European Heart Journal Advance Access originally published online on March 10, 2008
European Heart Journal 2008 29(7):837-839; doi:10.1093/eurheartj/ehn075
Developing the utility of blood biomarker associations beyond population sample linkage to events in cardiovascular patients
University Department of Medicine and Department of Cardiology, City Hospital, Dudley Road, Birmingham B18 7QH, UK
* Corresponding author. Tel: +44 121 507 4476/5080, Fax: +44 121 554 4083, Email: Robert.macfadyen{at}swbh.nhs.uk 
doi:10.1093/eurheartj/ehn007
This editorial refers to ‘The value of N-terminal fragment of brain natriuretic peptide and tissue inhibitor of metalloproteinase-1 levels as predictors of cardiovascular outcome in the LIPID study’
by M.J. West et al., on page 923
Footnotes
The opinions expressed in this article are not necessarily those of the Editors of the European Heart Journal or of the European Society of Cardiology.
West et al.1 report a small, well designed post hoc nested case–control sample from the LIPID trial of pravastatin.2 This was completed in low risk subjects with documented coronary disease in the form of a previous confirmed acute coronary syndrome (ACS). This additional report examines biomarker linkage over a relatively short time frame of 2–3 years and highlights two measurements perhaps not intuitively linked to the biology of coronary disease progression. The conditional logistic regression analysis shows a significant linkage association following point analysis for N-terminal brain natriuretic peptide (NT-BNP) and tissue metalloproteinase-1 (TIMP-1). TIMP-1 has been linked to cardiovascular events in low risk patients previously, in patients with less well confirmed coronary disease.3 Its re-affirmation here as a population predictor of events in this group of patients is of value. Similarly for NT-BNP, which is frequently presumed to be linked to cardiac events largely via the presence of association with ventricular dilatation, remodelling, or ventricular ‘dysfunction’ (not factors intuitively present even in patients with a confirmed yet uncomplicated ACS), this is not the first report to link it to coronary outcomes even in the absence of overt ventricular impairment.4 More interestingly, in this report there was a significant linkage expressed by both markers to subsequent emergent stroke disease.5 Both NT-BNP (indirectly) and TIMP-1 (directly) can be linked, at least theoretically, to structural change within the heart, be it myocytic, functional (i.e. linked to dynamic contractile changes), geometric (fixed changes in cardiac shape or wall thickness), or within the connective tissue support structure.6,7 Both are of course linked in the biology of hypertension and/or hypertensive heart disease and vascular atherosclerosis,8,9 but this and other conventional risk factors were clearly accounted for in the analysis of West et al.1 These authors were concerned to address the apparent lack of statistical association seen with high sensitivity C-reactive protein (hs-CRP) and interleukin-6 (IL-6) in this study. These are well described and useful biomarkers,10 but may simply not be sensitive or specific enough to stand out using this form of analysis (the number, frequency, or timing of sampling in a multicentre trial) or in this sample (in this number of patients, with this level of disease risk, and at this stage in the evolution of atherosclerosis).
As here for the LIPID study, most large-scale clinical trials collect large numbers of blood samples taken as a protocol routine. The subsequent analyses conducted are published as valid secondary end-points that give a degree of useful, if limited, information. However, in the setting of multicentre trials, the analyte selected for the population sample studied is not directed at new underlying biology. This is understandable as the blood biomarker is often not the primary goal of the study. In many instances, despite large numbers of repeated visits, the biomarker data reported are often a single measure in a subset of the trial population.
The association of both NT-BNP and TIMP-1 with the pathology of atherosclerosis is not in doubt. As in the report of West et al.,1 the statistically significant population associations of biomarkers that are demonstrable will vary dependent on the clinical patient sample and the point of analysis. This is because the evolution of disease and its controlling or mediating factors (which many biomarkers represent) are not likely to be either linear or a static hierarchy (with different factors dominating disease progression at different times in individual patients). In turn the selected biomarker (or physical biomeasurement for that matter) will indicate predominantly one aspect of the biology of disease progression (e.g. for TIMP-1 the connective tissue composition and turnover of blood vessels or myocardium). In addition, the accuracy of blood biomarker estimates is dependent on variance attributable to both the biology and the analytical technique, sampling density, storage, and the stage of the disease at the time of sampling.11 The accuracy of a physical biomeasurement will be operator and technique dependent.
