February 2009
Feature Story

Karen Lusky

Already playing first chair among tests used to detect acute myocardial infarction, cardiac troponin may have other instrumental roles in its future. New, more sensitive troponin assays on the near horizon hold the promise of improving diagnosis in acute coronary syndrome, as well as risk stratifying patients in numerous settings, including potentially primary care.

In fact, when asked about new cardiac biomarkers in the works to augment troponin, Robert Christenson, PhD, has begun to answer: “troponin.”

Using the current generation of troponin assays to identify risk, “we are probably only looking at the tip of the iceberg above the water line,” says Dr. Christenson, director of clinical chemistry laboratories and professor of pathology and medical and research technology at the University of Maryland. Below that water line, where the next generation of troponin assays will allow people to look, “there’s a lot of danger for adverse outcomes.”

Contemporary troponin assays in use pick up only about 10 percent of troponin values in the normal range, “and everything else is below the limit of detection,” says Fred Apple, PhD, medical director of clinical laboratories at Henne­pin County Medical Center and professor of laboratory medicine and pathology at the University of Minnesota School of Medicine. Yet companies developing the next generation of assays—Nano­sphere, Singulex, Beckman, Roche, Abbott, Ortho-Clinical Diagnostics, and others—are producing data showing a “nice bell-shaped curve for their normal reference populations,” rather than just the “tail end of the curve,” he says.

Emerging research shows that adults have troponin in their blood that comes from remodeling of heart muscle in a healthy state. The amounts appear to increase with age and conditions that stress or injure the heart.

Nanosphere, which expects to have a next-generation troponin assay on the market by mid-year on its Verigene platform, has found normal people ages 18 to 65 have troponin levels of 1 to 1.5 pg/mL, says Bill Moffitt, CEO of the company in Northbrook, Ill. Yet, even within that normal range, the upper quartile or top 25 percent contains a “preponderance of smokers.”

Putting the picograms in perspective, Alan Wu, PhD, points out that the best troponin assays in use today have a 99th percentile cutoff of 40 pg/mL for the normal reference range, and can measure as low as 10 pg/mL. By contrast, the next generation has a 99th percentile four times lower than that—and can detect troponin 10 times lower, says Dr. Wu, professor of laboratory medicine at the University of California at San Francisco.

The new assays, says Harvard cardiologist Marc S. Sabatine, MD, MPH, associate physician at Brigham and Women’s Hospital, will represent a “sea change in terms of how we think about troponin,” making it like the majority of tests, such as creatinine, sodium, or potassium, where everyone has a quantifiable level.

“The next step will be to decide whether an elevation means someone has a disease, and if so, what actions to take,” which will require additional studies and clinical trials, some of which are already underway, Dr. Sabatine says.

If all goes well, the next wave of troponin assays could end up identifying more cases of acute MI in the emergent care setting, experts predict. That’s what appears to have occurred even with the current generation of assays, which are more sensitive than their predecessors.

Dr. Apple notes that the global task force of cardiology groups used the current generation of assays, which are very good, to develop the universal definition of AMI. And that definition, he says, includes using troponin as the preferred marker with the 99th percentile as the cutoff for normal, above which indicates myocardial injury.

Using that definition over the past several years, Hennepin County Medical Center has seen its rate of MIs rise from eight percent at the outset to 12 to 13 percent. With the lower cutoff, some publications are reporting that the number of diagnosed MIs has climbed almost 50 to 75 percent, Dr. Apple says.

With the next round of even more sensitive assays, the numbers will rise even more. “There’s no question,” says Allan S. Jaffe, MD, professor of medicine and a consultant in cardiology and laboratory medicine at the Mayo Clinic in Rochester, Minn., that physicians will see a “marked” increase in the number of elevated troponins. “It’s quite clear that people with chronic heart disease are going to be found to have troponin elevations,” he says, pointing to one reason for the certain increase.

On the downside, that means already beleaguered emergency departments would have to take the time to do more serial troponin testing to make the appropriate diagnosis in patients with an initial troponin elevation.

With troponin, says Dr. Christenson, “three things are bad: troponin going up, troponin going down, and troponin being elevated and staying constant. Going up is bad because it’s likely acute injury,” he notes. Ditto for going down. “And staying elevated is something you’d expect to see in heart failure or renal failure that would confer higher risk. It shows that the person has an increased turnover at a constant rate in his or her heart cells.”

The good news is that the newer generation of assays may allow physicians to detect a change in troponin indicating acute MI faster than they can with contemporary troponin tests.

Dr. Apple says he and his colleagues have a paper in press at Clinical Chemistry showing that when using current assays, a less than 30 percent delta rise in troponin over six hours has a 91 percent specificity in ruling out acute MI compared with 70 percent based on a single concentration at presentation.

