Serial Measurement of Biomarkers after Acute Coronary Syndrome: Which One to Choose?

Measurement of circulating biomarkers is a cornerstone of cardiovascular medicine. Biomarkers reflect pathobiological mechanisms and the associated risks, often in a more precise manner than standard measures. Accordingly, biomarkers may be valuable complements to commonly used risk estimates. One of the most important areas where biomarkers add to clinical assessment is the management of patients with non–ST-elevation acute coronary syndrome (NSTEACS),2 both in the acute phase and during follow-up after clinical stabilization. However, despite a multitude of biomarkers individually having been shown to provide important information, current routines focus only on the measurement of cardiac troponin, estimates of renal function, and metabolic parameters (i.e., blood glucose and lipids), as their levels affect patient management. Why are cardiovascular biomarkers not used more frequently? The hurdles are high. To be regarded as clinically useful, a biomarker should not only specifically reflect a distinct disease process. In addition, the biomarker should act as a clue to the initiation of beneficial treatments, and the achieved modification of risk should go hand in hand with a change in the biomarker level (1) at the same time at which its relative prognostic value is known to change in relation to time elapsing from the acute event (2). Into this complicated scenario comes the analysis from Chan et al. presented in the current issue of Clinical Chemistry. The authors measured the concentrations of N-terminal pro–B-type natriuretic peptide (NT-proBNP) and C-reactive protein (CRP) in 1665 primarily medically managed NSTEACS patients enrolled in the TaRgeted platelet Inhibition to cLarify the Optimal strateGy to medicallY manage Acute Coronary Syndromes (TRILOGY-ACS) trial (3). The investigators measured biomarker concentrations at randomization (at a median of 4 days from hospital admittance), after 30 days, and at 6 months. NT-proBNP concentrations decreased by 28% and CRP concentrations by 62% during the observation period. Decreases in both biomarker concentrations were associated with a diagnosis of myocardial infarction (MI), reflecting more pronounced myocardial damage with higher biomarker concentrations initially, and coronary angiography, likely reflecting, although not formally tested, the benefit from coronary interventions. Increases in biomarker concentrations were associated with cardiovascular risk features including a longer delay from symptom onset to hospital admission, and, for CRP, better renal function and higher hemoglobin values, findings for which explanation remains elusive. The main result from the analysis was that NT-proBNP changes during the first 6 months predicted cardiovascular mortality during 30 months of follow-up, whereas CRP changes predicted MI. The effect sizes, however, were only moderate, with an 8%–9% adjusted risk of the composite end point of cardiovascular mortality, MI, or stroke per 40% change increments of each respective biomarker. The results for NT-proBNP are consistent with and extend data from previous studies. Higher B-type natriuretic peptide (BNP) concentrations are important predictors of mortality in NSTEACS (2, 4, 5). Samples obtained late during hospitalization or follow-up are preferable to those obtained very early, as they provide stronger associations with subsequent events (2, 4, 5). However, the evidence regarding whether temporal changes in BNP concentrations per se would be associated with a change in outcome has been conflicting so far (4, 5). Interestingly, randomization to an early invasive vs conservative strategy had no effect on NT-proBNP changes during the first 6 weeks after NSTEACS in FRISC II (Framingham and Fast Revascularization During Instability in Coronary Artery Disease II)(4). The discrepancy with the findings reported by Chan et al. (3) might reflect that coronary interventions in TRILOGY-ACS had been performed on the basis of clinical judgment, i.e., mainly in high-risk patients likely achieving greater treatment-related benefit along with decreases in NT-proBNP concentrations. The interpretation of the results regarding CRP changes in TRILOGY-ACS is more challenging. Baseline CRP concentrations were not prognostic at all, but increases independently predicted the composite end point, driven by an association with MI (3). These findings are novel but also evoke questions. Is the ischemic risk associated with a CRP increase of 40% the same for baseline concentrations of 1.7 or 13.1 mg/L (i.e., the 25th and 75th percentiles in the cohort