Growth differentiation factor‐15 and the effect of empagliflozin in heart failure: Findings from the EMPEROR program

Aims Growth differentiation factor‐15 (GDF‐15) is upregulated in part in response to cardiomyocyte stretch and stress, and it exerts a protective role that is mediated by its action to suppress signalling through insulin‐like growth factor (IGF) and enhance signalling through adenosine monophosphate‐activated protein kinase (AMPK). Sodium–glucose cotransporter 2 (SGLT2) inhibitors improve outcomes in heart failure, which has been experimentally linked to AMPK. This study aimed at evaluating the associations of GDF‐15 with baseline characteristics, the prognostic significance of GDF‐15, and the effect of empagliflozin on GDF‐15 in patients with heart failure with a reduced and preserved ejection fraction. Methods and results Growth differentiation factor‐15 was determined in serum samples from the EMPEROR‐Reduced and EMPEROR‐Preserved trials. Cox regression and mixed models for repeated measures were used to study the association with outcomes and the effect of empagliflozin on GDF‐15, respectively. We studied 1124 patients (560 placebo and 564 empagliflozin) with median GDF‐15 levels at baseline of 2442 (interquartile range 1603–3780) pg/ml. Patients with higher GDF‐15 levels were typically older men with more severe symptoms, higher N‐terminal pro‐B‐type natriuretic peptide levels, worse kidney function and who were prescribed metformin. Baseline levels of GDF‐15 were well correlated with levels of IGF‐binding protein 7 (rho = 0.64). Higher levels of GDF‐15 were independently associated with an increased risk of cardiovascular death, heart failure hospitalizations, and worse kidney outcomes. When considered as a continuous variable, for each doubling in GDF‐15, the adjusted hazard ratio for cardiovascular death or heart failure hospitalization was 1.40 (95% confidence interval 1.15–1.71; p < 0.001). The relative effect of empagliflozin on cardiovascular death and hospitalization for heart failure was most pronounced in patients with higher baseline levels of GDF‐15 (interaction p ‐trend = 0.031). At week 52, when compared with placebo, empagliflozin increased GDF‐15 by an additional 8% ( p = 0.020), an effect that was primarily seen in patients not receiving metformin, a known AMPK activator. Conclusions Growth differentiation factor‐15 is a marker of worse heart failure severity, is an independent predictor of major heart failure outcomes and may be associated with more pronounced benefits of empagliflozin. GDF‐15 is increased among metformin users, and empagliflozin was associated with an increase in GDF‐15 levels, primarily in patients not receiving metformin.