Anne Paxton
October 2020—When Elizabeth Selvin, PhD, MPH, of Johns Hopkins Bloomberg School of Public Health, began her studies of high-sensitivity cardiac troponin assays, they had not yet been approved in the U.S., as they are now, for use in diagnosing myocardial infarction. But some of her studies and those of Amy K. Saenger, PhD, DABCC, medical director of clinical laboratories and director of clinical chemistry at Hennepin County Medical Center in Minneapolis, take high-sensitivity cardiac troponin in a new direction by exploring its potential use as an aid in monitoring cardiovascular risk in the general population.
The new high-sensitivity assays detect circulating concentrations of high-sensitivity cardiac troponins I and T in 50 to 100 percent of the general ambulatory population and may indicate subclinical myocardial injury if concentrations are increased above the sex-specific reference interval (99th percentile).
How far can troponin’s uses as a biomarker now be expanded beyond its essential role in diagnosing MI in the acute care setting? Two new Diabetes Care studies conducted by Drs. Selvin and Saenger and others report on the potential usefulness of high-sensitivity cardiac troponin as a biomarker for improving mortality risk stratification in adults with diabetes and for predicting cardiovascular disease and major adverse cardiovascular events in childhood-onset type one diabetes.
In the first study, Dr. Selvin, professor of epidemiology and principal investigator of several research grants, and her colleagues investigate how well high-sensitivity cardiac troponin assays perform at reflecting comorbidity burden and improving mortality risk stratification in older adults with diabetes (Tang O, et al. Diabetes Care. 2020;43[6]:1200–1208). Such a potential use would be valuable because, as the population ages, cardiovascular risk reduction in older adults continues to be of central importance but the heterogeneity in the health of older adults presents challenges in evaluating their risk, the study notes.
Dr. Selvin and coauthors hypothesized that hs-cTnI or hs-cTnT may either replace or supplement comorbidity burden in the prediction of mortality risk in older adults with diabetes. The results of their study confirm that both might well serve such a role.
Dr. Selvin
“I’m particularly interested in understanding use of troponin for risk stratification in certain populations, and in particular diabetes, because people with diabetes are in a very high-risk group for cardiovascular complications,” Dr. Selvin says. “New classes of diabetes medications, including SGLT2 inhibitors, have cardiovascular benefits. The question is who should be first in line to receive these medications, which should be targeted to people in highest cardiovascular risk groups.”
“How do we optimize who should get these more expensive and cardioprotective medications?”
One of the interests of Dr. Selvin and her colleagues was determining whether troponin could be an aid in understanding who is at the highest risk of disease. “We have demonstrated that in people with diabetes, individuals who have elevated high-sensitivity cardiac troponin concentrations are at the highest risk of death and cardiovascular outcomes, particularly heart failure,” she says.
Drawing on the data from the longitudinal Atherosclerosis Risk in Communities (ARIC) study, Dr. Selvin and colleagues included 1,835 older adults (ages 67 to 89) with diabetes and assessed these subjects’ comorbidities at visit five (2011–2013). High comorbidity burden was defined as three or more comorbidities (cancer history, arthritis, hypertension, emphysema, incontinence, etc.). The research team measured concentrations of hs-cTnI in plasma obtained from a biobank using the Abbott Architect Stat hs-TnI assay and hs-cTnT concentrations using the Roche Elecsys Troponin T Gen 5 Stat assay in all the participants.
Both hs-cTnI and hs-cTnT reflected similar comorbidity profiles. There were 418 deaths that occurred over a median follow-up period of 6.2 years. In the study, higher concentrations of hs-cTnI and hs-cTnT were strongly associated with higher all-cause mortality risk. Specifically, the four-year cumulative mortality among those with troponin values less than the 40th percentile in the ARIC population at visit five was 5.4 percent (hs-cTnI) and 5.4 percent (hs-cTnT). This is compared with 27.8 percent (hs-cTnI) and 29.4 percent (hs-cTnT) among those with hs-cTnI or hs-cTnT greater than or equal to the 85th percentile.