from TRILOGY-ACS)? Probably not. CRP concentrations during an acute phase of NSTEACS do not predict MI (6), and, unfortunately, the study from Chan et al. does not help us to relate the ischemic risk associated with CRP changes to a given baseline value. Under stable conditions, CRP concentrations reflect inflammation in the vasculature but are boosted in the acute phase of NSTEACS due to the acute-phase reaction resulting from myocardial necrosis. CRP concentrations correlate with the MI size (7), and accordingly, an association of their changes with cardiovascular mortality (6) would have been expected in the analysis from Chan et al. However, the adjusted hazard ratio of a 40% CRP change regarding this outcome was exactly 1.00. On the other hand, decreases in CRP concentrations have been shown to mirror plaque stabilization by high-dose statin treatment in patients with stable coronary artery disease and acute coronary syndrome (8, 9), and lower CRP concentrations following statin treatment predict greater risk reduction after NSTEACS (10, 11). Given the low rate of coronary interventions in TRILOGY-ACS, one could thus speculate that increasing CRP concentrations over time might reflect atherosclerotic plaque activation, for example, in the untreated coronary culprit, resulting in recurrent coronary instability and more frequent ischemic events. How do these data translate into possible improvements of the management of patients with NSTEACS? First, the results from Chan et al. support the notion that changes in biomarker levels reflect the efficacy of an early invasive approach. While this also applies to baseline concentrations of BNPs (12), baseline CRP concentrations have no proven role for the choice of such treatment, however (7). Second, BNP concentrations are known to reflect the degree of left ventricular dysfunction during NSTEACS. This is why it is conceivable that these concentrations might be useful for guidance of pharmacological treatment, aiming at reducing mortality by prevention of adverse myocardial remodeling. Results from previous studies have been disappointing in this regard, however (13, 14). Nonetheless, the data from Chan et al. indicate that serial NT-proBNP measurements still might be of clinical value in patients with a particularly vulnerable phenotype, i.e., noninvasively managed individuals with prevalent cardiovascular risk features and/or large MI (3). Several questions remain. What is the optimal time point for remeasurement of biomarkers? In other words, how long should one dare to wait, and how many events would be acceptable, in high-risk individuals who might have benefited from biomarker-guided interventions in the meantime? Unfortunately, 30-day changes in biomarker concentrations were mostly non-prognostic in the analysis from Chan et al. (3). This might reflect limitations in statistical power, but also selection bias, as the study population consisted of 6-month survivors. On the other hand, NT-proBNP concentrations at 30 days independently predicted both cardiovascular mortality and MI (3), in accord with other studies (2, 4, 5). Thus, the central question is whether delta changes or absolute values should be used as the benchmark to target patients for medical interventions, and which biomarker should be chosen. Considering their prognostic value as soon as 4–16 weeks after an acute episode of NSTEACS (2, 3, 4, 5), BNP concentrations appear to be preferable. Target values of 80 ng/L for BNP, and around 400 ng/L for NT-proBNP, have been tested or suggested (4, 5). CRP concentrations and their changes appear to gain prognostic strength at measurements that are more remote from the acute event (2, 15), and for this reason might not be the first choice. Whether other biomarkers, such as cardiac troponins measured with high-sensitivity assays, add prognostic increments in the post NSTE-ACS setting remains to be investigated. How shall these patients be managed in the meantime? Studies investigating the value of BNP-guided management strategies, similar to those performed in patients with stable congestive heart failure, would provide valuable information. The need for such trials was emphasized more than a decade ago (5). The analysis from Chan et al. emphasizes that such studies could be done and hopefully will be an important contribution to the improvement of follow-up routines in patients with a recent NSTEACS. non–ST-elevation acute coronary syndrome N-terminal pro–B-type natriuretic peptide C-reactive protein TaRgeted platelet Inhibition to cLarify the Optimal strateGy to medicallY manage Acute Coronary Syndromes myocardial infarction B-type natriuretic peptide.