was the largest study and it used the American Heart Association definition for obstructive CAD.^11, In this population of patients referred for nonurgent, noninvasive testing, the sensitivity was 73% (95% confidence interval [CI], 64% to 81%), the negative likelihood ratio was 0.56 (95% CI, 0.42 to 0.77), and the NPV was 94% (95% CI, 92% to 96%). The Rosenberg et al (2010),(Personalized Risk Evaluation and Diagnosis In the Coronary Tree[PREDICT])^12, and Thomas et al (2013),(Coronary Obstruction Detection by Molecular Personalized Gene Expression [COMPASS])^13, studies used a broader definition of obstructive CAD and enrolled few patients at intermediate risk (18% and 17%, respectively) based on clinical risk prediction rules. The sensitivities were 85% (95% CI, 79% to 90%) and 89% (95% CI, 78% to 95%) in PREDICT and COMPASS, respectively while the NPV rates were 83% (95% CI, 77 to 89) and 96% (95% CI, 93% to 99%). The thresholds used to identify obstructive CAD were not clear in Ladapo et al (2017).^8,  The studies are described in more detail in the following paragraphs.

Corus CAD score was validated in the prospective multicenter PREDICT study (2010) in which blood samples were collected from 526 nondiabetic patients who did not have systemic infectious or inflammatory conditions and who were not receiving immunosuppressive or chemotherapeutic agents with a clinical indication for coronary angiography but no known previous myocardial infarction, revascularization, or obstructive CAD (71% symptomatic).^12,This is the same cohort from which the second assay development case-control cohort was drawn.^5,Patients were sequentially allocated to development and validation sets. The development cohort was 58% male and 87% white. The validation cohort is described in the tables. Investigators defined obstructive CAD as 50% or greater stenosis in 1 or more major coronary arteries on quantitative coronary angiography, which they stated corresponded to 65% to 70% stenosis on clinical angiography. PREDICT compared the predictive accuracy of the GES test with clinical predictors and MPI stress testing. A 2014 follow-up publication, including patients from the gene discovery and algorithm development cohorts in combination with the validation cohort (n=1038), reported similar performance.^14,

In another follow-up from PREDICT, Lansky et al (2012) found that the Corus CAD score was an independent predictor of CAD in multivariate analysis, with odds ratios (ORs) of 2.53 (p=0.001) for the total study population and 1.99 (95%CI, 1.35 to 2.96; p=0.001) and 3.45 (95% CI, 1.97 to 5.91; p=0.001) for males and females, respectively.^15, In this analysis, MPI was not associated with any measures of CAD in the general population or when stratified by sex.

Thomas et al (2013) assessed the clinical validity and utility of the Corus CAD score for detection of obstructive CAD in symptomatic, nondiabetic patients without inflammatory conditions in a multicenter, prospective study, COMPASS.¹⁴ Obstructive CAD was defined as 50% or greater stenosis in 1 or more major coronary arteries on quantitative coronary angiography. The COMPASS sample base differed from the PREDICT sample by including patients who had received a referral for MPI but had not been referred for invasive coronary angiography. MPI-positive participants underwent invasive coronary angiography based on clinician judgment, and all other participants received coronary computed tomography angiography (CCTA). Of 537 enrolled patients, only 431 (80%) were evaluable, primarily due to refusal to undergo invasive coronary angiography or CCTA. The performance characteristics for MPI (core-lab) in this population were also provided as follows: sensitivity, 36% (95% CI, 24% to 50%); specificity, 90% (95% CI, 87% to 93%); PPV, 41% (95% CI, 28% to 56%); and NPV, 88% (95% CI, 84% to 92%). The sensitivity of MPI in COMPASS was lower than generally reported in the literature. Ladapo et al (2013) reported simulation analyses demonstrating how referral bias could have influenced the performance characteristics that have been reported in the literature.^16,

Voora et al (2017) evaluated the Corus CAD score in a cohort from the PROMISE trial funded by National Heart, Lung, and Blood Institute.^11, PROMISE was a randomized controlled trial (2015) that enrolled 10003 outpatients who were randomized to functional (ie, exercise, echocardiographic, or nuclear stress testing) or anatomic (ie,CTA) diagnostic testing.^17, Patients were symptomatic and at increased risk for CAD based on age and/or the presence of CAD risk factors, and presented with symptoms suggestive of obstructive CAD. An ancillary analysis of PROMISE patients was supported in part by the manufacturer and included 2370 PROMISE patients without diabetes who were not on anti-inflammatory medications and who had samples in the biorepository of sufficient quality for analysis. The definition of obstructive CAD was 70% or more stenosis in a major coronary artery or 50% or more left main stenosis using CTA data.

Several studies have evaluated Corus CAD in a cohort of patients from the PRESET (A Registry to Evaluate Patterns of Care Associated with the Use of Corus CAD in Real World Clinical Care Settings) registry. The PRESET registry is funded by the manufacturer. This registry enrolled patients from 21 primary care practices in the United States between August 2012 and August 2014. Patients had nonacute chest pain and typical or atypical symptoms of obstructive CAD without history of myocardial infarction or revascularization, diabetes, suspected acute myocardial infarction, high-risk unstable angina pectoris, New York Heart Association class III or IV heart failure symptoms, cardiomyopathy with an ejection fraction of 35% or less, severe cardiac valvular diseases, current systemic infectious or inflammatory condition, or recent treatment with an immunosuppressive or chemotherapeutic agent. A report by Ladapo et al (2017) is primarily focused on physician decision-making but includes a table of the Corus CAD score and advanced cardiac testing results for obstructive CAD in 84 patients.^8, Therefore, those data are included in the following tables. Subsequent reports focused on adults aged 65 and older (n=176) and women of all ages (n=288) with stable symptoms suggestive of obstructive CAD, showing higher referral rates for patients with a higher Corus CAD score.^18,19,

