Novel biomarkers--the long march from bench to bedside

In recent years, a deeper understanding of the pathobiology of atherothrombosis as the underlying mechanism of acute coronary syndromes (ACS) has directed scientific studies towards the evaluation of novel serum biomarkers as potential diagnostic tools for the clinical setting. Prerequisites for a biomarker to enter clinical routine in ACS are several fold. The most important features are that the marker can be offered on a routine platform, provides higher sensitivity and specificity than the electrocardiogram in predicting outcome, and it should impact therapeutic decision making. Markers reflecting increased risk for coronary artery disease, such asserum creatinine, lipid levels or glycated haemoglobin (HbA1C), are only useful in predicting long-term prognosis. For the acute phase, the highest value can be expected from biomarkers that closely reflect the underlying pathophysiology, i.e. inflammation and thrombus formation. In fact, cardiac troponins as surrogate markers for local thrombus formation with consecutive downstream embolization have become the cornerstone of risk stratification in patients presenting with ACS.1 Moreover, the therapeutic decision making in terms of invasive or conservative management as well as selection of antithrombotic treatment can be guided by the troponin result.2,3 Similarly, B-type natriuretic peptides (BNP and NT-proBNP) are valid markers reflecting cardiac function and myocardial ischaemia, and predicting adverse clinical events during midterm follow-up. Although there are some data suggesting a benefit from an invasive management of patients with elevated BNP/NT-proBNP levels,4 the link to therapeutic measures could not yet be established as convincingly as for troponins. However, this marker has been recommended for risk stratification in current guidelines on the treatment of ACS and thus entered routine use. Inflammation represents key pathophysiological mechanisms in ACS. Accordingly, elevated high sensitive C-reactive protein (hsCRP) levels could be linked to increased long-term mortality and in fact this has been confirmed in various studies.5 However, CRP as a global acute phase reactant is not specific for vascular processes and can be found elevated in a variety of inflammatory conditions. Thus local inflammation is less well reflected by hsCRP, probably precluding its value for the acute phase of ACS. There are still a considerable numbers of patients who are at increased risk undetectable by markers currently available for routine use, i.e. troponins, BNP, and hsCRP. Since the pathophysiology is complex, not a single marker but a combination of markers will probably be used in the future, representing different mechanisms. Accordingly, the search goes on for combinations of ideal markers. A large number of candidate biomarkers have been investigated in numerous studies, with the aim to improve diagnosis and risk assessment further. Many can never be considered for clinical routine because the analytical requirements preclude this. Myeloperoxidase (MPO) and soluble CD40 ligand (sCD40L) are two biomarkers out of this multitude which were considered to have reached the level of clinical routine. MPO is a member of the haem peroxidase superfamily, traditionally regarded as a microbicidal enzyme. It is released from activated polymorphonuclear neutrophils and monocytes where it is stored in azurophilic granules at sites of inflammation. Furthermore, in ACS, increased numbers of MPO-expressing macrophages have been described in eroded or ruptured plaques. In contrast, macrophages of human fatty streaks contained no or only small amounts of MPO. It actively promotes endothelial dysfunction and plaque destabilization. Therefore, it can be viewed as a factor closely reflecting local processes. Consequently, the predictive value of MPO has consistently been described in apparently healthy individuals, and throughout the entire spectrum of cardiovascular diseases from heart failure to stable coronary artery disease and to ACS. Notably, the predictive value of MPO was independent of other biomarkers, in particular of troponins.6–10 The CD40 and CD40 ligand (CD40L) system is expressed on a variety of cell types including activated platelets, vascular endothelial cells, vascular smooth muscle cells, monocytes, and macrophages. Following expression on the cell surface, CD40L is partly cleaved by proteases and subsequently released into the circulation as soluble CD40 ligand (sCD40L) which can be detected in serum and plasma. The main source of circulating sCD40L is platelets.11 The binding of CD40L enhances the inflammatory response, acts prothrombotically, leads to plaque destabilization, and inhibits endothelial regeneration. From several clinical studies it has consistently been reported that sCD40L is elevated in patients with ACS and that it provides prognostic information with therapeutic implications independent of established cardiac markers, e.g. cardiac troponins.12 However, pre-analytical conditions are decisive for the assessment of sCD40L and may preclude routine clinical use.13 Morrow et al.14 present retrospective data on MPO and sCD40L in patients with ACS. The key findings of this study are that baseline concentrations of MPO in patients with ACS and simultaneous pharmacological treatment are associated with an increased short-term (30 days) risk for non-fatal recurrent ischaemic events, whereas the concentration of sCD40L was without predictive value. Even though this is not the first report of MPO and sCD40L in patients with ACS, and therefore more or less validates previous findings, it has implications for the future use of biomarkers. The potential conclusion is that the combination of troponin, BNP, and MPO seems currently to provide the best risk stratification offered by biomarkers. The data presented challenge the role of hsCRP. In a previous report from the same group, a multimarker strategy for risk assessment of patients with ACS integrating troponin I, BNP, and hsCRP had been proposed.15 In the current analysis, however, a panel of multiple biomarkers with troponin I, BNP and MPO instead of hsCRP was superior. Should MPO therefore replace hsCRP in future as a better inflammatory marker? Despite the potential superiority of this marker over hsCRP, there is also an important flaw to MPO. The therapeutic consequences of elevated MPO levels in patients with ACS remain unclear. Since in the present study MPO failed to predict a benefit from an early invasive treatment as compared with a conservative approach, other strategies such as the use of statins need to be explored. Why did sCD40L fail as a risk marker? After the first reports on sCD40L as a marker for platelet activation and inflammation in patients with ACS, there has been great enthusiasm that sCD40L might become a clinically important diagnostic tool, particularly useful to decide on optimal medical treatment. However, pre-analytical factors seem to limit a broader application in daily practice. Furthermore, just recently, contradictory results have been reported, raising concerns of the reproducibility of sCD40L as a biomarker for risk stratification.16 The present study confirms the limitations of sCD40L as a clinical marker. Like novel drugs seeking labelling, biomarkers must go through different phases of evaluation. This roadmap includes various steps from the first experimental studies, followed by clinical studies as a proof of concept, to confirmatory and comparative analyses, as well as interventional trials (Figure 1). Roadmap for the evaluation of novel biomarkers. The present study must be viewed in this context. MPO but not sCD40L has the potential to replace hsCRP as an inflammatory biomarker for risk stratification in ACS. The introduction of MPO into clinical routine should be considered. Conflict of interest: none declared.