A multi-biomarker risk score improves prediction of long-term mortality in patients with advanced heart failure.

BACKGROUND

Accurate risk prediction is important for an adequate management of heart failure (HF) patients. We assessed the incremental prognostic ability of a multi-biomarker approach in advanced HF.

METHODS

In 349 patients with advanced HF (median 75 years, 66% male) we investigated the incremental prognostic value of 12 novel biomarkers involved in different pathophysiological pathways including inflammation, immunological activation, oxidative stress, cell growth, remodeling, angiogenesis and apoptosis.

RESULTS

During a median follow-up of 4.9 years 55.9% of patients died. Using multivariable Cox regression and bootstrap variable-selection age, chronic obstructive pulmonary disease, N-terminal pro-B-type natriuretic peptide (NT-proBNP) and the following 5 novel biomarkers were retained in the best mortality prediction model: the chemokine fractalkine, the angiogenic and mitogenic hepatocyte growth factor (HGF), the growth differentiation factor 15 (GDF-15) influencing cardiac remodeling and apoptosis, and the 2 pro-apoptotic molecules soluble apoptosis-stimulating fragment (sFAS) and soluble tumor necrosis factor-related apoptosis-inducing ligand (sTRAIL). This multi-biomarker score had strong discriminatory power for 5-year mortality (area under the Receiver Operating Characteristic curve [AUC]=0.81) and improved risk prediction beyond the prognostic power of a comprehensive conventional risk score including known clinical predictors and NT-proBNP (AUC=0.77). Net reclassification confirmed a significant improvement of individual risk prediction (p=0.003).

CONCLUSIONS

Risk prediction by a multi-biomarker score is superior to a conventional risk score including clinical parameters and NT-proBNP. Additional predictive information from different biological pathways reflects the multisystemic character of HF.