Blocking the Ghrelin Receptor Did Not Reduce Alcohol Cravings but Changed Food Choices in People With Alcohol Use Disorder

This was a randomized, double-blind, placebo-controlled, within-subject crossover study. Forty-two individuals with alcohol use disorder received PF-5190457 100 mg twice daily or placebo in two counterbalanced stages. Assessments included: (1) an alcohol cue-reactivity experiment in a bar-like laboratory setting measuring craving; (2) a virtual food choice experiment in a cafeteria-like virtual reality environment measuring calorie selection; and (3) for a subset of 12 participants, a cue-reactivity task during functional MRI to measure brain activation patterns. The trial was registered on ClinicalTrials.gov (NCT02707055).

Alcohol use disorder is a major public health problem with limited pharmacological treatments. The ghrelin system has been a promising therapeutic target based on strong animal evidence. While this trial's negative result for alcohol craving is disappointing, it provides crucial human data that narrows the path forward — either the ghrelin system's role in human alcohol behavior differs from animal models, or different approaches to targeting it may be needed. The positive food-related finding also highlights the ghrelin-appetite connection as a viable therapeutic direction.

Ghrelin is a 28-amino-acid peptide hormone that links appetite, reward, and metabolic regulation. While animal studies consistently show that blocking ghrelin signaling reduces alcohol consumption, this is one of the first rigorous human trials testing this approach — and the results suggest the translation from animal models to humans may not be straightforward for alcohol outcomes. The food-related finding adds to the evidence supporting ghrelin system targeting for obesity and eating-related disorders, an area where GLP-1 agonists have already achieved clinical success.

The sample size was modest (42 enrolled, only 29 completers), limiting statistical power to detect smaller effects. The within-subject crossover design, while reducing individual variability, may introduce carryover effects. The fMRI subset was very small (n=12). Cue-reactivity in a laboratory setting may not reflect real-world alcohol craving or consumption. The study measured craving rather than actual drinking behavior. PF-5190457 is an inverse agonist/competitive antagonist — different ghrelin receptor targeting strategies might yield different results.