Clearing a path for new
  cardiac markers

February 2005
Feature Story

Although the troponins remain the cornerstone of risk stratification in acute coronary syndrome patients, investigators continue to search for cardiac markers that can help identify a select group. That is the 20 percent of cardiac troponin-negative unstable angina/non-ST segment elevation ACS patients who will visit U.S. emergency departments with chest pain and go on to experience a cardiac event within six months of being diagnosed. Under investigation—alone and in conjunction with one another—are a plethora of biomarkers involved in the atherothrombotic process, including markers of hemostasis, platelet function, hemodynamic stress, inflammation, ischemia, and necrosis.

A multi-marker strategy for risk stratification of ACS patients is not recommended by cardiac marker guidelines, such as those being drafted by the National Academy of Clinical Biochemistry, or NACB. However, the academy’s draft laboratory medicine practice guideline on the use of biomarkers in ACS and heart failure does allow for the possibility that multiple-marker strategies could in the future yield better information for diagnosis and risk stratification than the measurement of a single cardiac marker.

The new draft practice guideline, posted on the NACB Web site, www.nacb.org, addresses the use of biomarkers across the spectrum of ACS and heart failure, discussing analytical issues, diagnosis, risk stratification, and more. The draft guideline says that for risk stratification of ACS patients, cardiac troponin is the preferred marker and, if available, should be measured in all patients with suspected ACS. In patients with a clinical syndrome consistent with ACS, the draft guideline says, a maximal concentration exceeding the 99th percentile of values for a reference control group (with acceptable precision) should be considered indicative of increased risk of death or myocardial infarction. (Using the 99th percentile implies that there will be at least a one percent value above the 99th percentile producing a false-positive result.)

This is the only class I recommendation made in the document, which divides its advice into class I, class IIa, class IIb, or class III, based on the recommendation classes of the American College of Cardiology and American Heart Association. Class I represents conditions or recommendations for which there is general agreement or evidence, or both, that a procedure is useful or effective, while class IIa and IIb recommendations represent conditions in which opinions diverge, with class IIa carrying a weight of evidence or opinion in favor of the usefulness or effectiveness of a procedure, and class IIb carrying a weight of evidence in which the usefulness or efficacy of the procedure is less well established. Class III signifies conditions for which a procedure is not useful or effective and, in some cases, may be harmful.

In its recommendations regarding whether a multi-marker approach should be used for risk stratification, the draft guideline says there is less evidence for measuring hs-CRP, BNP, and NT-proBNP in addition to a cardiac troponin for assessing risk in patients with a clinical syndrome consistent with ACS than using cardiac troponin alone, but the document gives the recommendation a class IIa nod. Measurement of markers of myocardial ischemia in addition to cardiac troponin and electrocardiogram to aid in the short-term risk stratification of patients with suspected ACS, and in excluding ACS in patients with low clinical probability of ischemia, carries less evidence, and was categorized as class IIb. Using a multi-marker strategy that includes measuring two or more path o bio logi cal ly diverse biomarkers in addition to cardiac troponin for risk stratification is also now a class IIb recommendation.

Yet the evidence base for adding markers to troponin for risk assessment continues to grow. "The natriuretic peptides appear to be strong markers for risk stratification in known acute coronary syndrome patients and, given the evidence, might have even merit ed a class I recommendation. However, the committee felt that the benefits of therapy based on elevated BNP or NT-proBNP levels in the ACS population remain uncertain, and therefore we felt that a class IIa recommendation was appropriate," explains Robert Christenson, PhD, professor of pathology and director of rapid response laboratories at the University of Maryland Medical Systems, Baltimore, and chair of the group writing the practice guideline on the use of biomarkers in ACS and heart failure.

"Generally speaking, in the ED right now, most people do not measure BNP or NT-proBNP for chest pain unless there is associated atrial fibrillation or symptoms of shortness of breath," says Alan Maisel, MD, professor of medicine at the University of California, San Diego, and director of the critical care unit in the heart failure program at the San Diego VA Healthcare System. "However, before a patient presents with acute coronary syndrome, when you have suspicion for underlying disease but plaque has not yet been ruptured, BNP or NT-proBNP could potentially be part of the evaluation."

