Machine learning-based biomarker profile derived from 4210 serially measured proteins predicts clinical outcome of patients with heart failure

Abstract

Aims: Risk assessment tools are needed for timely identification of patients with heart failure (HF) with reduced ejection fraction (HFrEF) who are at high risk of adverse events. In this study, we aim to derive a small set out of 4210 repeatedly measured proteins, which, along with clinical characteristics and established biomarkers, carry optimal prognostic capacity for adverse events, in patients with HFrEF.

Methods and results: In 382 patients, we performed repeated blood sampling (median follow-up: 2.1 years) and applied an aptamer-based multiplex proteomic approach. We used machine learning to select the optimal set of predictors for the primary endpoint (PEP: composite of cardiovascular death, heart transplantation, left ventricular assist device implantation, and HF hospitalization). The association between repeated measures of selected proteins and PEP was investigated by multivariable joint models. Internal validation (cross-validated c-index) and external validation (Henry Ford HF PharmacoGenomic Registry cohort) were performed. Nine proteins were selected in addition to the MAGGIC risk score, N-terminal pro-hormone B-type natriuretic peptide, and troponin T: suppression of tumourigenicity 2, tryptophanyl-tRNA synthetase cytoplasmic, histone H2A Type 3, angiotensinogen, deltex-1, thrombospondin-4, ADAMTS-like protein 2, anthrax toxin receptor 1, and cathepsin D. N-terminal pro-hormone B-type natriuretic peptide and angiotensinogen showed the strongest associations [hazard ratio (95% confidence interval): 1.96 (1.17-3.40) and 0.66 (0.49-0.88), respectively]. The multivariable model yielded a c-index of 0.85 upon internal validation and c-indices up to 0.80 upon external validation. The c-index was higher than that of a model containing established risk factors (P = 0.021).

Conclusion: Nine serially measured proteins captured the most essential prognostic information for the occurrence of adverse events in patients with HFrEF, and provided incremental value for HF prognostication beyond established risk factors. These proteins could be used for dynamic, individual risk assessment in a prospective setting. These findings also illustrate the potential value of relatively 'novel' biomarkers for prognostication.

Clinical trial registration: https://clinicaltrials.gov/ct2/show/NCT01851538?term=nCT01851538&draw=2&rank=1 24.

Keywords: Elastic net; Heart failure; NT-proBNP; Prediction; Proteomics; Repeated measurements.

Graphical Abstract

Figure 1

Proteins with a significantly different average evolution between patients with and without the primary endpoint. The estimated average evolution (per year) of circulating proteins with a significantly different trajectory (FDR < 0.05 and |relative difference in slope| > 10%) is depicted separately in patients with the primary endpoint and those who remained endpoint free. Average evolutions are estimated using linear mixed-effect regression models and are adjusted for the MAGGIC risk score. The red box depicts the average evolution of proteins in patients who reached the study endpoint, and the blue box depicts the average evolutions in patients who remained endpoint free.

Figure 2

Average evolutions of the 10 proteins selected based on penalized regression. The average evolution of circulating proteins is depicted during the 2 years preceding a primary endpoint in patients with chronic heart failure who reached the study endpoint and last sample moment in patients who remained endpoint free. ‘Time zero’ is defined as the occurrence of the endpoint or censoring and is depicted on the right side of the x-axis; inherently to this representation, baseline sampling preceded this ‘time zero’. The solid red line depicts the average evolution of proteins in patients who reached the study endpoint, and the solid blue line depicts the average evolution in patients who remained endpoint free. The dashed lines represent the 95% confidence intervals.