Table 1. Clinical Validity Study Characteristics of the Corus CAD Score for Diagnosing Obstructive CAD

Study	Study Populationa	Design	Reference Standard for Obstructive CAD	Threshold Score for Positive Corus CAD Score Test	Timing of Reference and Corus CAD Score Tests	Blinding of Assessors	Comment
Rosenberg  et al(2010)12, PREDICT	·   Referred for ICA ·   Mean age, >60 y ·   90% White ·   43% women 48% low risk, 18% Intermediate Risk, 34% high risk	Prospective	≥50% stenosis in ≥1 major coronary arteries by quantitative CA	14.75	Blood samples drawn before CA	Yes	·   PREDICT study validation cohort ·   Funded by manufacturer
Thomas  et al(2013)13, COMPASS	·   Referred for MPI stress testing ·   Mean age, 56 y ·   89% White ·   48% women 58% low risk, 17% Intermediate Risk, 25% high risk	Prospective	≥50% stenosis in ≥1 major coronary arteries by quantitative CA or CCTA	15	Blood samples  drawn before MPI and CA	Yes	·   COMPASS study ·   Funded by manufacturer
Voora et al (2017)11,PROMISE	·   Referred for nonurgent, noninvasive testing for suspected CAD ·   Median age, >60 y ·   91% White ·   53% women	Nonconcurrent, prospective	≥70% stenosis in a major coronary artery or ≥50% left main stenosis using CCTA	15	Blood samples  drawn before CA	Yes	·   PROMISE trial funded by NHLBI ·   PROMISE ancillary analysis funded by manufacturer
Ladapo  et al (2017)8, PRESET	·   Evaluated in primary care and referred for advanced cardiac testing ·   Proportion of women among those referred for advanced testing not reported	Prospective	Cardiac stress test or ICA (thresholds NR)	15	Blood samples  drawn before further testing	NR	·   PRESET registry ·   Funded by manufacturer

CA: coronary angiography; CAD: coronary artery disease; CCTA: coronary computed tomographic angiography; ICA: invasive coronary angiography;
MPI: myocardial perfusion imaging; NHLBI: National Heart, Lung, and Blood Institute; NR: not reported.

^a In all studies, patients were nondiabetic, without inflammatory conditions, and were not receiving immunosuppressive or chemotherapeutic agents.

Table 2. Clinical Validity Results of the Corus CAD Score for Diagnosing Obstructive CAD

Study	Initial N	Final N	Excluded Samples	Prevalence of Obstructive CAD	Sensitivity (95% CI), %	Specificity (95% CI), %	PPV (95% CI) %	NPV (95% CI) %	AUC (95% CI) %
Reference standard: ≥50% stenosis in ≥1 major coronary arteries by quantitative CA
Rosenberg et al (2010)12, PREDICT	649	525	·   Insufficient sample volume RNA yield: 43 ·   Genomic DNA: 78 ·   Quality control analysis: 2 ·   Unknown: 1	37%	85 (79 to 90)a	43 (38 to 49)a	46 (41 to 52)a	83 (77 to 89)a	0.70 (NR)
Thomas et al (2013)13, COMPASS	537	431	·   Refused CTA after negative MPI: 90 ·   Other incomplete data: 16	15%	89 (78 to 95)b	52 (47 to 57)b	24 (19 to 30)b	96 (93 to 99)b	0.79 (0.72 to 0.84)
Reference standard: ≥70% stenosis in a major coronary artery or ≥50% left main stenosis using CCTA
Voora et al (2017)11, PROMISE	2370	1137	Did not have site-read CTA data	10%	73 (64 to 81)a	48 (45 to 51)a	14 (11 to 17)a	94 (92 to 96)a	0.63 (0.57 to 0.68)
Reference standard: cardiac stress test or ICA (thresholds NR)
Ladapo et al (2017)8, PRESET	126	84	Testing results not available	12%	100 (59 to 100)a	18 (10 to 28)a	14 (7 to 25)a	100 (66 to 100)a	NR

AUC: area under the curve; CA: coronary angiography; CAD: coronary artery disease; CI: confidence interval; CCTA: coronary computed tomography angiography; CTA: computed tomography angiography; ICA: invasive coronary angiography; MPI: myocardial perfusion imaging; NPV: negative predictive value; NR: not reported; PPV: positive predictive value.

^a CIs not reported in publication; calculated based on data provided.

^b The performance characteristics for MPI (core-lab) in this population were also provided: sensitivity, 36% (95% CI, 24% to 50%); specificity, 90% (95% CI, 87% to 93%); PPV, 41% (95% CI, 28% to 56%); and NPV, 88% (95% CI, 84% to 92%).

