Preclinical and clinical evidence for suppression of alcohol intake by apremilast

Abstract

Treatment options for alcohol use disorders (AUDs) have minimally advanced since 2004, while the annual deaths and economic toll have increased alarmingly. Phosphodiesterase type 4 (PDE4) is associated with alcohol and nicotine dependence. PDE4 inhibitors were identified as a potential AUD treatment using a bioinformatics approach. We prioritized a newer PDE4 inhibitor, apremilast, as ideal for repurposing (i.e., FDA approved for psoriasis, low incidence of adverse events, excellent safety profile) and tested it using multiple animal strains and models, as well as in a human phase IIa study. We found that apremilast reduced binge-like alcohol intake and behavioral measures of alcohol motivation in mouse models of genetic risk for drinking to intoxication. Apremilast also reduced excessive alcohol drinking in models of stress-facilitated drinking and alcohol dependence. Using site-directed drug infusions and electrophysiology, we uncovered that apremilast may act to lessen drinking in mice by increasing neural activity in the nucleus accumbens, a key brain region in the regulation of alcohol intake. Importantly, apremilast (90 mg/d) reduced excessive drinking in non-treatment-seeking individuals with AUD in a double-blind, placebo-controlled study. These results demonstrate that apremilast suppresses excessive alcohol drinking across the spectrum of AUD severity.

Keywords: Addiction; Clinical Trials; Drug therapy; Neuroscience.

Figure 1. Apremilast reduces binge-like drinking behavior and ethanol motivation in mice selectively bred for drinking to intoxication.

(A) Binge-like ethanol intake (g/kg/4 hours) for HDID-1 (n = 10–12/sex/apremilast dose; main effect of apremilast [F(2, 61) = 21.0, P < 0.0001], with no sex or sex × treatment interactions. Both doses of apremilast reduced ethanol (EtOH) intake in HDID-1 mice. (B) Blood alcohol levels (BALs, mg%) in HDID-1; main effect of apremilast [F(2, 64) = 9.73, P < 0.001]; both doses of apremilast reduced BALs compared with 0 mg/kg. (C) Average 4-hour ethanol intake over 6-week test (week 1, baseline; weeks 2–5, treatment; week 6, washout) for apremilast-treated HDID-1 mice (n = 10–12/sex/apremilast treatment); main effect of time [F(2, 78) = 5.68; P < 0.01] and a time × treatment interaction [F(2, 78) = 17.56; P < 0.0001]; 40 mg/kg reduced ethanol intake compared with baseline and washout intake was higher than baseline. (D) BALs (mg%) for end of week 5, 4-hour drinking; no effect of apremilast (2-tailed Student’s t test ). (E) Highest operant response ratio reached (breakpoint) during PR testing (marker of ethanol motivation) for iHDID-1 (n = 10/12/sex/apremilast treatment); main effect of treatment [F(2, 64) = 4.47; P < 0.05]; 40 mg/kg reduced breakpoint iHDID-1 mice. (F) Ethanol reinforcers earned during quinine-adulterated testing; main effect of apremilast treatment [F(1, 134) = 37.90; P < 0.0001], with no effect of quinine or apremilast × quinine interaction; 40 mg/kg apremilast reduced the number of reinforcers earned for iHDID-1 mice at all quinine concentrations tested. *P < 0.05; **P < 0.05; ***P < 0.001; ****P < 0.0001 by 2-way ANOVA followed by Dunnett’s post hoc test (A–C, E, and F).

Figure 2. Apremilast reduces binge-like drinking behavior through increasing excitability of D1, but not D2, MSNs.

