Occupancy of the kappa opioid receptor by naltrexone predicts reduction in drinking and craving

The efficacy of naltrexone to treat alcohol use disorder (AUD) is modest. A better understanding of the neurobiology underlying naltrexone effects could optimize treatments. We evaluated the occupancy of the kappa opioid receptor (KOR) by naltrexone measured with [¹¹C]-LY2795050 positron emission tomography (PET) as a predictor of response to naltrexone. Response to naltrexone was defined as the difference in craving and the difference between the number of drinks consumed during an alcohol drinking paradigm (ADP) before and after 1 week of supervised 100 mg daily oral naltrexone. Forty-four (14 F) nontreatment seeking heavy drinkers meeting criteria for AUD were enrolled. Participants drank 47 ± 16 drinks per week and were balanced in family history of alcoholism (FH, 26 positive). High KOR occupancy (92 ± 1%) was achieved. Occupancy was negatively associated with number of years drinking (YOD) in FH positive, but not FH negative, participants (t_3,42 = 4.00, p = 0.0003). Higher KOR occupancy by naltrexone was associated with higher alcohol craving during the ADP (F_1,81 = 4.88, p = 0.030). The reduction in drinking after naltrexone was negatively associated with KOR occupancy, with significant effects of FH status (t_1,43 = -2.08, p = 0.044). A logistic regression model including KOR occupancy, YOD, and FH variables achieved an 84% prediction accuracy for ≥50% reduction in drinking. These results confirm that naltrexone binds at the KOR site and suggest that KOR occupancy by naltrexone may be related to clinical response. Based on our results, we propose that differential affinities for the mu and KOR could explain why lower doses of naltrexone can have greater clinical efficacy.