Table 3. Relevance Gaps for Clinical Validity Studies of the Corus CAD Score for Diagnosing Obstructive CAD

Study	Population	Intervention	Comparator	Outcomes	Duration of Follow-Up
Rosenberg  et al (2010)12, PREDICT	2. Test use in current diagnostic pathway unclear 4. Study only includes patients referred for ICA and only 18% of patients were at intermediate risk 5. Racial minorities were not well-represented	None noted	2. Used broad obstructive CAD definition	3. Diagnostic performance characteristics not provided for clinical risk models; performance characteristics by sex not provided	None noted
Thomas et al (2013)13, COMPASS	2. Test use in current diagnostic pathway unclear 4. Only 17% of patients were at intermediate risk 5. Racial minorities were not well-represented	None noted	2. Used broad obstructive CAD definition	3. Diagnostic performance characteristics not provided for clinical risk models performance characteristics by gender not provided	None noted
Voora et al (2017)11, PROMISE	2. Test use in current diagnostic pathway unclear 5. Racial minorities were not well-represented	None noted	3. Performance characteristics for comparators not provided	3. Diagnostic performance characteristics calculated based on data provided; performance characteristics not provided for clinical risk models; performance characteristics by sex not provided	None noted
Ladapo et al (2017)8, PRESET	2. Test use in current diagnostic pathway unclear	None noted	1. Thresholds for diagnosis not given	3. Diagnostic performance characteristics not provided for clinical risk models; performance characteristics by sex not provided	None noted
Key	1.Intended use population unclear 2.Clinical context for test is unclear 3.Study population unclear 4.Study population not representative of intended clinical use 5.Study population is subpopulation of intended use	1.Classification thresholds not defined 2.Version used unclear 3.Not version currently in clinical use	1.Classification thresholds not defined 2.Not compared to credible reference standard 3.Not compared to other tests in use for same purpose	1.Study does not directly assess a key health outcome 2.Evidence chain or decision model not explicated 3.Key clinical validity outcomes not reported (sensitivity, specificity, predictive values) 4.Reclassification of diagnostic or risk categories not reported 5.Adverse events of the test not described (excluding minor discomforts and inconvenience of venipuncture or noninvasive tests)	1.Follow-up duration not sufficient with respect to natural history of disease (TP, TN, FP, FN cannot be determined)

CAD: coronary artery disease; FN: false negative; FP: false positive; ICA: invasive coronary angiography; TN: true negative; TP: true positive.

Table 4. Study Design and Conduct Gaps for Clinical Validity Studies of the Corus CAD Score for Diagnosing Obstructive CAD

Study	Selection	Blinding	Delivery of Test	Selective Reporting	Completeness of Follow-Up	Statistical
Rosenberg et al (2010)12, PREDICT	None noted	None noted	None noted	None noted	None noted	1.CIs not reported, calculated based on data provided
Thomas et al (2013)13, COMPASS	None noted	None noted	None noted	None noted	2. 90 patients with negative MPI refused CTA and were excluded; no description of these patients was provided	None noted
Voora et al (2017)11, PROMISE	None noted	None noted	None noted	None noted	None noted	1.CIs not reported, calculated based on data provided 2. No comparison to noninvasive testing provided
Ladapo et al (2017)8, PRESET	None noted	1. Blinding not reported	None noted	None noted	None noted	1. CIs not reported, calculated based on data provided 2. No comparison to noninvasive testing provided
Key	1.Selection not described 2.Selection not random nor consecutive (ie, convenience)	1.Not blinded to results of reference or other comparator tests	1.Timing of delivery of index or reference test not described 2.Timing of index and comparator tests not same 3.Procedure for interpreting tests not described 4.Expertise of evaluators not described	1.Not registered 2.Evidence of selective reporting 3.Evidence of selective publication	1.Inadequate description of indeterminate and missing samples 2.High number of samples excluded 3.High loss to follow-up or missing data	1.CIs and/or p values not reported 2.No statistical test reported to compare to alternatives

CAD: coronary artery disease; CI: confidence interval; CTA: computed tomography angiography; MPI: myocardial perfusion imaging.

Net reclassification for the Corus CAD score compared with other tests for the diagnosis of obstructive CAD was performed in Rosenberg et al (2010)^12, and Thomas et al (2013)^13, and are shown in Table 5 below. In Rosenberg et al (2010), the Corus CAD, Diamond-Forrester, and expanded clinical model scores were prospectively categorized as low (0% to <20%), intermediate (≥20% to <50%), or high (≥50%) risk for obstructive CAD. MPI results were categorized as negative (no defect or possible fixed or reversible defect) or positive (fixed or reversible defect). In Thomas et al (2013), Corus CAD scores were categorized as low (≤15), intermediate (16-27), and high (≥28). The Diamond-Forrester and Morise scores were categorized as low (<15%), medium (≥15 to ≤50%), or high likelihood (>50%). It was not clear how the cutoffs were chosen in Thomas et al (2013).

As described in the Clinical Context section of this review, the pretest probability cutoffs from clinical models used for risk stratification vary in the literature, but intermediate risk frequently ranges from 10% to 90%. Net reclassification using this cutoff has not been reported.

Table 5. Net Reclassification Index for the Corus CAD Score vs Other Modalities for Diagnosing Obstructive CAD

Author (Year)	Net Reclassification Improvementa for Corus CAD score vs Second Modality (95% CI)
	Myocardial Perfusion Imaging
	Site-Read	Core-Lab	Diamond-Forrester	Morise	Expanded Clinical Model
Rosenberg et al (2010)12, PREDICT	21% (NR)	NR	20% (NR)	NR	16% (NR)
p	<0.001		<0.001		<0.001
Thomas et al (2013)13,COMPASS	26% (NR)	11% (NR)	28% (NR)	60% (NR)	NR
p	NR	NR	NR	NR	NR

CI: confidence interval; NR: not reported; CAD: coronary artery disease.

^a Net reclassification improvement quantifies the difference between the proportion of patients correctly reclassified from an incorrect initial classification and the proportion incorrectly reclassified from a correct initial classification.