(A) Relative NAc Pde4a gene expression for female HDID-1 mice (46–48/fluid group); significant effect of fluid type. Ethanol mice express higher levels of PDE4a. EtOH, ethanol. (B) Relative gene expression of NAc Pde4b; significant effect of fluid type. Ethanol mice express higher levels of PDE4b. (C) Ethanol intake (g/kg/2 hours) following intra-NAc apremilast infusions (0 or 2.2 μg/μL/side for male HDID-1 mice (n = 19–20/fluid group/infusion group) shows a significant effect of apremilast. (D) Blood alcohol levels (mg%); significant effect of apremilast. (E) Apremilast suppressed synaptic inhibition of NAc MSNs (n = 6–8/group). Inhibitory synaptic drive = frequency × current of spontaneous inhibitory postsynaptic currents (sIPSCs). § = main effect of treatment [F(1, 25) = 6.53, P < 0.05]. (F) Apremilast promoted synaptic excitation of NAc MSNs (n = 10–16/group). Excitatory synaptic drive = frequency × amplitude of spontaneous excitatory postsynaptic potentials (sEPSPs). §§ = main effect of treatment [F(1, 48) = 11.08, P < 0.01]. *P < 0.05, effect of treatment in D1 MSNs. (G) Apremilast promoted NAc output by lowering the threshold for MSN action potential (AP) firing (n = 19–24/group). § = main effect of treatment [F(1, 82) = 6.26, P < 0.05]. *P < 0.05, effect of treatment in D1 MSNs. V, vehicle (0.002% DMSO); A, apremilast (1 μM). Dashed lines indicate the AP threshold for each example trace. *P < 0.05, **P < 0.05, ***P < 0.001 by 2-tailed Student’s t test (A–D). Data in E–G were analyzed using 2- or 3-way ANOVA, with cell type (D1 or D2 MSN) and treatment condition (vehicle or apremilast) as between-groups factors. The effect of treatment within each MSN subtype was analyzed using Bonferroni’s multiple-comparison test.

Figure 3. Apremilast reduces dependence-induced escalations in ethanol drinking in C57BL/6J mice.

(A) Ethanol intake (g/kg/2 hours) for male C57BL/6J mice (n = 9–10/vapor group/stress group/apremilast treatment) during baseline and tests 1 and 2; main effect of group [F(3, 114) = 15.22; P < 0.001], phase [F(2, 114) = 60.80; P < 0.001], and a group × phase interaction [F(6, 228) = 13.25; P < 0.001]; CIE and CIE + FSS had higher intake compared with baseline and control (CTL) values (*P < 0.05) and their own baseline (^P < 0.05); CIE + FSS had higher intake in test 2 than all other groups and their own baseline (^#P < 0.05). (B) Ethanol intake (g/kg/2 hours) during test 3; main effect of group [F(3, 106) = 16.28; P < 0.001], apremilast [F(2, 106) = 21.83; P < 0.001], and a group × treatment interaction [F(6, 106) = 3.25; P < 0.01]; for mice that received vehicle, ethanol intake was higher for CIE mice compared with CTL mice (*P < 0.05) and higher for CIE + FSS compared with the 3 groups that also received vehicle (^#P < 0.05). CIE + FSS mice that received 20 mg/kg apremilast continued to drink more ethanol than CTL mice (*P < 0.05). However, this dose reduced ethanol intake compared with its vehicle condition group (^P < 0.05). The 40 mg/kg apremilast dose resulted in a significant decrease in ethanol intake in CIE and CIE + FSS mice compared with their vehicle equivalent (^P < 0.05). (C) Ethanol (EtOH) intake (g/kg/2 hours) for female and male C57BL/6J mice (n = 10/vapor group/apremilast treatment) following 3 weeks of CIE, main effect of vapor exposure, whereby ethanol vapor increased intake (*P < 0.05 by 1-way ANOVA). (D) Ethanol intake (g/kg/2 hours) during test week, main effect of treatment, 40 mg/kg (p.o.) reduced intake in ethanol vapor and air exposed mice. *P < 0.05, ***P < 0.001, ****P < 0.0001 by 2-way (A and B) or 1-way (C and D) ANOVA followed by Newman-Keuls post hoc test.

Figure 4. Apremilast reduces alcohol intake in non–treatment-seeking individuals with an AUD.

(A) Apremilast (90 mg/d) significantly (z = 2.24; P < 0.025) reduces the number of drinks per day relative to placebo in 51 non–treatment-seeking individuals with alcohol use disorder of moderate severity or greater. A negative binomial latent growth curve model was used to calculate an effect size for apremilast versus placebo in the decrease in drinks per day from baseline through 11 days of ad libitum drinking. This procedure generated 11-day change values totaling 2.74 drinks per day for apremilast and 0.48 for placebo, and yields a Cohen’s d value of 0.77, which can be interpreted as a “large” effect of apremilast on decreasing drinking. (B) Proportion of heavy drinking days (4+ for women, 5+ for men) is significantly (z = 2.17, P = 0.030) reduced for apremilast versus placebo. A latent logistic regression model (retransformed to units of proportion) was used to calculate differences in daily risk of heavy drinking through 11 days of treatment. Risk reduction between day 1 and day 11 was 0.39 for apremilast versus 0.05 for placebo; Cohen’s d for this difference was 0.26, a “small-medium” value.