Heart failure with reduced ejection fraction (HFrEF) remains a major cause of morbidity and mortality. Molecular phenotyping may identify patient subgroups potentially benefitting from distinct therapeutic approaches. We developed a high-resolution mass spectrometry (MS)-based workflow enabling deep proteome analyses of endomyocardial biopsy (EMB) specimens.
We included 60 patients (18 years or older) with NYHA class II–III symptoms, undergoing EMB for evaluation of non-ischemic, non-valvular, non-hypertensive HFrEF. Left ventricular EMBs were obtained without procedure-related complications. Patients were followed for a median of 35 months for a composite endpoint of all-cause mortality, left ventricular assist device implantation, or heart transplantation. To generate deep proteomic data from small tissue samples (ca. 2 mg), we optimized tissue homogenization, protein extraction, and sample preparation for nano-flow liquid chromatography-coupled MS that was operated in data dependent acquisition (DDA) mode. We detected 6,612 proteins across all patients, constituting the largest heart tissue proteome dataset of HFrEF patients to date.
Using unsupervised hierarchical clustering with Ward’s method and Euclidean distance as a similarity metric, we divided the sample cohort into three patient subgroups. Applying functional enrichment cluster analysis, subgroup 1 (n=16) was characterized by increased inflammation and decreased cell survival; subgroup 2 (n=18) by improved cell survival and increased cell movement and angiogenesis; and subgroup 3 (n=26) by reduced inflammation and increased oxidative phosphorylation (all P<0.001). Notably, subgroup 1, despite having the highest LVEF (%) [(median (IQR) 32 (19–37) vs. 19 (11–23) and 24 (16–28) in subgroups 2 and 3; P=0.002], had the worst prognosis (50% incidence of the combined endpoint vs. 8% and 11%; P=0.001). In a multivariable Cox regression analysis adjusting for age and sex, the proteomic subgroups remained significantly associated with outcome (P=0.001).
In conclusion, deep myocardial proteome profiling identified HFrEF patient subgroups with dilative cardiomyopathy with distinct pathophysiological signatures and prognostic trajectories. Proteome-based patient stratification may ultimately guide individualized treatment strategies in heart failure.