Voros et al (2014) pooled results from PREDICT and COMPASS to compare Corus CAD score with computed tomography imaging for detecting plaque burden (coronary artery calcium [CAC]), and luminal stenosis.^20, Six hundred ten patients, 216 from PREDICT (19% of enrolled patients) and 394 from COMPASS (73% of enrolled patients), who had undergone CAC scoring, CTA, and Corus CAD score were included. Mean age was 57 years; 50% were female, and approximately 50% used statin medication. Prevalence of obstructive CAD (≥50% stenosis) was 16% in the PREDICT cohort (patients referred for coronary angiography) and 13% in the COMPASS cohort (patients referred for MPI). In linear regression analyses, Corus CAD scores statistically and significantly correlated with CAC (r=0.50), the number of arterial segments with any plaque (r=0.37), overall stenosis severity (r=0.38), and maximum luminal stenosis (r=0.41) (all p<0.01), but the strength of the correlations was modest. Several Corus CAD score cutoffs were explored (eg, to maximize diagnostic accuracy). Results using a cutoff of 15 points are shown in Table 6. For detecting luminal stenosis of 50% or greater, the Corus CAD score PPV and NPV were 23% and 95%, respectively. For detecting clinically significant CAC (≥400), the Corus CAD score PPV and NPV were 14% and 97%, respectively. Limitations of the study included a lack of clinical outcomes (eg, survival, morbidity) and lack of comparison with CAC and CTA for predicting these outcomes (ie, incremental Corus CAD score predictive value was not assessed).

Table 6. Performance of Corus CAD and Diamond-Forrester Classification for Coronary Artery Plaque Burden and Luminal Stenosis: Pooled PREDICT and COMPASS Analysis

Outcome	Corus CAD AUROC (95% CI)	Diamond-Forrester AUROC (95% CI)	Sensitivity, %	Specificity, %	PPV, %	NPV, %
Plaque burdena
CAC >0	0.75 (0.71 to 0.79)	0.65 (0.61 to 0.69)	71	62	65	68
CAC ≥400	0.75 (0.68 to 0.82)	0.61 (0.53 to 0.69)	84	49	14	97
Luminal stenosis by CTA
≥50%	0.75 (0.70 to 0.80)	0.65 (0.59 to 0.71)	84	51	23	95
≥70%	0.75 (0.67 to 0.83)	0.63 (0.53 to 0.73)	90	48	8	99

Adapted from Voros et al (2014).^20,
AUROC: area under the receiver operating characteristic curve; CAC: coronary artery calcium; CAD: coronary artery disease; CI: confidence interval; CTA: computed tomography angiography; NPV: negative predictive value; PPV: positive predictive value.

^a Long-term outcomes are generally excellent for patients with CAC >0 and substantially worse for patients with CAC >400.

Subsection Summary: Diagnostic Performance

The diagnostic pathway for CAD includes information from medical history, along with age and sex, stress testing, and imaging. It is not clear how the Corus CAD gene expression test fits in the current diagnostic pathway and how results would be used to change current guideline-based risk stratification before and/or after other noninvasive testing. Results of two validation studies (PREDICT, COMPASS) have reported the test may improve CAD prediction beyond the Diamond-Forrester prediction model. In the COMPASS study, the sensitivity and NPV of the Corus CAD score in diagnosing obstructive CAD was superior to MPI in patients referred for MPI testing. However, in that study, the reported sensitivity of MPI was considerably lower than that generally reported in the literature. Neither PREDICT nor COMPASS used the guideline definition of obstructive CAD as the reference standard and had relatively few patients at intermediate risk based on clinical prediction rules. The sensitivity and NPV of clinical models were not reported. An analysis of a cohort from the PROMISE trial including patients with an intermediate pretest probability of obstructive CAD confirmed a high NPV for the Corus CAD score.

The test excludes patients with diabetes, acute and chronic inflammatory conditions, and such patients are expected to be common among those being evaluated for obstructive CAD. Thus applicability to clinical practice may be narrow. Although the test is marketed as a sex-specific test, performance characteristics by sex and age were not provided. One study reported that the Corus CAD score was associated with obstructive CAD in both men (OR=1.99; 95% CI, 1.35 to 2.96) and women (OR=3.45; 95% CI, 1.97 to 5.91). The gene selection, algorithm development, and validation studies have been performed in populations that were approximately 90% white.

Net reclassification has been reported comparing the Corus CAD score with other clinical prediction tools and MPI. While the pretest probability cutoffs from clinical models used for risk stratification vary in the literature, intermediate risk frequently ranges from 10% to 90% and net reclassification using this cutoff has not been reported.

Prognostic Performance

Publications from four of the previously described studies have reported performance of the Corus CAD score in the prognosis of cardiovascular events. Table 7 summarizes the results. Rosenberg et al (2012) published a follow-up report from PREDICT on the association between Corus CAD score and subsequent major adverse cardiac events (MACE), including myocardial infarction, stroke/transient ischemic attack, all-cause mortality, and coronary revascularization.^21,

In Thomas et al (2013), patients were followed for 6 months after Corus CAD testing, with 420 of 431 completing follow-up.^13, MACE (nonfatal myocardial infarction, stroke/transient ischemic attacks, or all-cause mortality) and revascularization events were recorded. Only two MACE events occurred.

Voora et al (2017) included analysis of 2370 PROMISE patients with samples in the biorepository who were followed for a median of 25 months.^11, The association between the Corus CAD score and a composite outcome of death, myocardial infarction, revascularization, or unstable angina was statistically significant after adjustment for the Framingham Risk Score. The association was driven primarily by the revascularization component. When revascularization was removed from the composite, there was no longer a significant association between the Corus CAD score and the outcome after adjusting for the Framingham Risk Score. A low Corus CAD score was associated with a low-risk (1.6%) of revascularization and an NPV of 98% (CI not reported).