“The hope is that the new high-sensitive assays would be able to identify that delta over one to two hours,” he says. That’s because as the 99th percentile reference (normal) cutoff value gets lower and lower, Dr. Apple says, “you can see smaller, real troponin changes at the low concentration end.”

Also, says Dr. Christenson, while the best current troponin assays in use today have values that are “tight” (that is, they have a low coefficient of variation) around the true value at the 99th percentile, as you go lower, there’s “more and more noise in the measurement.” In contrast, with the hypersensitive assays in development, the precision does not deteriorate as you go lower and lower.

The amount of change in troponin indicative of an acute MI may be different using the next generation of assays, however. In an AACC audioconference on troponin, Dr. Wu pointed to research he and colleagues conducted using such an assay to identify the change in troponin levels that exceed biological variability. “It turns out to be about 50 percent over several hours,” he says.

“Others have suggested 30 percent,” adds Dr. Jaffe, who was a presenter with Dr. Wu in the audioconference. “But we should eventually be able to come to terms about what sort of percent change will define when an acute event is likely.” That doesn’t mean there won’t be exceptions, he says, or that elevated troponin in some patients is not an additional risk marker where physicians will want to admit those patients to the hospital. “But it will help to eventually refine how we ought to use troponin to identify who is at risk.”

One question in play is whether increasingly sensitive assays will turn troponin into a de facto marker for cardiac ischemia.

If the answer is yes, “that would be wonderful,” says Cleveland Clinic emer­gency medicine physician W. Frank Peacock, MD, “because we don’t have an ischemia test right now. And we can’t exclude an acute coronary syndrome with a negative troponin. Currently, if you have a negative troponin, you haven’t excluded anything yet.”

Luis LaSalvia, MD, head of integrated diagnostics and market development for Siemens Healthcare Diagnostics, Tarrytown, NY, which is evaluating the best path forward with next-generation troponin testing, notes that the U.S. spends an es­timated $8 billion to $13 billion per year in managing chest pain patients in the ED. “And approximately up to 80 percent of these patients don’t have ACS.”

The jury is, of course, still way out on whether troponin has a role as an ischemia marker in detecting ACS without myocardial necrosis. But Nanosphere CEO Moffitt points to an intriguing, decade-old study in Europe in which researchers were able to measure troponin in patients undergoing brief periods of cardiac ischemia followed by reperfusion during beating heart coronary artery revascularization (Su­lei­man MS, et al. Clin Chim Acta. 1999 Jun 15; 284[1]:25–30).

The researchers found that “although significant amounts of troponin I are released into the coronary sinus, only high concentration can be translated into measurable values in the systemic circulation. Therefore, the issue becomes one of sensitivity,” Moffitt says. And higher sensitivity is what the next generation of troponin assays brings to the diagnostic table.

Harvard’s Dr. Sabatine believes one can see a release of troponin in someone with reversible cardiac ischemia. In a study at Brigham, he and colleagues were able to correlate circulating troponin levels in 120 study subjects to how they fared on cardiac stress testing performed with nuclear perfusion imaging (Sabatine MS, et al. Eur Heart J. 2009 Jan; 30[2]:162–169. Epub 2008 Nov).

The researchers used Singulex’s highly sensitive troponin assay, which uses technology based on single molecule counting. The researchers also tested patients undergoing stress tests using a current-generation troponin assay, which showed the subjects had “essentially flat” troponin levels, because the assay didn’t allow you to “probe down to the subtle elevations,” Dr. Sabatine says. “However, when one uses a far more sensitive assay like Singulex’s, you can actually appreciate changes in circulating troponin levels related to the amount of ischemia on stress testing.”

The study participants underwent stress testing following the usual protocols, which require monitoring of the patient’s symptoms and EKG throughout the procedure. Following the gold standard, the participants also received myocardial perfusion imaging before and at peak exercise to look for perfusion problems. Blood samples for troponin testing were obtained from patients before their stress test, immediately after, and two and four hours post-test, respectively, according to the study report. Thus, the patient served as his or her own control, Dr. Sabatine says.

Using the Singulex assay, the researchers found that even participants with a “stone cold normal stress test” had quantifiable troponin levels in single-spot measurements before the test, though the level didn’t rise, Dr. Sabatine says. However, participants with a mildly positive stress test had an intermediate rise in troponin, and those with a “very positive test” showed an even greater increase.

The researchers did not see a strong correlation between the amount of ischemia, EKG findings, and patient symptoms. “We know that symptoms and EKG aren’t sufficient to diagnose ischemia,” Dr. Sabatine explains, “which is why we do stress testing combined with perfusion imaging.”

“And EKG changes or angina during stress testing aren’t very specific”—for example, one can see EKG changes with tachycardia that do not reflect true ischemia. And someone can have pain symptoms caused by movement of the chest wall during the stress test.

