Clinical Psychoneuroendocrinology and Neuropsychopharmacology (CPN)

National Institute On Drug Abuse

Investigators

Linked publications, trials & patents

Abstract

There has been growing interested in the role of the glucagon-like peptide-1 (GLP-1) system in addictive behaviors and its potential as a pharmacotherapeutic target (Bruns, Tressler, et al., Pharmacol Res 2024). In a series of studies aiming to understand the link between alcohol use and the GLP-1 system, we found that A) genetic variants on the GLP-1 receptor (GLP-1R) gene, that result in amino-acid substitutions and putative changes in the GLP-1R structure/function, are associated with AUD risk/severity and related brain functional connectivity; B) alcohol intake, with different doses and routes of administration, reduces peripheral GLP-1 levels; and C) GLP-1R gene expression in postmortem brain tissues from individuals with AUD is higher than that in controls (Suchankova, et. al., Transl Psychiatry 2015; Farokhnia, Fede et. al., Sci Rep 2022; Farokhnia, et. al., Addict Biol 2022; Molina-Castro, et. al., Addict Biol 2024). On the behavioral pharmacology side, we previously showed that liraglutide and semaglutide, two long-acting GLP-1R agonists (GLP-1RAs) clinically approved for type 2 diabetes mellitus and obesity, significantly reduced voluntary alcohol intake/preference in male rats, binge-like alcohol drinking in male and female mice, and binge-like and dependence-associated alcohol drinking in male and female rats (Marty, Farokhnia, et. al., Front Neurosci 2020; Chuong, Farokhnia, Khom, et. al., JCI Insight 2023). To translate these findings to humans, we conducted a pharmacoepidemiological cohort study, using electronic health records (EHR) data from the Department of Veterans Affairs, and found that receipt of GLP-1RAs agonists for any indication was associated with significantly greater reduction in alcohol use severity (measured via alcohol use disorder identification test – consumption (AUDIT-C)), compared to propensity-score matched controls. Conversely, receipt of dipeptidyl peptidase-4 inhibitors (DPP-4Is), another class of GLP-1 therapies that are FDA-approved for the treatment type 2 diabetes mellitus, was not associated with changes in AUDIT-C scores. We back translated the latter finding by testing two DPP-4Is, linagliptin (peripherally restricted) and omarigliptin (brain penetrant), which showed no impact on binge-like alcohol drinking in mice or dependence-associated alcohol drinking in rats, while expectedly reducing blood glucose levels (Farokhnia, et. al., JCI 2025). In two additional pharmacoepidemiological studies, using data from Kaiser Permanente North California and the NIH All of Us program, we have replicated the finding that receipt of GLP-1RAs is associated with significant reductions in drinks per week and AUDIT-C scores, compared to propensity-score matched controls (Palzes, et. al., under review; Tyndall, et. al., under review). At this juncture, RCTs are needed to test the safety and efficacy of GLP-1RAs for ASUDs (Leggio, et al., Nature Medicine 2023). In a 20-week, outpatient, randomized, between-subject, double-blind, placebo-controlled trial, we are testing semaglutide (up to 2.4 mg/week) in individuals AUD (recruitment ongoing). Outcomes include real-world safety (e.g., number/severity of adverse events) and efficacy (e.g., number of drinks per drinking day) measures, as well as experimental medicine procedures (e.g., cue-reactivity, brain fMRI). We are also examining the impact of GLP-1RAs, alone and combined with other agents, on opioid use and related outcomes, using validated rodent models such as fentanyl vapor self-administration in mice and intravenous oxycodone self-administration in rats (Ahearn, et. al., in preparation). Blocking the ghrelin receptor (GHSR) represents another novel pharmacotherapeutic approach for AUD. In a Phase 1b clinical trial, we previously showed that the GHSR blocker PF-5190457 is safe/tolerable, does not influence alcohol pharmacokinetics, and may reduce cue-elicited alcohol craving in non-treatment-seeking individuals with AUD (Lee et al., Mol Psychiatry 2020). We completed a Phase 2a clinical trial in treatment-seeking individuals with AUD and found that while PF-5190457 did not reduce cue-elicited alcohol craving nor fMRI-based brain activity, it did reduce the number of calories selected in a virtual buffet (Faulkner, et al., JCI Insight 2024). We also examined pharmacokinetic and pharmacodynamic parameters in this study and found significant reductions in growth hormone (GH) and liver-expressed antimicrobial peptide 2 (LEAP2) concentrations under PF-5190457 vs placebo, with no other impact on endocrine or immune systems (Tyler, et. al., under review). LEAP2 has recently been identified as an endogenous GHSR antagonist that influences metabolic and reward-related pathways. In a series of secondary analyses of human studies, we found that A) LEAP2 concentrations negatively correlated with priming- and cue-elicited alcohol craving, and B) oral alcohol administration reduced LEAP2 concentrations, an effect that was attenuated by co-administration of PF-5190457. Consistently, intraperitoneal alcohol administration reduced LEAP2 concentrations in wild-type but not GHSR-knockout rats (Leko, et. al., under review). In