Ladapo et al (2018) and Gul et al (2019) evaluated the association between Corus CAD scores and cardiovascular events at 12 months in elderly adults (n=176) and women (n=288) from the PRESET registry.^18,19,  In adults 65 years of age or older the incidence of major adverse cardiovascular events or revascularization was 0% in patients with a low Corus CAD score and 10% in patients with a higher Corus CAD score (p=0.04). In the cohort of women of all ages, the incidence of major cardiac events was not statistically different between women with a low Corus CAD score (1.3%) and those with a higher Corus CAD score (4.2%, p=0.16).

Table 7. Clinical Validity Results of the Corus CAD Score for Prognosticating Cardiovascular Events

Author	N	Event	Incidence	Sens (95% CI)	Spec (95% CI)	PPV (95% CI)	NPV (95% CI)	Association (95% CI)
Rosenberg et al (2012)21,	1160	12-mo MACEa	1.5	82 (NR)	34 (NR)	1.8 (NR)	99 (NR)	OR=2.41 (0.74 to 10.5)
		12-mo MACEa or revascularizations	25	86 (NR)	41 (NR)	33 (NR)	90 (NR)	OR=4.32 (3.02 to 6.25)
Thomas et al (2013)13,	420	6-mo revascularizations or MACEa	6.7	96 (NR)	NR	NR	99 (NR)	NR
Voora et al (2017)11,	2370	Death, MI, or UA with median 25-mo follow-up	2.6	NR	NR	NR	NR	HR=0.98 (0.52 to 1.87)b
		Death, MI, UA, or revascularization with median 25-mo follow-up	6.0	NR	NR	NR	NR	HR=1.70 (1.10 to 2.64)b

Values are percent unless otherwise indicated.
CI: confidence interval; MACE: major adverse cardiac events; MI: myocardial infarction; NPV: negative predictive value; NR: not reported; OR: odds ratio; HR: hazard ratio; PPV: positive predictive value; Sens: sensitivity; Spec: specificity; UA: unstable angina

^a MACE included MI, stroke/transient ischemic attack, all-cause mortality.

^b Adjusted for Framingham Risk Score.

Subsection Summary: Prognostic Performance

There is less evidence on the association between the Corus CAD score and cardiovascular events. The available evidence provides a preliminary indication that a Corus CAD score of 15 or less identifies a group unlikely to require revascularization within 2 years. No data was given regarding which revascularizations were planned vs emergent; eg, information is needed describing how many revascularizations were performed to alleviate symptoms, for progression to unstable angina, or to decrease the risk of cardiac outcomes such as death, heart failure, or myocardial infarction. More data are needed on coronary events other than revascularizations. Notably, CIs for performance characteristics are lacking in these studies.

Section Summary: Clinically Valid

There is uncertainty regarding the role of the test in the diagnostic pathway. The proposed strategy for integrating the results of the test with current guidelines for risk stratification before and/or after other noninvasive testing is not clear. The diagnostic strategy incorporating the Corus CAD test should be explicitly described so that it is clear which existing data are relevant for evaluating the proposed use. Proposed changes in stratification compared with existing guidelines are needed so that net reclassification analyses compared with guideline recommended stratification can be constructed. Decision models of a strategy incorporating the Corus CAD score into the guideline recommendations would be useful.

The Corus CAD score is correlated with the presence of obstructive CAD. The PREDICT and COMPASS studies reported that the GES is superior to the Diamond-Forrester model and to MPI for predicting obstructive CAD. However, the available studies do not specify the use of the test in the guideline, recommended diagnostic pathway for stable ischemic heart disease. Therefore, it is not possible to make conclusions about clinical validity. The test provides scores that are age- and sex-specific and the manufacturer’s website states the test is the “the first sex-specific test for obstructive CAD”.^7,Performance characteristics by sex and age were reported from a safety analysis of registry data. A high Corus CAD score was associated with adverse cardiac events in older adults (both men and women), but this association was not statistically significant when assessed in the cohort of women.

Clinically Useful

A test is clinically useful if the use of the results informs management decisions that improve the net health outcome of care. The net health outcome can be improved if patients receive correct therapy, or more effective therapy, or avoid unnecessary therapy, or avoid unnecessary testing.

Direct Evidence

Direct evidence of clinical utility is provided by studies that have compared health outcomes for patients managed with and without the test. Because these are intervention studies, the preferred evidence would be from randomized controlled trials.

There is no direct evidence from randomized controlled trials.

Chain of Evidence

Indirect evidence on clinical utility rests on clinical validity. If the evidence is insufficient to demonstrate test performance, no inferences can be made about clinical utility.

To develop a chain of evidence or a decision model requires explication of the elements in the model and evidence that is sufficient to demonstrate each of the links in the chain of evidence or the validity of the assumptions in the decision model. A chain of evidence or decision model must be constructed so to permit comparison between a diagnostic strategy including Corus CAD testing and a strategy of no Corus CAD testing. The Corus CAD test is associated with the diagnosis of obstructive CAD. The Corus CAD test classifies patients into clinically credible diagnostic groups (low- and high-risk of obstructive CAD) that were defined a priori and evaluated in prospective studies. However, it is not clear how the test fits in the current diagnostic pathway and how results would be used to change current guideline-based risk.

Patients managed without the Corus CAD test should be evaluated according to established guidelines for the noninvasive evaluation of patients with stable ischemic heart disease.^2, Studies examining patient outcomes of Corus CAD testing have primarily analyzed changes in physician management as an outcome.