Recent data from the Platelet Receptor Inhibition in Ischemic Syndrome Management (PRISM) trial published in the November issue of Circulation (2004;110:3206-3212) showed that when measured at baseline and at 48 and 72 hours in 1,791 patients with non-ST-segment-elevation ACS, NT-proBNP levels greater than 250 ng/L were associated with higher event rates at 30 days. In troponin T-negative patients, NT-proBNP identified a subgroup of high-risk patients whose risk was comparable to that of troponin T-positive patients. In addition, the researchers found that clinical stabilization without refractory ischemia was associated with a rapid and significant decline in NT-proBNP levels, while the lack of a rapid decline in NT-proBNP levels in patients with baseline levels that were 250 ng/L or higher was linked to an adverse short-term prognosis. Rising NT-proBNP levels were also associated with an adverse 30-day prognosis.

A study underway at the SanDiego VA is demonstrating similar findings. "In our experience, if we see BNP levels rise over a six- to 12-hour period, we know that something is going wrong with the patient, even if the troponins are negative," Dr. Maisel says. "In the first couple of hundred patients that we’ve looked at in the study we are now doing, in the setting of acute coronary syndrome, about 34 percent ended up developing clinical CHF [congestive heart failure] in the hospital, and virtually every one of those patients had BNP concentrations that rose in the first 12 hours or so."

Biosite, the first company to bring BNP to market, is examining how the assay can be used as part of a multi-marker strategy, but it has taken a slightly different approach than most of the researchers who are looking at such strategies. The company has been evaluating an index response panel, called the CardioProfiler, which uses BNP in conjunction with traditional cardiac markers for ACS diagnosis and prognosis.

Using an algorithm derived from thousands of values from ACS and noncardiac chest pain patients, the CardioProfiler will provide a single index value after analyzing a blood sample using BNP, CK-MB, troponin I and complexes, and myoglobin assays. "What we have found," says Ken Buechler, PhD, president and chief scientific officer at Biosite, "is that if we use a multi-marker index, which is an algorithm that uses each one of the concentrations of the analytes involved and calculates a single value, the multi-marker index approach does a much better job at predicting early MI than individual markers alone." Dr. Maisel, who is doing research with the CardioProfiler, adds, "Within two hours of the onset of chest pain, data suggests that the index could diagnose MI with about 94 percent sensitivity, whereas troponin alone is only about 30 percent sensitive within two hours of the onset of chest pain."

In addition to the natriuretic peptides, the National Academy of Clinical Biochemistry includes high-sensitivity C-reactive protein as a potential additional marker for risk stratification under a class IIa recommendation, but Dr. Christenson has concerns about CRP’s lack of specificity when used in the general ED population. "If you look at an isolated population of ACS patients, hs-CRP will be predictive. However, when you start to look at the entire population in which ACS is suspected, the test is nonspecific. In addition, it’s not like D-dimer or IMA [ischemia-modified albumin], where a negative value predicts low risk of disease. The data strongly indicate that specificity is going to be a problem with general use of hs-CRP," he says.

This doesn’t mean, however, that CRP would not add value to a multi-marker panel for risk-stratifying ACS patients. "If you take troponin, CRP, and BNP and use them in a multi-marker strategy," Dr. Maisel says, "each one will give individual additional prognostic risk information."

While BNP, NT-proBNP, the troponins, and hs-CRP have a growing body of literature that can be used to determine whether they are ready for use in a multi-marker risk-stratification strategy, there are other tests that show promise but need additional studies. Myeloperoxidase, or MPO, a biomarker that has been reported to be useful in identifying inflammation in the walls of coronary arteries and therefore risk for heart disease or MI, has seemingly found a number of champions and may be the next new cardiac marker to gain Food and Drug Administration approval. Prognostix Inc., of Cleveland, has submitted an MPO test to the FDA for review, while Abbott Diagnostics, Dade Behring, and Biosite all have MPO assays in development.

