Prognostic Performance Of Multiple Biomarkers In The Context Of Conventional Risk Factors In Patients With Coronary Artery Disease

Background: Cardiovascular disease (CVD) is the leading cause of death and disability worldwide. This highlights the need for accurate risk stratification. An increasing number of novel biomarkers have been identified to predict cardiovascular events. So far,few data compare cardiovascular biomarkers from different pathways or assess the incremental performance of a multimarker panel for risk prediction in patients with established coronary artery disease (CAD).Object: To evaluate whether simultaneous testing of these biomarkers in a population with CAD adds clinically useful incremental information. To evaluate whether a panel of these cardiac-derived biomarkers is superior to strategies that use single biomarker assessments or established risk factors only for cardiovascular risk prediction.Method: We evaluated the ability of 8 biomarkers representing distinct pathophysiological pathways related to cardiovascular events in CAD. Biomarkers representing inflammation (growth differentiation factor-15 [GDF-15], fibrinogen), renal function (creatinine), endothelial activation (uric acid [UA]), cardiovascular stress(N-terminal-pro-B-type natriuretic peptide [NT-proBNP], soluble ST2 [sST2]), plaque instability (pregnancy-associated plasma protein A [PAPP-A] ),and activated coagulation(D-dimer) in 3440 patients with CAD and assessed their association with myocardial infarction, stroke, cardiovascular death, heart failure and all cause death (clinical events,n=419) over 2.9 years.Results: The strongest association with outcome in multivariable-adjusted analyses was observed for NT-proBNP (hazard ratio [HR] for one standard deviation [SD] increase 1.67, 95% confidence interval [Cl] 1.55-1.8, C-index 0.824), GDF-15 (HR 1.74, 95% CI 1.52-2, C-index 0.806), fibrinogen (HR 3.83, 95% CI 2.66-5.52, C-index 0.798), UA (HR 2.7, 95% CI 1.92-3.8, C-index 0.792), D-dimer (HR 1.51, 95% CI 1.37-1.67, C-index 0.8)and creatinine (HR 2.51, 95% CI 2.1-3.01, C-index 0.8). Each of these single markers and their combination (C-index 0.835) added predictive information beyond that obtained by baseline model and led to substantial reclassification (P-integrated discrimination improvement <0.05). Combination of the top six biomarkers (NT-proBNP, GDF-15, UA,fibrinogen, D-dimer, creatinine) calculate a multimarker score. The multimarker panel was also superior to a model containing clinical risk factors and led to significant improvements in classification and discrimination accuracy for composite of cardiovascular events and all cause death (P<0.001). However, the NRI and IDI for the combination of six biomarkers were not significantly different from those resulting from NT-proBNP. The C-index improved for all single biomarkers and the combination of the top six biomarkers (all p<0.001). The C-index for the combination of all top six biomarkers was significantly increased compared with that obtained from the strongest single biomarker NT-proBNP (C-index 0.835 vs. 0.824, p=0.0045).Conclusion: GDF-15, fibrinogen, UA, D-dimer, NT-proBNP, and creatinine are complementary prognostic biomarkers. They are independently and significantly enhanced prognostic performance and reclassification of combined outcomes of cardiovascular events and all course death. Importantly, when combined in risk model, the multimarker approach added incremental risk information beyond traditional cardiovascular risk factors,and showed the highest model stability and precision. NT-proBNP in direct comparison proved to be the strongest single biomarker with substantial reclassification. Notably, the multimarker approach significantly improved risk prediction and stratification compared with NT-proBNP, however, the improvement in risk reclassification is very modestly. The muitimarker score by a unique concurrent analysis of this pathobiologically broad group of biomarkers of in a large well-characterized population with CAD may have potential utility as a screening tool, a premise that would require additional studies. Mutimarker approach may improve identification of high-risk patients with CAD in need of more aggressive therapeutic interventions.