The Investigation of a Molecular PersonAlized Coronary Gene Expression Test (IMPACT)-CARDiology Practice Pattern study (2013) compared a prospective cohort with matched historical controls to evaluate whether the Corus CAD test altered cardiologist evaluation and clinical management of CAD.^22, CAD was categorized by authors as no CAD (0% stenosis), CAD with 50% or less stenosis, or CAD with more than 50% stenosis. Eighty-eight patients were enrolled and 83 included in the final analysis. The matched cohort comprised 83 patients selected with similar distributions of age, sex, and clinical risk factors evaluated at a participating clinic within the past 3 to 30 months. Diagnostic testing plans were changed for 58% of patients in the prospective cohort (95% CI, 46% to 69%; p<0.001) with a greater reduction in testing intensity (39%) compared with increased testing intensity (19%). Compared with the historical control group, the prospective cohort had a 71% reduction in overall diagnostic testing (p<0.001).

IMPACT-Primary Care Practice Pattern (PCP) (2014) evaluated whether having the Corus CAD altered primary care providers’ diagnostic evaluation and clinical management of stable, nonacute, nondiabetic patients presenting with CAD symptoms.^23, Nine primary care providers at 4 centers evaluated 261 consecutive patients, 251 (96%) of whom were eligible for participation. Clinicians documented their pretest impressions and recommendations for further evaluation and management on a clinical report form. All patients underwent Corus CAD testing. The primary outcome was the change in patient management between preliminary and final treatment plans. Diagnostic testing plans were changed for 58% of patients, with reductions in testing intensity more common (64%) than increases (34%; p<0.001). No study-related MACE were observed in 247 (98%) patients who had at least 30 days of follow-up.

The REGISTRY 1 study (2015) assessed the impact of having the Corus CAD on patient management decisions by examining the association between Corus CAD results and posttest referral patterns.^24, Primary care practitioners at 7 centers evaluated 342 stable, nonacute, nondiabetic patients presenting with CAD symptoms. All patients underwent Corus CAD testing. Of 167 patients with low (≤15) Corus CAD score, 10 (6%) were referred for further cardiac evaluation compared with 122 (70%) of 175 patients in the high Corus CAD score group (p<0.001). Over a mean follow-up of 264 days, there were 5 MACE, 2 in the low Corus CAD score group and 3 in the high Corus CAD score group. Of 21 patients who underwent elective invasive coronary angiography, 1 (50%) of 2 in the low Corus CAD score group and 8 (42%) of 19 in the high Corus CAD score group had obstructive findings.

Ladapo et al (2015) pooled results for women who participated in the IMPACT-PCP (n=140) and REGISTRY 1 (n=180) studies to evaluate the impact of Corus CAD score on further cardiac evaluation (n=320).^25, Referral rate for further cardiac evaluation was 4% for women with low Corus CAD score (n=248) vs 83% for women with elevated Corus CAD score (n=72).

The Ladapo et al (2017) analysis of the 566 patients from the PRESET registry (described previously) evaluated the association between the Corus CAD score and cardiac referrals (referral to cardiology or further cardiac testing).^8, Ten percent (26/252) of low Corus CAD score patients were referred vs 44% (137/314) of high Corus CAD score patients. After adjusting for age, sex, body mass index, smoking status, hypertension, and dyslipidemia, the association between Corus CAD score and referral rate remained statistically significant (OR=0.15; 95% CI, 0.10 to 0.24; p<0.001). With 1 year of follow-up, MACE and revascularizations were noted in 3 (1.2%) of 252 low Corus CAD score patients and 14 (4.5%) of 314 high Corus CAD score patients (p=0.03).

Section Summary: Clinically Useful

There are no rigorous studies comparing clinical outcomes for patients managed using Corus CAD testing with alternative methods for stable ischemic heart disease (ie, no direct evidence that the test is clinically useful). Currently, it is unclear whether a chain of evidence can be constructed because of the lack of evidence on use of the test in the intermediate risk population. .

Summary of Evidence

For individuals who have suspected stable ischemic heart disease without diabetes or inflammatory conditions who receive gene expression testing, the evidence includes retrospective case-control and prospective cohort studies. The relevant outcomes are overall survival, disease-specific survival, test accuracy and validity, change in disease status, morbid events, and resource utilization. The diagnostic pathway for CAD includes information from a medical history, along with age and sex, stress testing, and imaging. Newer noninvasive methods are being tested, such as gene expression testing. It is not clear how the Corus CAD gene expression test fits in the current diagnostic pathway and how results would be used to change current guideline-based risk stratification before and/or after other noninvasive testing. Results of two validation studies (PREDICT, COMPASS) have reported the test may improve CAD prediction beyond the Diamond-Forrester prediction model. In the COMPASS study, the sensitivity and NPV of the Corus CAD score in diagnosing obstructive CAD was superior to MPI in patients referred for MPI testing. However, in that study, the reported sensitivity of MPI was considerably lower than that generally reported in the literature. Neither PREDICT nor COMPASS used the guideline definition of obstructive CAD as the reference standard and had relatively few patients at intermediate risk based on clinical prediction rules. The sensitivity and NPV of clinical models were not reported. An analysis of a cohort from the PROMISE trial including patients with intermediate pretest probability of obstructive CAD confirmed a high NPV for the Corus CAD score. The test also has been shown to have some predictive ability of future revascularization; too few major cardiac events have been observed during the limited duration of follow-up to assess predictive ability for that outcome. Evidence for the Corus CAD score has not directly demonstrated that the test is clinically useful and a chain of evidence cannot be constructed to supports its utility. The evidence is insufficient to determine the effects of the technology on health outcomes.