Stanley Hazen, MD, PhD, section head of preventive cardiology and cardiac rehabilitation at the Cleveland Clinic Foundation, and his colleagues first published data on the prognostic value of MPO in 2003 (N Engl J Med. 2003; 349:1595-604), which showed that the marker may be particularly beneficial in chest pain patients who exhibit low initial troponin levels. "We looked at initial MPO levels in more than 600 patients immediately upon presentation to the emergency room with the complaint of chest pain," Dr. Hazen says. "By using MPO testing in addition to current laboratory-based risk assessments, we increased our ability to identify subjects who experienced a major adverse cardiac event over the next 30 days to six months from 54 percent to 85 percent of the time." An even larger study performed in Europe (Circulation. 2003;108:e9034-9035) has confirmed these results.

"Myeloperoxidase is strongly and independently correlated with measures of endothelial dysfunction in humans, which is consistent with it playing a deleterious role by consuming nitric oxide in the artery wall," explains Dr. Hazen. "MPO is markedly enriched in culprit lesions in subjects with sudden cardiac death at sites of fissure or rupture of plaque and at superficial erosions. Erosions as a site for intracoronary thrombus formation occur to a greater extent in women and smokers, where something makes endothelial cells on the surface of vulnerable plaque die and slough off, exposing a prothrombotic surface. MPO-induced apoptosis of coronary artery endothelial cells appears to play a role in that."

In recent published studies by Peter Libby, MD, PhD, and colleagues from Brigham and Women’s Hospital, Harvard Medical School (Arterioscler Thromb Vasc Biol. 2004:24;1309-1314), very low levels of MPO such as those found within vulnerable atherosclerotic plaque, resulted in endothelial cell apoptosis and increased tissue-factor expression. Tissue factor is seen as key in the prothrombotic phenotype. Based on what researchers are seeing, Dr. Hazen thinks MPO testing may have two potential uses in ACS subjects. "One will be in identifying higher-risk groups, meaning that if your MPO level is above a certain cutoff, it denotes that you have a substantial increase in risk for having a heart attack, needing bypass surgery, or dying in the near-term," he says. The other scenario is to look at what low levels of MPO mean. "We are seeing a definite population that has a low MPO concentration and an initial negative troponin, and the negative predictive value for cardiac events is markedly improved in this population versus those in whom only the troponins are measured," Dr. Hazen says, noting that the addition of a low MPO value to a negative troponin cuts in half the number of missed cases of cardiac events over the subsequent six-month interval after a patient is diagnosed with unstable angina/non-ST-segment elevation ACS.

Says Biosite’s Dr. Buechler: "It’s clearly been demonstrated in the literature that MPO is an important prognostic marker. It appears to predict bad outcomes such as death or MI, and we are studying that in our discovery process right now." Biosite licensed the MPO technology from the Cleveland Clinic Foundation and its affiliate, Prognostix, about a year ago.

Abbott Diagnostics and Dade Behring recently announced that they have agreements with the Cleveland Clinic Foundation and Prognostix to develop MPO assays and are looking to bolster their cardiac marker offerings with other emerging cardiac markers as well. "I think clinical research is focused right now on plaque destabilization and rupture, and consequently, we’re learning more about the role of emerging markers in these processes," says John Shaw, PhD, vice president of technology assessment at Dade Behring. The markers they are pursuing for secondary risk stratification and specifically for plaque vulnerability are MPO, placental growth factor, and sCD40L.

Though published studies examining sCD40L’s usefulness in risk stratification have been few, those that have made it into the peer-reviewed literature indicate it’s a promising marker, and it’s on the radar screen of many cardiac marker researchers. "I really believe that sCD40L is going to be a big player in risk stratification in the future," Dr. Christenson says, "mainly because it gives you more information about platelets, which many of the other emerging markers do not."

A recent study performed by Christopher Heeschen, MD, and his colleagues at the University of Frankfurt in Germany, who have performed a number of multi-marker studies in the risk stratification arena, demonstrated the potential value of sCD40L when used in conjunction with yet another emerging marker for risk stratification, pregnancy-associated plasma protein-A, or PAPP-A, a zinc-binding matrix metalloproteinase that appears to be expressed abundantly in eroded and ruptured plaques in ACS patients.