A high comorbidity burden was associated with elevated mortality risk compared with those with a low burden, but including either cardiac troponin in risk equations improved prediction even above and beyond comorbidity burden. Inclusion of the two high-sensitivity cardiac troponin tests predominantly down-classified those who did not die during follow-up.
When the researchers assessed the use of troponin to risk stratify beyond health status levels defined by a high comorbidity burden or dementia or frailty, those with high hs-cTnI or hs-cTnT and the presence of a high comorbidity burden or frailty or dementia were at the highest risk of mortality. “Even among those with a low comorbidity burden, no dementia, and no frailty,” they write, “high hs-cTnI or hs-cTnT was independently associated with a higher mortality risk.”
The results led the coauthors to conclude that there may be a role for measuring hs-cTnI or hs-cTnT to supplement the overall clinical impression of the patient, thus refining comorbidity burden to improve individualization of cardiovascular risk reduction strategies in older age.
In their commentary published in the same issue, Ian J. Neeland, MD, and James A. de Lemos, MD, of the University of Texas Southwestern Medical Center, say “the cohort is reflective of the contemporary older adult population with diabetes and study findings are likely generalizable to the aging U.S. population.”
The ARIC study, Dr. Selvin notes, has been ongoing for decades and is one of the biggest cohort studies of cardiovascular disease in the United States. “We’ve been following these patients for over 30 years. Originally we recruited almost 16,000 people in the late 1980s when they were middle-aged. We’re seeing all the participants annually; they are all over 75 now.” The average age was about 75 at the time she and colleagues did their study.
“Traditional risk factors do a very good job with cardiovascular risk prediction, so it’s hard to get the prediction statistics to budge,” Dr. Selvin says. With this study, the researchers wished to know whether these troponin assays are independently able to improve prediction for cardiovascular events, particularly heart failure and mortality.
“What our research has showed,” she says, “is that a single elevated troponin can help us identify people who are at the highest risk of poor outcomes. That was an exciting, important finding.”
The researchers categorized subjects in the study according to the number of different comorbidities in their profile. In theory, “Physicians should holistically think of their patients and think about the morbidities and the effects they have,” Dr. Selvin says. “In the American Diabetes Association guidelines, it is recommended that treatment decisions in older adults should depend on number of comorbidities.” But it’s not clear how clinicians should operationalize the guidelines, she adds.
“We looked at how these comorbidities are related to outcomes. For the most part, what we showed is that if you have high comorbidity, you’re likely to have an elevated high-sensitivity troponin, but that the troponin itself is just much more discriminatory. Using an objective biomarker such as troponin that is easily measured may be an approach to risk stratification that overcomes some of the complexity of trying to think about all the wide variety of comorbidities in the context of patients with diabetes.”
A criticism of the ADA guidelines for older adults, she says, is the wide differences among comorbidities, which can range from cancer, dementia, or hypertension to frailty or incontinence. “Metastatic cancer is very different from osteoarthritis or urinary incontinence, and treatment decisions should not treat them equally. Troponin is an objective biomarker that reflects undetected, chronic damage to the heart. It is an incredibly potent marker. We’re identifying people who have heart damage and are much more likely to experience heart failure or a cardiovascular event and die. And we think troponin could be a useful way to monitor risk and identify individuals in clinical practice who should be targeted for more intensive cardiovascular intervention.”
The robust association of high-sensitivity cardiac troponin with cardiovascular and even microvascular outcomes is now well established, Dr. Selvin says. “The protein is a highly specific biomarker of damage to the heart, but many conditions can cause elevations in troponin. That’s what makes it so useful as a global measure of end organ damage. We’re seeing the underlying chronic cardiac damage when we measure a troponin in the population.”
One important implication of the study is that high-sensitivity cardiac troponins could improve discrimination more than other biomarkers such as C-reactive protein. Could high-sensitivity troponin take CRP’s place? “It wasn’t a question we specifically set out to answer in this study,” Dr. Selvin says. “But CRP is a nonspecific marker of inflammation, and elevations in CRP are not a specific biomarker of underlying cardiac disease. Elevations in CRP can be caused by many different things, and the test is less reliable than many other standard laboratory measures. It’s not as stable a marker as troponin.”