Currently routine management of acute and chronic disease is traditionally defined by symptomatic assessment, physical examination, and, for some cardiovascular states, intermittent physical measurements. These processes are clearly operator dependent, relying on an accurately obtained and truthful symptomatic enquiry, and skill in physical examination and/or technical biomeasurement and their interpretation in individual circumstances (such as blood pressure, echocardiography; electrocardiography; angiographic or other imaging technologies, etc). Currently the most obvious area where blood biomarker measurement has revolutionized standards of definition of disease and fundamentally altered patterns of cardiovascular care would be in the clarification and redefinition of treatment in ACS.12 Previously defined by relatively insensitive and non-specific symptomatic scoring, cardiac troponin measurements have revolutionized the recognition of and provided targeting for the accelerated management of coronary disease. However, the interpretation and application of biomarker and/or physical biomeasurements by repeat measures within an individual are now needed to move this technology beyond linkage analysis and implication in the general processes of pathobiology.
Many valid studies similar to that of West et al.1 reaffirm previously documented biomarker associations but they do not develop new ideas or test new hypotheses. The ability to take these data forward in a practical clinical setting is dependent on the application of repeated measures within individuals, which could potentially define clinical stability and/or allow anticipation of specific clinical events.13 As above, the biomarker (or biomarkers) selected, their sampling frame (both patient and disease state), the technology of measurement, and the pathobiology of disease progression would determine the events predicted. Biomarker selection should be based on a specific target, either an aetiological feature or a prognostic feature. Thus, for example in left ventricular systolic dysfunction (LVSD), one biomarker might express appropriate sensitivity and specificity for the presence and or progression of ischaemia, whereas a completely different biomarker might illustrate myocyte remodelling, and another might illustrate fluid or volume loading of the heart. It is of interest that traditional mediators of events in LVSD such as norepinephrine, angiotensin II, aldosterone, or renin have rarely been linked to individual circumstances and even more rarely measured serially within individual patients. In the same way, if we integrate their role in defining outcome and/or if changes in renal function and therapy play a big role in this, then we need to appreciate that a biomarker screen could easily build up a picture of the dynamic stability of the patients and processes involved in LVSD.
We already have intra-individual repeated measures monitoring technology based on physical biomeasures within most modern implantable heart failure devices. These are based on biomeasures such as simple indices of heart rate variability and bioimpedance.14 Routine trend analyses can be employed to identify significant changes and establish linkage to individual disease state stability sufficient to anticipate events and guide therapeutic interventions. While available for some years, the application of this technology is limited to those with an implanted resynchronization or anti-tachyarrhythmia device. The development of more rapid, simple, and reliable analyses of cardiovascular biomarkers and application of these principles of repeated measures within the individual is a simple and perhaps more easily extendable extension of this clinical management structure.
The time for practical testing of these systems has come. The research challenge is to define the operating characteristics underlining the importance of selecting the correct biomarker for the correct stage of disease. The specific analysis will depend on which aspect of pathobiology it is designed to illustrate and define the correct interpretation of the change in levels that it might indicate (whether this is a relative increase, relative decrease, or altered intra-subject variance). The change of course is within the individual subject. Finally, we must ensure and test that such systems perform better than the conventional standard using realistic operators (neither individuals with unmatched skill, nor practitioners who pay scant attention to clinical presentation and examination). This is a new and exciting challenge stemming from population research findings such as those summarized by West et al., but importantly extends their impact to the management of individual patients. This holds greater potential for rewards for the considerable research effort by many multicentre trial groups. Finally this application of relevant biomarker technology can be used in many aspects of medical care and not simply cardiovascular disease, far less coronary disease alone.
Conflict of interest: none declared.
Footnotes
The opinions expressed in this article are not necessarily those of the Editors of the European Heart Journal or of the European Society of Cardiology.
References
- West MJ, Nestel PJ, Kirby AC, Schnabel R, Sullivan D, Rimes RJ, Pollicino C, Lubos E, Münzel TF, White HD, Tonkin AM, Bickel C, Tiret L, Blankenburg S. for the LIPID Study Investigators. The value of N-terminal fragment of brain natriuretic peptide and tissue inhibitor of metalloproteinase-1 levels as predictors of cardiovascular outcome in the LIPID study. Eur Heart J (2008) 29:923–931. First published on February 22, 2008. doi:10.1093/eurheartj/ehn007.
[Abstract/Free Full Text] - The Long Term Intervention with Pravastatin in Ischaemic Disease (LIPID) Study Group. Prevention of cardiovascular events and death with pravastatin in patients with coronary heart disease and a broad range of initial cholesterol results. N Engl J Med (1998) 339:1349–1357.
