Research strategies to assess pharmacotherapies for alcoholism

1. The development and implementation of standardized basic and clinical research strategies is necessary for the discovery and testing of new and effective pharmacotherapies for alcohol dependence. While many animal studies have indicated the involvement of central serotonin, endogenous opioids and dopamine, their exact mechanisms of action have not been fully discerned. One problem is the use of various animal models, making it difficult to compare or replicate results from different laboratories. 2. Once a drug appears to decrease alcohol intake in animals, assessing its effects in humans can be even more challenging. 3. Most drugs were developed and may be even registered for another indication, so Phase I trials must be repeated in heavy alcohol users in case the drug has behavioural, pharmacokinetic or pharmacodynamic interactions with alcohol. 4. Phase II trials, assessing the effects of the drug on alcohol intake, are usually radically different from the animal models: most are long (> 6 months), use varied designs, and accept only abstinence as successful outcome. Moderately dependent alcoholics, who comprise the majority of patients, prefer a goal of nonhazardous drinking. 5. In Phase III multicentre trials, especially international ones, it is almost impossible to control for potentially confounding variables, such as patient characteristics and concomitant psychosocial treatment. 6. Also, research in pharmacotherapies for alcoholism is influenced by external factors such as sources of funding and other resources. Thus, this methodical progression from Phase I to Phase III clinical trials is often abbreviated. 7. The authors developed a brief early Phase II clinical paradigm to assess pharmacotherapies for moderately dependent alcoholics, and conducted 3 studies with the serotonin uptake inhibitors, citalopram and fluoxetine, and a serotonin antagonist, ritanserin. This paradigm, which includes a short (1 to 2 weeks) outpatient phase and two experimental drinking sessions, measures alcohol intake, desire to drink, alcohol's subjective effects, and intoxication. While it may have limited relevance to a goal of abstinence, it is efficient and economical to screen drug effects on alcohol intake and suggest mechanisms (desire to drink, subjective effects). It is also the only human model concordant with results in animals. 8. Development of new pharmacotherapies for alcoholism may benefit from acceptance of both abstinence and moderation of drinking as acceptable treatment goals.