“Your CRP today may be quite different from your CRP a few weeks from now whereas troponin does not have as much variability. So, absolutely, I think troponin could be more useful than CRP in many circumstances if we’re talking about cardiovascular risk stratification.”
The authors note that their findings about high-sensitivity troponin might be more significant for down-stratifying cardiovascular risk than for up-stratifying patients according to risk. Says Dr. Selvin: “Showing improvements in prediction means we are reclassifying people according to their risk, either up or down. The improvement in down-classification means that oftentimes these comorbidities are classifying people as high risk, when really, they’re not. And so, when you measure troponin, it correctly classifies these individuals downward into their correct risk group.”
Thus, the cost ramifications of the study are significant. “Again, we didn’t study cost-effectiveness specifically. But if we’re doing a more accurate job of classifying people according to their risk, that means more efficiency and hopefully better treatment decisions, overall care, and patient outcomes,” Dr. Selvin says.
High-sensitivity troponin can clearly improve cardiovascular stratification, she says. “So we think there is great potential for these high-sensitivity troponin assays to be used in clinical practice routinely to help monitor risk.”
“A 70 year old is not a 70 year old is not a 70 year old. Some older adults are still working and even running marathons,” Dr. Selvin says, “while others are frail and unable to perform routine daily tasks without assistance. Using objective biomarkers can help us understand that heterogeneity in older patients, and accurately assess mortality risk in what is clearly not a homogeneous group.”
The second Diabetes Care study, by Dr. Saenger and colleagues, also supplies useful information to guide selection of biomarkers to predict development of CVD and major adverse cardiovascular events (MACE) in diabetes. With their focus on childhood-onset type one diabetes, the authors studied the ability of high-sensitivity cardiac troponin and N-terminal prohormone B-type natriuretic peptide to predict CVD and MACE. They examined a cohort of childhood-onset type one diabetes patients, the subject of a historic study, and tested biobanked specimens available at their first presentation following their diabetes diagnosis, which occurred more than 20 years earlier (Costacou T, et al. Diabetes Care. 2020;43[9]:2199–2207).
The two biomarkers, hs-cTnT and NT-proBNP, have been suggested for use in general and at-risk populations with acute illness from CVD or heart failure, but their potential value as biomarkers in stable settings for prediction of long-term cardiovascular events has been less well studied. The predictive value of the two biomarkers has been fairly well characterized for individuals with type two diabetes (and found to contribute to improved CVD prediction) but not in type one, says Dr. Saenger.
“It may in part be due to the fact that there has been a focus more on type two diabetes simply because there is a greater prevalence of type two compared to type one diabetes, and easier in one sense to enroll patients in trials and studies,” explains Dr. Saenger, who serves as the AACC liaison to the CAP Clinical Chemistry Committee. “However, now treatments have improved for type one diabetes, people are diagnosed earlier, they’re living longer, but they’re developing more cardiovascular-related complications. With type two, they’re generally not diagnosed until middle age and can also already manifest a lot of other comorbidities.”
Dr. Saenger
People with type one diabetes are developing similar cardiovascular complications compared with those without diabetes, she says, “but seemingly at an earlier phase because of the toll that diabetes has taken on their bodies.” Dysglycemia and traditional risk factors are known to contribute to the disproportionate cardiovascular risk in type one diabetes compared with the general population—a more than 30-fold risk among young adults. But she and her coauthors believe identification of factors beyond those is long overdue. The team chose hs-cTnT and NT-proBNP as biomarkers of myocardial injury and heart failure, respectively.
To conduct the study, the researchers drew upon data from the Epidemiology of Diabetes Complications study, a prospective investigation of a childhood-onset (<17 years) type one diabetes cohort diagnosed, or seen within one year of diagnosis, at Children’s Hospital of Pittsburgh between 1950 and 1980. The EDC study ran from 1986 to 1988 and continued to follow subjects biennially with surveys for 25 years.