SUPPLEMENTAL INFORMATION

Practice Guidelines and Position Statements

American Heart Association

The AHA(2012) released a policy statement on genetics and cardiovascular disease.^26, Gene expression testing is not specifically mentioned. Generally, the AHA supported recommendations issued in 2000 by a now defunct Advisory Committee to the U.S. Department of Health and Human Services, which stated: “No test should be introduced in the market before it is established that it can be used to diagnose and/or predict a health-related condition in an appropriate way.”^27,

The AHA (2017) released a scientific statement on the expressed genome in cardiovascular diseases and stroke.^28, The statement summarized the clinical validity and utility evidence for the Corus CAD score, stating “…the Corus CAD test is a clinically available diagnostic test that has been evaluated, has been deemed to be valid and useful.…”

American College of Cardiology Foundation et al

The joint guidelines of the American College of Cardiology Foundation (2012) and 6 other medical societies for the diagnosis and management of patients with stable ischemic heart disease did not mention the gene expression score.^2, The 2014 update to these guidelines also did not mention the gene expression score.^6,

U.S. Preventive Services Task Force Recommendations

Not applicable.

Ongoing and Unpublished Clinical Trials

A search of ClinicalTrials.gov in December 2017 did not identify any ongoing or unpublished trials that would likely influence this review.]
________________________________________________________________________________________

Horizon BCBSNJ Medical Policy Development Process:

This Horizon BCBSNJ Medical Policy (the “Medical Policy”) has been developed by Horizon BCBSNJ’s Medical Policy Committee (the “Committee”) consistent with generally accepted standards of medical practice, and reflects Horizon BCBSNJ’s view of the subject health care services, supplies or procedures, and in what circumstances they are deemed to be medically necessary or experimental/ investigational in nature. This Medical Policy also considers whether and to what degree the subject health care services, supplies or procedures are clinically appropriate, in terms of type, frequency, extent, site and duration and if they are considered effective for the illnesses, injuries or diseases discussed. Where relevant, this Medical Policy considers whether the subject health care services, supplies or procedures are being requested primarily for the convenience of the covered person or the health care provider. It may also consider whether the services, supplies or procedures are more costly than an alternative service or sequence of services, supplies or procedures that are at least as likely to produce equivalent therapeutic or diagnostic results as to the diagnosis or treatment of the relevant illness, injury or disease. In reaching its conclusion regarding what it considers to be the generally accepted standards of medical practice, the Committee reviews and considers the following: all credible scientific evidence published in peer-reviewed medical literature generally recognized by the relevant medical community, physician and health care provider specialty society recommendations, the views of physicians and health care providers practicing in relevant clinical areas (including, but not limited to, the prevailing opinion within the appropriate specialty) and any other relevant factor as determined by applicable State and Federal laws and regulations.

___________________________________________________________________________________________________________________________

Index:
Gene Expression Testing in the Evaluation of Patients With Stable Ischemic Heart Disease
Gene Expression Testing to Predict Coronary Artery Disease
Corus CAD

References:
1. Xu J, Kochanek KD, Murphy SL, et al. Deaths: final data for 2007. Natl Vital Stat Rep. May 2010;58(19):1-19. PMID 25075874.

2. Fihn SD, Gardin JM, Abrams J, et al. 2012 ACCF/AHA/ACP/AATS/PCNA/SCAI/STS Guideline for the diagnosis and management of patients with stable ischemic heart disease: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines, and the American College of Physicians, American Association for Thoracic Surgery, Preventive Cardiovascular Nurses Association, Society for Cardiovascular Angiography and Interventions, and Society of Thoracic Surgeons. J Am Coll Cardiol. Dec 18 2012;60(24):e44-e164. PMID 23182125.

3. Patel MR, Peterson ED, Dai D, et al. Low diagnostic yield of elective coronary angiography. N Engl J Med. Mar 11 2010;362(10):886-895. PMID 20220183.

4. Wingrove JA, Daniels SE, Sehnert AJ, et al. Correlation of peripheral-blood gene expression with the extent of coronary artery stenosis. Circ Cardiovasc Genet. Oct 2008;1(1):31-38. PMID 20031539.

5. Elashoff MR, Wingrove JA, Beineke P, et al. Development of a blood-based gene expression algorithm for assessment of obstructive coronary artery disease in non-diabetic patients. BMC Med Genomics. Mar 28 2011;4(1):26. PMID 21443790.

6. Fihn SD, Blankenship JC, Alexander KP, et al. 2014 ACC/AHA/AATS/PCNA/SCAI/STS focused update of the guideline for the diagnosis and management of patients with stable ischemic heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines, and the American Association for Thoracic Surgery, Preventive Cardiovascular Nurses Association, Society for Cardiovascular Angiography and Interventions, and Society of Thoracic Surgeons. J Am Coll Cardiol. Nov 04 2014;64(18):1929-1949. PMID 25077860.

7. CardioDx. Corus CAD Product Overview. 2018; http://www.cardiodx.com/corus-cad/product-overview/. Accessed January 8, 2018.

8. Ladapo JA, Budoff M, Sharp D, et al. Clinical utility of a precision medicine test evaluating outpatients with suspected obstructive coronary artery disease. Am J Med. Apr 2017;130(4):482.e411-482.e417. PMID 27993573.

9. Levine GN, Bates ER, Blankenship JC, et al. 2011 ACCF/AHA/SCAI Guideline for Percutaneous Coronary Intervention. A report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines and the Society for Cardiovascular Angiography and Interventions. J Am Coll Cardiol. Dec 6 2011;58(24):e44-122. PMID 22070834.

10. Douglas PS, Hoffmann U, Lee KL, et al. PROspective Multicenter Imaging Study for Evaluation of chest pain: rationale and design of the PROMISE trial. Am Heart J. Jun 2014;167(6):796-803.e791. PMID 24890527.

11. Voora D, Coles A, Lee KL, et al. An age- and sex-specific gene expression score is associated with revascularization and coronary artery disease: Insights from the Prospective Multicenter Imaging Study for Evaluation of Chest Pain (PROMISE) trial. Am Heart J. Feb 2017;184:133-140. PMID 28224927.

