Old, the new, more or less and the conundrum for biomarkers in cardiovascular diseases?

Abstract

Biomarkers are critical for diagnostic, monitoring and management of cardiovascular diseases. Opportunities to improve how biomarkers are used are important. This editorial on biomarkers in coronary artery disease provides commentary on the study as well as broad benchmarks on what make ideal biomarkers. Biomarker discovery is difficult and opportunities to expand and improve their use should be encouraged.

Key Words: Biomarker; Panel; Coronary artery disease; Heart failure; Observation trial

Core Tip: Biomarkers are among a few pillars of medical care however caution is raised and encouragement is provided for observations into biomarker studies. Combination of biomarkers are a novel method and may increase the appeal of single biomarkers. Design of experiments to evaluate biomarkers and diseases are as important as the specific biomarkers being tested.

TO THE EDITOR

Li et al[1] in their recent observational study on biomarkers for coronary heart disease (CHD) and heart failure (HF), the authors, remarked that serum cystatin C, monocyte/high-density lipoprotein cholesterol ratio, along with serum uric acid are appear as useful biomarkers to diagnose these cardiovascular conditions with the potential for combined CHD and HF diagnosis. The authors went on to comment on ease of access, routine availability, reliability, enhancement of clinical decisions and risk assessment. Furthermore, their role in panels could enhance their accuracy and prognostic evaluation. These are bold claims in an era where we are well advanced in the use of biomarkers for a range of clinical scenarios from diagnosis, monitoring to risk stratification and prognostication. In a sense we are at a juncture where we are looking to reinvent or expand the role of biomarkers for clinical practice. For CHD and HF let explore this from what is established and what is unknown.

What makes a good biomarker? This is a complex question and has had opinions from leaders, where Koenig et al[2] suggested that for primary prevention, the traditional risk factors can act a first screen to risk stratify future cardiovascular events. These risk scores such as Framingham, the European Society of Cardiology Systematic Coronary Risk Evaluation, or the Prospective Cardiovascular Münster score, and others derived from large datasets analysed through multivariable statistical models, can miss ‘a considerable’ number of otherwise at-risk patients[2-4]. This knowledge justifies the search for how to improve utility of know biomarkers and the search for non-traditional risk factors. On this point we applaud any new publication looking to improve the knowledge base in this area. Li et.al and co-authors suggested that ‘serum cystatin C, monocyte/high-density lipoprotein cholesterol ratio, and uric acid are valuable biomarkers for diagnosing CHD and HF, with the potential for combined CHD + HF diagnosis’.

In this paper the authors are exploring novel avenues of using established biomarkers to diagnose cardiovascular disease (CVD), which have existing biomarkers which have a high degree on sensitivity and specificity in diagnosis (Table 1). Vasan et.al and other papers have listed in detail the definition and types of biomarkers[5] and their characteristics[6]; extended panel of biomarkers[7], including with novel techniques[8], and those that have pooled data on biomarkers and risk[9]. In the traditional sense single biomarkers or limited panels suggest that troponin (CK-MB) and brain natriuretic peptides alone for CHD and CHF or together for both have high sensitivity and specificity, predictive, robust, stable, non-invasive and with preclinical and clinical value[6,7,9]. When we add in clinical history and examination, electrocardiogram and echocardiography this equation improves. Thus, before venturing further into the debate of scientific merit, is this combination of biomarkers relevant. Firstly, if cost and access are factors then yes in the right jurisdiction this observation could have merit. An example would be if cost is prohibitive for laboratories or point of care device or we don’t have the ability to use more than one modality to interrogate the heart (e.g. remoteness).

Table 1 Common biomarker and interpretation.

	Troponin (CK-MB)	BNP	Cystatin-C	Monocyte/high-density lipoprotein cholesterol ratio	Uric acid
Source	Myocardium	Atrium	Nucleated cell	Balance between inflammatory and oxidative stress of monocytes and HDL-C	Purine breakdown waste product
CHD	++	+	+	+/-	+/-
CHF	+	++	+	+/-

BNP: Brain natriuretic peptide; CHD: Coronary heart disease; CHF: Congestive heart failure; HDL-C: High-density lipoprotein cholesterol.

On the scientific merit Troponin and BNP are cardiac specific. The three biomarkers studied do not have that degree of specificity on its origin relevant to CVD. Wu et al[10] noted that cardiac Troponins (T and I) with NT-proBNP adds incremental value in with risk factors for the PREDICT score assessing incident CVD in primary care; 4102 asymptomatic patients were assessed within a randomised and controlled trial using Vitamin D[10,11]. Thus, from this paper it can similarly be extracted that with the right use, protocols and suspicion of CAD and or CHF this novel panel could add value. However there remains a gap from this early observation to what the settings are (ambulatory primary, secondary prevention; subacute symptomatic and hospital patients), the confounders and the list of biomarker credentials raised earlier[5,6]. While we should be encouraging of all levels of research; findings and causation are a sacred cow. Drawing such conclusions warrant the most robust of evidence. Early observation of note, and further evidence encouraged.