The study, published in the Journal of the American College of Cardiology (2005;45:229-237), examined the predictive value of PAPP-A in 547 patients with angiographically validated ACS and in an emergency department population of 644 patients with acute chest pain, in addition to troponin T, vascular endothelial growth factor (VEGF), hs-CRP, interleukin-10, and sCD40L. Patients were followed for the occurrence of death or myocardial infarction. In patients with ACS, elevated PAPP-A levels (>12.6 mIU/L) indicated an increased risk. When the analysis was restricted to TnT-negative patients, PAPP-A still identified a subgroup of high-risk patients. In a multivariable model, PAPP-A, sCD40L, IL-10, and VEGF were independent predictors. Patients negative for TnT, sCD40L, and PAPP-A were at very low cardiac risk.

Placental growth factor is just one of a variety of pro-inflammatory markers, including sCD40L, Il-6, PAPP-A, MPO, and monocyte chemoattractant protein-1, as well as anti-inflammatory markers, including IL-10 and activin A, that have been suggested to provide relevant prognostic information in ACS patients. "In the future," says Dade Behring’s Dr. Shaw, "we see risk stratification going toward a panel of markers of inflammation combined with troponin and NT-proBNP, and that’s why we’re looking at markers such as placental growth factor, MPO, and sCD40L."

Abbott is investigating whole-blood choline as a potential new cardiac marker. Few studies of whole-blood choline’s use in risk assessment have been published, but an article in the American Journal of Cardiology (2003;91:1060-1067) demonstrated that the biomarker could be useful in risk stratification based on a prospective study of 327 patients with suspected ACS that focused on the analysis of troponin-negative patients. In the study, choline was measured using high-pressure liquid chromatography mass spectrometry in whole blood, and patients were followed for 30 days. In patients with negative troponin I test results at admission, choline was a predictor of cardiac death and nonfatal cardiac arrest, life-threatening arrhythmias, heart failure, and coronary angioplasty. In multivariate analysis of troponin-negative patients, choline was the strongest predictor of cardiac death or arrest.

Jessie Shih, PhD, associate research fellow for cardiovascular at Abbott Diagnostics, says Abbott is looking at other new cardiac markers as well. "Cardiac is a growing area for us, and we are actively investigating new markers as they are identified in press or as people contact us," she says. Adds Robert Olson, Abbott’s scientific affairs director for cardiac, "While troponin is very good, there is still a need to find an early ischemia marker, and that’s a major focus for us as well."

The only marker for ischemia on the market is ischemia-modified albumin from Ischemia Technologies, which Dr. Christenson says is another marker that can be explored for risk stratification. Its most efficacious use at the moment appears to be in ruling out ACS. "There’s compelling data that IMA’s negative predictive value is very good, and that the test will probably be clinically useful. The literature is still developing, however, on how exactly to interpret a positive test," Dr. Christenson says.

For risk stratification, Drs. Christenson and Maisel believe that the future is in multi-marker panels. "They’ll have to be markers that reflect the spectrum of the physiology of acute coronary syndrome, and we’ll have to find the best mix," Dr. Christenson says. In the meantime, he adds, it’s important to continue to standardize troponin I assays and to continue to move toward troponin I tests that can dem on strate acceptable precision at the 99th percentile of values for a reference control group.

The American Association of Clinical Chemistry’s troponin I standardization subcommittee is collecting data from a third round-robin event that will aid standardization. "I would expect in the next year or so that there will be a big push to standardize troponin I assays," Dr. Christenson predicts.

As for the status of the NACB’s guideline on markers of ACS and heart failure, a second draft is scheduled to be placed on the academy’s Web site early this spring, and the guideline probably will be finalized this summer. To date, at least a dozen groups have signed on to the guideline, with 25 or so others being represented by the National Heart Attack Alert Program.

What does the immediate future hold for the multitude of emerging risk-stratification markers for ACS? "The data are out there for using multiple markers, and I think we are seeing different utilities for each of the markers," says Abbott’s Dr. Shih. "Some of them may be more useful for shorter-term risk stratification, and some are more useful further out. But what we need to do is to get more robust assays out there and understand their clinical utility better with larger multi-site studies."

Yes, the plethora of emerging markers for ACS risk stratification is exciting, but the true test will come, Dr. Christenson says, when clinicians begin using them in the clinical setting. "A multi-marker strategy is definitely coming. But no matter what stage we get to," he cautions, "there will still be an art to using cardiac markers for risk stratification, because the markers will always have to be interpreted in the clinical context of the patient."

Sue Parham is a writer in Edgewater, Md.