In explaining the study findings showing troponin elevations associated with positive stress tests, Dr. Sabatine says it could be that although myocytes survive ischemia, it compromises the cell membrane, allowing troponin to leak out. “We used to think that troponin, which is part of the skeleton of the cell, would only be detected in the circulation if the cell breaks down.” But now it’s known that some troponin floats in the cytoplasm unbound to the “mechanical machinery,” he says. “That could be what’s being released quickly.”

“Whatever the source of the release, we know the insult is related to the amount of transient ischemia.” But it’s possible a person could have a tiny infarct during the stress testing, leading to troponin release, Dr. Sabatine says. He views that as unlikely given that the physicians who run the stress test stop it when the patient becomes symptomatic. Thus, the patient undergoing the testing may have angina for a “minute or two at the most,” whereas typically people have to experience 15 to 20 minutes of angina before an actual infarction is seen, he says.

“It could end up being a semantical or philosophical issue in terms of whether a few myocytes have died or are transiently injured” as the cause of the elevated troponin levels. Yet either way, Dr. Sabatine adds, “from a clinician’s standpoint, the very interesting finding is that you can now biochemically identify myocardial ischemia.”

Another study using Roche Diagnostics’ next-generation troponin assay did not produce findings similar to those of the Brigham study, he says. “But there are different thresholds of sensitivity [between the two tests], and the Singulex assay measures troponin I, and the Roche assay, troponin T. So it’s not entirely apples to apples. You also look at how many subjects were in the study and how well the outcomes were classified.”

Mayo’s Dr. Jaffe doesn’t think the issue of the impact of the different assays on study findings can be “adjudicated” at this point. He does say, however, that if troponin does what he thinks it will eventually be able to do—“that is, augment the stress test—it will be useful.”

Prevention is generally the most cost-effective medicine in human and monetary terms. If the new, more sensitive troponin assays in the works pan out, some predict the marker could become part of an annual physical to detect patients at risk for negative outcomes who appear healthy based on today’s routine evaluations.

In fact, troponin’s star as a risk stratifier has already begun to rise based on studies using the current generation of assays.

As one example, in a study conducted in Rancho Bernardo, Calif., of 957 older, apparently healthy community-dwelling adults, researchers found that subjects with detectable troponin T (=0.01 ng/mL) using an existing assay had a higher risk of death from cardiovascular and all causes (Daniels LB, et al. J Am Coll Cardiol. 2008 Aug 5;52[6]:450–459). In the study abstract, the authors write: “Those with both TnT and NT-proBNP elevations are at even higher risk, and the increased risk persists for years.”

Julian Braz, PhD, manager of clinical and scientific affairs at Roche Diagnostics, which declines to discuss assays in development, says that by investing in appropriate “due diligence” in terms of what to do with the results of more sensitive troponin assays, “we could come up with a way of risk stratifying the normal population.”

“In theory, a cardiac panel approach to identify a patient’s level of risk might prove the most valuable. For example, if a patient is positive for troponin and another cardiac marker such as NT-proBNP, CRP, or a novel marker, then that patient would be placed in an increased risk category. Or if a patient has several cardiac markers elevated in combination, then that patient would be placed in a high-risk category,” Dr. Braz says. “The nice part about using a panel for risk is that each of the cardiac markers acts independently of one another.”

Nanosphere is conducting research to look not only at the ability of its assay to diagnose acute MI earlier, but also to identify meaningful elevations of troponin “against the background of other comorbidities,” including diabetes, kidney disease, hypertension, and congestive heart failure, Moffitt says. The prospective, multicenter, blinded clinical trial, which just started, will run in 12 to 18 sites in the United States and Europe. (The names of the principal investigators will be released later this year, Moffitt notes.)

Mayo’s Dr. Jaffe predicts that eventually physicians will use higher-sensitivity troponin testing as a screening tool to detect someone, for example, with mild hypertension who is beginning to develop cardiac hypertrophy long before it is clinically evident.

He sees other uses for the more sensitive troponin assays. Developing drugs to identify cardiotoxic pharmaceuticals before they’re out of the gate is one possibility.

John A. Todd, PhD, vice president of research and development for Singulex, Alameda, Calif., says the next-generation assays might be used as a biomarker for cardiotoxicity caused by, for example, anthracyclines or Herceptin. “In this context,” he says, “the idea under study” is that you could use a cardiotoxic drug more aggressively in someone who did not show an increase in troponin as opposed to someone who did.

Troponin is also “wildly prognostic” both short term and long term in critically ill patients, showing cardiac injury that needs to be managed, Dr. Jaffe says. But at present, how best to do that is unclear.

Much work needs to be done to get those new assays ready for prime time, he adds. “Those of us invested in biomarkers can’t run so fast with [the next generation of troponin testing] that we outstrip the ability of the clinical community to use that data.”

The more sensitive assays themselves will cause people to ask more questions about how they can be used, Nanosphere’s Moffitt predicts. In his view, “we have an old friend in the troponin biomarker turned into a new tool” that will offer new opportunities to improve medical care.