Dr. Saenger’s study determined that quartiles of NT-proBNP were linked with CVD incidence in both sexes and incidence of CVD increased incrementally with each increasing quartile of hs-cTnT in males. Both hs-cTnT and NT-proBNP independently and significantly predicted the incidence of cardiovascular events and MACE more than 20 years later, although sex differences were observed. They are strong, independent predictors of CVD and MACE among individuals with childhood-onset type one diabetes, the authors say. Unfortunately, however, the addition of both biomarkers to a model that included traditional CVD risk factors (BMI, smoking, non-HDL-C, albumin excretion) did not significantly improve the prediction of CVD, Dr. Saenger says.
Similarly, when the two biomarkers were simultaneously added to the model for predicting MACE (which included HbA1c, hypertension status, non-HDL-C, WBC count, and albumin excretion), NT-proBNP was a better predictor of MACE compared with hs-cTnT, but addition of both biomarkers did not improve the prediction of MACE.
The authors write, “Although we were unable to show that their assessment improves outcome prediction beyond that offered by traditional cardiovascular risk factors, it would be unlikely that a single biomarker or even a combination of two biomarkers would significantly improve disease prediction, given the multitude of factors contributing to the pathogenesis of cardiovascular complications in diabetes.”
The observed differences in the distribution of the two biomarkers and the strength of their association with the outcomes studied by sex highlights the value of disaggregating study findings by sex, the authors say. The concentrations of cardiac troponin in females, when using high-sensitivity troponin assays, is lower compared with males. But, they write, “the ability of hs-cTnT to independently predict cardiovascular outcomes has generally not been evaluated separately in men and women, with few exceptions.”
“Our results suggest that compared with men,” they continue, “women are dramatically more likely to experience a subsequent cardiovascular event with hs-cTnT concentrations below the assay limit of detection (LoD) (i.e., <5 ng/L), with no further increase in risk at higher hs-cTnT concentrations.” Confirmation of their results in other cohorts and/or diverse populations with or without diabetes is important, they say, to determine clinical relevance, “because it would suggest that separate prognostic cutoff points should be established for men and women.”
“This is similar to what is noted for a normal, healthy population,” Dr. Saenger says, “and is why sex-specific cutoffs are advocated for when using high-sensitivity troponin assays, with females having a lower 99th percentile compared to males. We were hoping to see more of a signal in our cohort, which was approximately half female, but we were right at the limit of detection and the hs-cTnT assay isn’t the best in terms of sensitivity. But it’s what we had available.” According to the International Federation of Clinical Chemistry and Laboratory Medicine specifications for designations for high-sensitivity troponin assays, the assay needs to detect more than 50 percent of normal individuals in males and females. “The Roche electrochemiluminescence immunoassay, due to its analytical design, does meet that criteria for males but not for females,” Dr. Saenger says of the assay used in the study. She maintains an updated list of analytical characteristics for hs-cTn assays on the International Federation of Clinical Chemistry and Laboratory Medicine website (http://bit.ly/2qKSK25).
The study also determined that NT-proBNP didn’t add a great deal of predictive ability. “It was a stronger predictor of major adverse cardiac events in females compared to males, but we didn’t find this association to be as strong as we thought we would.”
However, Dr. Saenger says, “With an analytically superior high-sensitivity assay, if you’re able to essentially follow a patient’s risk over time, it would be important to know at what timepoint it can accurately predict development of cardiovascular disease. We don’t know that just yet. It would be of value in the outpatient setting to prospectively measure some of these biomarkers in real time and implement routine testing in the future if they show promise.”
She continues to work with her study coauthors by looking at other high-sensitivity cardiac troponin assays as the analytical and diagnostic capabilities of those assays evolve. “Another hs-cTn assay may demonstrate significant results in our type one diabetes cohort, simply because it is analytically superior and can quantitate more values in females,” as one example, she says.
A study that applies the same methodology used in her type one diabetes study to a different cohort might be worthwhile, Dr. Saenger says. “The patients in our study were followed pretty closely and are on the physicians’ radar. It would be really interesting if there were a similar cohort that had specimens collected over decades to look at each individual’s biomarker patterns, in particular with females since a large proportion of type one diabetics are women, and evaluate the correlation of biomarkers with cardiovascular disease or some type of subclinical myocardial dysfunction that we didn’t know about.”
Anne Paxton is a writer and attorney in Seattle.