A human laboratory pilot study with baclofen in alcoholic individuals

Abstract

Preclinical and clinical studies show that the GABA(B) receptor agonist baclofen may represent a pharmacotherapy for alcohol dependence (AD). However, the mechanisms by which baclofen affects drinking are not well characterized; thus this pilot study investigated possible baclofen's biobehavioral mechanisms. The design was a double-blind controlled randomized human laboratory pilot study. Fourteen non-treatment seeking alcohol-dependent heavy drinking subjects received either baclofen 10mg t.i.d. or an active placebo (cyproheptadine 2mg t.i.d., to control for sedation) for a 7-day period. At day 8, participants performed an alcohol cue-reactivity (CR) followed by an alcohol self-administration (ASA). Additionally, we explored possible moderators that might guide future larger studies, i.e. anxiety, family history and onset of alcoholism, and D4 dopamine receptor (DRD4) and 5-HTTLPR polymorphisms. The main results were a significant effect of baclofen for increasing stimulation (p=.001) and sedation (p<.01). Furthermore, when drinking during the ASA and the 2 days before was analyzed as a composite variable, there was a significant effect of baclofen to reduce alcohol consumption (p<.01). As for the exploratory analyses, baclofen's effects to increase alcohol sedation and to reduce alcohol consumption were limited to those individuals with DRD4 ≥7 repeats (DRD4L). Yet, baclofen's effects on alcohol consumption were also moderated by 5-HTTLPR LL genotype. In conclusion, baclofen's ability to reduce alcohol drinking may be related to its effects on the biphasic effects of alcohol, but larger studies are needed to confirm these preliminary findings.

Fig. 1

Stimulation and sedation subscales of the Biphasic Alcohol Effects Scale (BAES) between the baclofen and active placebo groups during the alcohol “priming” (a) and the alcohol self-administration (ASA) experiment (b). After consuming the priming drink, there was a significant effect of baclofen, compared to active placebo, in the biphasic effects of alcohol, i.e. significant increased stimulation and sedation. Fig. 1a shows the biphasic effects of alcohol after the alcohol priming, but before the ad-libitum session; Fig. 1b shows the biphasic effects of alcohol during the ad-libitum session.

Fig. 2

Estimated means (and standard error bars) for the Standard Drink Units (SDUs) assessed 2 days prior and during the alcohol self-administration (ASA). In order to post-hoc analyze the data with a repeated measures ANCOVA with all 3 time points (ASA and self-reported alcohol drinking 1 day and 2 days before ASA), on the same metric, drinks during the ASA were “scaled up” to the standard drinking units (SDUs) for the 2 days before, by multiplying by eight (the sample mean for the 2 days before was 7.0, and the sample mean for the day before was 5.7; the average of these two values is 6.4; the sample mean for ASA drinks consumed was 0.8. 6.4/0.8=8.) These three dependent measures were entered into repeated measures ANCOVA with ADS as a covariate.

Fig. 3

Moderating effects for Standard drinking units (SDUs) consumed during 2 days prior and during the during the Alcohol Self-Administration (ASA). Moderation effects (Fig. a — DRD4 Status, Fig. b — 5-HTTLPR status) for Standard Drink Units (SDUs) consumed during the 2 days prior and during the laboratory session.