addition to GHSR blockade, the ghrelin system could also be suppressed by inhibiting Ghrelin-O-Acyl-Transferase (GOAT), the enzyme responsible for ghrelin acylation/activation. We completed a Phase 1b/2a clinical trial to test the safety and efficacy of GLWL-01, a GOAT inhibitor, in non-treatment-seeking individuals with AUD (data analysis ongoing). We have also examined non-pharmacological correlates of the gut-brain axis in relation to addictive behaviors. In a case-control survey study focused on bariatric surgery, we found that problematic alcohol, opioid, amphetamine, and cannabis use were reduced, and participants expressed more "guilt" associated with alcohol and other substance use post-surgery compared to pre-surgery (Speed, et. al., Drug Alcohol Depend 2025). In another case-control study focused on gut microbiome/metabolome, we found that metabolites derived from the lipid super-pathway were more abundant among recently abstinent individuals with AUD compared to currently drinking individuals with AUD and controls without AUD. The abstinent group was most clinically different from currently drinking and control groups with respect to their gut microbiome and metabolome (Piacentino, Vizioli, Barb, et al., PLoS One 2024). We have also used this rich dataset to answer other clinically relevant questions such as consumption of ultra-processed foods in the context of alcohol use, circulating immune and endocrine markers in currently drinking and abstinent individuals with AUD (Tyler et. al., Addict Biol 2025), and correlates of alcohol cue-reactivity in these populations (Vizioli, et. al., in preparation). In addition to the gut-brain axis, it is also important to study the role of other peripheral-central pathways in addictions. One such pathway is the aldosterone/mineralocorticoid receptor (MR) system with growing evidence indicating its role in alcohol seeking and consummatory behavior and its potential as a novel pharmacotherapeutic target for AUD. We previously showed that the prototypic MR antagonist spironolactone reduced binge-like alcohol drinking in mice and dependence-associated alcohol drinking in rats; in two pharmacoepidemiological cohort studies, spironolactone prescription for any indication was associated with reductions in alcohol use (Palzes, et. al., Neuropsychopharmacology 2021; Farokhnia, Rentsch, Chuong, McGinn. Et. al., Mol Psychiatry 2022). In another pharmacoepidemiological study using data from the TriNetX platform, we took a target trial emulation approach and found that spironolactone prescription was associated with higher rates of AUD remission and lower rates of alcohol intoxication compared with other classes of antihypertensive drugs such as angiotensin-converting-enzyme inhibitors, calcium channel blockers, and beta blockers (Loften, Wang, et. al., in preparation). We started a randomized, double-blind, placebo-controlled, ascending dose, Phase 1b study with spironolactone in individuals with AUD (recruitment ongoing). Given that spironolactone is a nonselective MR antagonist with additional activity on other systems such as androgen and progesterone receptors, to better understand the specific role of MR in alcohol use behaviors, we tested the effects of selective MR antagonists (finerenone, eplerenone, and PF-03882845), spironolactone’s active metabolites with MR antagonism properties (7αTMS and canrenone), selective MR agonists (aldosterone and fludrocortisone), an androgen receptor antagonist (darolutamide), and a progesterone receptor agonist (dydrogesterone), on binge-like alcohol drinking in mice. The selective MR antagonists reduced alcohol intake, while other compounds had no effect. Eplerenone and PF-03882845, but not finerenone, also reduced the intake of sweet solution without alcohol, pointing to finerenone as the preferred compound for clinical testing (Wilkinson, et. al., under review). We are also examining the impact of MR antagonists on opioid use and related outcomes using validated rodent models and a target trial emulation pharmacoepidemiological study (Wang, Whiting, Bruns, et. al., in preparation). With a holistic view in addiction treatment which requires combining pharmacotherapy with evidence-based behavioral and psychosocial interventions, we have started to study the impact of environmental enrichment (EE) on alcohol-related outcomes. EE refers to the experience of an organism upon interaction with rich, complex, multisensory, and often novel stimuli in the environment. Aspects of EE have shown therapeutic potential for neuropsychiatric conditions including addiction. We designed and started a pilot clinical trial to study the effects of cognitive and sensorimotor EE in people with AUD (recruitment ongoing). Following a within-subjects counterbalanced design, each participant undergoes an EE phase (including virtual reality, video games, computer games, books, arts and crafts, puzzles, brain games, music, fragrance, and other stimuli and activities) and a control condition. The primary outcome is self-reported alcohol use via ecological momentary assessment. Secondary outcomes include psychosocial variables associated with AUD, alcohol craving during a cue-reactivity task, alcohol demand during an alcohol purchase task, and subjective responses during an alcohol challenge test.