12. 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. Oct 5 2010;153(7):425-434. PMID 20921541.

13. Thomas GS, Voros S, McPherson JA, et al. A blood-based gene expression test for obstructive coronary artery disease tested in symptomatic nondiabetic patients referred for myocardial perfusion imaging the COMPASS study. Circ Cardiovasc Genet. Apr 2013;6(2):154-162. PMID 23418288.

14. Daniels SE, Beineke P, Rhees B, et al. Biological and analytical stability of a peripheral blood gene expression score for obstructive coronary artery disease in the PREDICT and COMPASS studies. J Cardiovasc Transl Res. Oct 2014;7(7):615-622. PMID 25119856.

15. Lansky A, Elashoff MR, Ng V, et al. A gender-specific blood-based gene expression score for assessing obstructive coronary artery disease in nondiabetic patients: results of the Personalized Risk Evaluation and Diagnosis in the Coronary Tree (PREDICT) trial. Am Heart J. Sep 2012;164(3):320-326. PMID 22980297.

16. Ladapo JA, Blecker S, Elashoff MR, et al. Clinical implications of referral bias in the diagnostic performance of exercise testing for coronary artery disease. J Am Heart Assoc. Dec 13 2013;2(6):e000505. PMID 24334965.

17. Douglas PS, Hoffmann U, Patel MR, et al. Outcomes of anatomical versus functional testing for coronary artery disease. N Engl J Med. Apr 2 2015;372(14):1291-1300. PMID 25773919.

18. Ladapo, JJ, Budoff, MM, Sharp, DD, Kuo, JJ, Huang, LL, Maniet, BB, Herman, LL, Monane, MM. Utility of a Precision Medicine Test in Elderly Adults with Symptoms Suggestive of Coronary Artery Disease. J Am Geriatr Soc, 2017 Dec 7;66(2). PMID 29210056.

19. Gul, BB, Lansky, AA, Budoff, MM, Sharp, DD, Maniet, BB, Herman, LL, Kuo, JJ, Huang, LL, Monane, MM, Ladapo, JJ. The Clinical Utility of a Precision Medicine Blood Test Incorporating Age, Sex, and Gene Expression for Evaluating Women with Stable Symptoms Suggestive of Obstructive Coronary Artery Disease: Analysis from the PRESET Registry. J Womens Health (Larchmt), 2019 Jan 18. PMID 30653377.

20. Voros S, Elashoff MR, Wingrove JA, et al. A peripheral blood gene expression score is associated with atherosclerotic plaque burden and stenosis by cardiovascular CT-angiography: results from the PREDICT and COMPASS studies. Atherosclerosis. Mar 2014;233(1):284-290. PMID 24529158.

21. Rosenberg S, Elashoff MR, Lieu HD, et al. Whole blood gene expression testing for coronary artery disease in nondiabetic patients: major adverse cardiovascular events and interventions in the PREDICT Trial. J Cardiovasc Transl Res. Jun 2012;5(3):366-374. PMID 22396313.

22. McPherson JA, Davis K, Yau M, et al. The clinical utility of gene expression testing on the diagnostic evaluation 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. Crit Pathw Cardiol. Jun 2013;12(2):37-42. PMID 23680805.

23. Herman L, Froelich J, Kanelos D, et al. Utility of a genomic-based, personalized medicine test in patients presenting with symptoms suggesting coronary artery disease. J Am Board Fam Med. Mar-Apr 2014;27(2):258- 267. PMID 24610188.

24. Ladapo JA, Lyons H, Yau M, et al. Enhanced assessment of chest pain and related symptoms in the primary care setting through the use of a novel personalized medicine genomic test: results from a prospective registry study. Am J Med Qual. Jul-Aug 2015;30(4):345-352. PMID 24798176.

25. Ladapo JA, Herman L, Weiner BH, et al. Use of a blood test incorporating age, sex, and gene expression influences medical decision-making in the evaluation of women presenting with symptoms suggestive of obstructive coronary artery disease: summary results from two ambulatory care studies in primary care. Menopause. Nov 2015;22(11):1224-1230. PMID 25828395.

26. Ashley EA, Hershberger RE, Caleshu C, et al. Genetics and cardiovascular disease: a policy statement from the American Heart Association. Circulation. Jul 3 2012;126(1):142-157. PMID 22645291.

27. Secretary's Advisory Committee on Genetic Testing, National Institutes of Health. Enhancing the oversight of genetic tests: recommendations of the SACGT. Betheday, MD: NIH; 2000 July.

28. Musunuru K, Ingelsson E, Fornage M, et al. The expressed genome in cardiovascular diseases and stroke: refinement, diagnosis, and prediction: a scientific statement from the American Heart Association. Circ Cardiovasc Genet. Aug 2017;10(4). PMID 28760750.

29. Centers for Medicare & Medicaid Services. Local Coverage Determination (LCD): CORUS CAD Test (L36713). 2016; https://www.cms.gov/medicare-coverage-database/details/lcd- details.aspx?LCDId=36713&ver=4&CoverageSelection=Both&ArticleType=All&PolicyType=Final&s=All&KeyWor d=corus&KeyWordLookUp=Title&KeyWordSearchType=And&bc=gAAAACAAAAAA&. Accessed January 16, 2018.

Codes:
(The list of codes is not intended to be all-inclusive and is included below for informational purposes only. Inclusion or exclusion of a procedure, diagnosis, drug or device code(s) does not constitute or imply authorization, certification, approval, offer of coverage or guarantee of payment.)

CPT*

81493
81599