Acyl ghrelin infusion increases circulating growth hormone in patients with heart failure and reduced ejection fraction

Ghrelin is a 28 amino-acid anabolic peptide hormone released from the stomach in response to fasting and weight loss. It stimulates appetite and release of growth hormone (GH) via the GH secretagogue receptor 1a (GHSR-1a) in healthy individuals. Both ghrelin and the GHSR1a are expressed in the myocardium. When acylated (activated) ghrelin binds to GHSR1a it activates signalling pathways associated with cardiomyocyte survival, contractility and suppression of inflammation¹ suggesting both GH dependent and independent mechanisms.² In patients with heart failure (HF), GH is dysregulated, with relative GH deficiency and GH resistance.³

In the Karolinska Acyl Ghrelin Trial (ClinicalTrials.gov NCT05277415), a recent double-blind randomized trial in HF with reduced ejection fraction (HFrEF), intravenous acyl ghrelin but not placebo increased cardiac output by 28%.⁴ We assessed acyl ghrelin versus placebo on GH release and the pharmacodynamics of acyl ghrelin treatment on the GH response in this study. In brief, 31 patients with chronic HFrEF were randomized to human acyl ghrelin (0.1 μg/kg/min; n = 15) or placebo (NaCl; n = 16) intravenously over 120 min. Blood sampling was performed prior to (T0) and after 60 (T60) and 120 (T120) min infusion, and 30 min after stopping infusion (T150) (detailed methods in online supplementary Appendix S1).

In patients randomized to acyl ghrelin, high GH response was defined as above and low GH response as equal to or below median of the area under the curve for GH (AUC_GH). Baseline characteristics according to GH response (high vs. low) are expressed as median and quartiles [Q1-Q3] or number and percentages (%). GH responses according to timepoints were analysed by cross-correlation. Associations between baseline characteristics and below/above median AUC_GH were assessed by univariable logistic regression. Difference in GH and insulin concentration between intervention/placebo groups and below/above median AUC_GH was assessed by repeated measures analysis of variance. The Karolinska Acyl Ghrelin Trial was approved by the regional ethics committee and complies with the Declaration of Helsinki. All participants provided written informed consent.

At baseline, fasting GH did not differ between intervention (n = 15) and placebo groups (n = 15; one patient excluded due to premature interruption of placebo infusion) (0.4 [0.2–1.5] vs. 0.3 [0.1–1.2] μg/L; p = 0.422). Displayed in Figure 1A, GH increased rapidly during infusion in the acyl ghrelin-treated group (T60: 26 [20–38] μg/L), began to decline even before stopping infusion (T120: 9.1 [6.9–13] μg/L), and declined further after stopping infusion (T150: 2.4 [1.7–4.1] μg/L) compared to placebo (T60: 0.5 [0.3–0.9]; T120: 0.6 [0.4–0.9]; T150: 0.7 [0.3–1.2] μg/L; p acyl ghrelin vs. placebo <0.001). GH responses for individual patients are presented in Figure 1B. There was a correlation between concentrations of acyl ghrelin achieved and concentrations of GH during infusion (cross-correlation 0.75). Online supplementary Table S1 displays baseline characteristics in acyl ghrelin-treated patients divided according to response with GH concentrations by AUC (AUC_GH): high/above (n = 7) or low/below (n = 8) median AUC_GH. Acyl ghrelin-treated patients with a higher GH response more often had a HF aetiology of dilated cardiomyopathy and numerically higher baseline heart rate, higher baseline E/e′ and higher baseline N-terminal pro-B-type natriuretic peptide. Patients with lower GH response all had ischaemic heart disease and more often history of malignancies, higher waist circumference and numerically more frequently diabetes and insulin resistance. Associations between selected baseline characteristics and GH response in acyl ghrelin-treated patients are depicted in online supplementary Figure S1. Insulin concentrations during acyl ghrelin infusion after standardized breakfast were numerically lower in patients with high GH response (n = 7) (T0: 22 [20–36]; T60: 6.5 [6.2–12]; T120: 8.1 [5.7–14]; T150: 9.1 [7.6–11] μg/L) compared to low GH response (n = 6, excluding 2 patients receiving insulin) (T0: 54 [31–60]; T60: 22 [9.1–33]; T120: 8.4 [8.0–25]; T150: 16 [8.7–25] μg/L; p = 0.377.

In the present randomized trial, we previously reported that in patients with HFrEF, intravenous acyl ghrelin significantly increased cardiac output.⁴ Here we report that acyl ghrelin caused an appropriate increase in circulating GH. This is consistent with a previous study in HFrEF, where intravenous ghrelin increased GH and cardiac index.⁵ We also report a correlation between acyl ghrelin and GH concentrations. GH decreased between 60 and 120 min, consistent with GH release occurring in a pulsatile rhythm. Taken together these data suggest that in patients with HFrEF, treatment with acyl ghrelin results in an appropriate rapid GH response as well as a more prolonged increase in cardiac output.

The GH response was potentially greater with more advanced HF. Myocardial GHSR expression may be increased in HF and more so with more advanced HF, and lower after heart transplantation.^{6, 7} In these patients with end-stage HF, GHSR and tissue ghrelin expression were both correlated with lower left ventricular ejection fraction and higher tissue B-type natriuretic peptide.⁷ Circulating total ghrelin,⁸ acyl ghrelin⁹ and GH³ are elevated in advanced HFrEF. Additionally, GHSRs may be suppressed in presence of impaired glucose tolerance.¹⁰ Taken together, this suggests that elevated ghrelin may be an adaptive compensatory response in HF, similarly to natriuretic peptides, which are adaptive, and differently from catecholamines, which are maladaptive.

Our study is limited by a modest sample size. Nevertheless, it seems reasonable to conclude that patients with HFrEF treated with acyl ghrelin respond appropriately with both increased circulating GH and increased cardiac output. This may be helpful for pharmacodynamic and pharmacokinetic assessments in future trials of treatment with ghrelin, ghrelin analogues, or ghrelin receptor agonists.

Conflict of interest: C.H. reports consulting fees from Novartis, Roche Diagnostics and AnaCardio, research grants from Bayer and speaker and honoraria from AstraZeneca and Novartis; supported by the Swedish Research Council [grant 20 180 899]. M.S. reports consulting fees from AnaCardio; speaker's honoraria from Orion Pharma; research grants from Swedish Heart and Lung Foundation. T.T. reports speaker's honoraria from Orion Pharma, Boehringer Ingelheim, Novartis. U.L.F. reports consulting fees from Orion Pharma and AnaCardio. P.M.H. reports consulting fees from Pharmanovia, Celltrion Healthcare and NV Rose; supported by the Swedish Research Council [grant 2017–02243]. D.C.A. reports consulting fees from AnaCardio; speaker's honoraria from Pfizer; supported by grants from the Heart Lung Foundation, Swedish Society for Medical Research (SSMF), Swedish Medical Society, and Harald and Greta Jeansson memorial foundation. L.H.L. reports grants from AstraZeneca, Vifor, Boston Scientific, Boehringer Ingelheim, Novartis, MSD; consulting fees from Vifor, AstraZeneca, Bayer, Pharmacosmos, MSD, MedScape, Sanofi, Lexicon, Myokardia, Boehringer Ingelheim, Servier, Edwards Life Sciences, Alleviant; speaker's honoraria from Abbott, OrionPharma, MedScape, Radcliffe, AstraZeneca, Novartis, Boehringer Ingelheim, Bayer; Patent: AnaCardio; stock ownership: AnaCardio. All other authors have nothing to disclose.