[Abstract/Free Full Text] - Lubos E, Schnabel R, Rupprecht HJ, Bickel C, Messow CM, Prigge S, Cambien F, Tiret L, Münzel T, Blankenberg S. Prognostic value of tissue inhibitor of metalloproteinase-1 for cardiovascular death among patients with cardiovascular disease: results from the Athero Gene study. Eur Heart J (2006) 27:150–156.
[Abstract/Free Full Text] - Bibbins-Domingo K, Gupta R, Na B, Wu AH, Schiller MB, Whooley MA. N-Terminal fragment of the pro-hormone brain-type natriuretic peptide (NT-proBNP), cardiovascular events, and mortality in patients with stable coronary heart disease. JAMA (2007) 297:169–176.
[Abstract/Free Full Text] - Castellanos M, Serena J. Applicability of biomarkers in ischemic stroke. Cerebrovasc Dis (2007) 24(Suppl 1):7–15.[CrossRef][Medline]
- Orn S, Manhenke C, Squire IB, Ng L, Anand I, Dickstein K. Plasma MMP- 2, MMP-9 and N-BNP in long-term survivors following complicated myocardial infarction: relation to cardiac magnetic resonance imaging measures of left ventricular structure and function. J Card Failure (2007) 13:843–849.[Web of Science][Medline]
- Tayebjee MH, Lip GYH, MacFadyen RJ. Matrix metalloproteinases in coronary artery disease: clinical and therapeutic implications and pathological significance. Curr Med Chem (2005) 12:917–925.[CrossRef][Web of Science][Medline]
- Pedersen F, Raymond I, Kistorp C, Sandgaard N, Jacobsen P, Hildebrandt P. N-terminal pro-brain natriuretic peptide in arterial hypertension: a valuable prognostic marker of cardiovascular events. J Card Fail (2005) 11(5 Suppl):s70–s75.[CrossRef][Web of Science][Medline]
- Tayebjee MH, Nadar SK, MacFadyen RJ, Lip GY. Tissue inhibitor of metalloproteinase-1 and matrix metalloproteinase-9 levels in patients with hypertension: relationship to tissue Doppler indices of diastolic relaxation. Am J Hypertension (2004) 17:770–774.[Web of Science][Medline]
- Cesari M, Penninx BW, Newman AB, Kritchevsky SB, Nicklas BJ, Sutton-Tyrrell K, Rubin SM, Ding J, Simonsick EM, Harris TB, Pahor M. Inflammatory markers and onset of cardiovascular events: results from the Health ABC study. Circulation (2003) 108:2317–2322.
[Abstract/Free Full Text] - Chuen MJNK, MacFadyen RJ. Sources of variance in blood biomarker measurement relevant to their use in describing the physiology and pathology of left ventricular systolic dysfunction. Biomarkers Med (2008) 1:355–374.[CrossRef]
- Silber S, Albertsson P, Avilés FF, Camici PG, Colombo A, Hamm C, Jørgensen E, Marco J, Nordrehaug JE, Ruzyllo W, Urban P, Stone GW, Wijns W. Task Force for Percutaneous Coronary Interventions of the European Society of Cardiology. Guidelines for percutaneous coronary interventions. The Task Force for Percutaneous Coronary Interventions of the European Society of Cardiology. Eur Heart J (2005) 26:804–847.
[Free Full Text] - MacFadyen RJ. Intraindividual temporal variance of biomarkers and the definition of individualized prognosis in cardiovascular patients. J Hum Hypertension (2006) 20:383–386.[CrossRef][Web of Science][Medline]
- Wang L. Fundamentals of intrathoracic impedance monitoring in heart failure. Am J Cardiol (2007) 99:3G–10G.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
Related articles in EHJ:
- The value of N-terminal fragment of brain natriuretic peptide and tissue inhibitor of metalloproteinase-1 levels as predictors of cardiovascular outcome in the LIPID study
- Malcolm J. West, Paul J. Nestel, Adrienne C. Kirby, Renate Schnabel, David Sullivan, R. John Simes, Christine Pollicino, Edith Lubos, Thomas F. Münzel, Harvey D. White, Andrew M. Tonkin, Christoph Bickel, Laurence Tiret, Stefan Blankenberg, and for the LIPID Study Investigators
EHJ 2008 29: 923-931.[Abstract] [FREE Full Text]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||