By Will Boggs, MD
NEW YORK (Reuters Health) Apr 16 – The GABA-transaminase inhibitor vigabatrin failed to help patients with cocaine dependence in a randomized multisite study.
“Based on our clinical trial it seems unlikely that vigabatrin will ever be found to be effective in the treatment of cocaine dependence,” Dr. Eugene C. Somoza from the Cincinnati Veterans Affairs Medical Center and the University of Cincinnati in Ohio told Reuters Health.
Dr. Somoza and colleagues compared vigabatrin with placebo in a 12-week clinical trial with 186 participants at 11 U.S. sites. Everyone received weekly computerized cognitive behavioral therapy plus biweekly half-hour individual sessions with a counselor. One hundred forty-one patients completed the study, according to a report April 10th online in JAMA Psychiatry.
Cocaine abstinence was defined by two measures: self-report of cocaine use, and quantitative measurements of the cocaine metabolite benzoylecgonine and creatinine.
Overall, about 25% of urine samples were missing. Roughly 11% of samples in each group went missing during active treatment. Another 14% in each group were missing due to study dropout. And in both groups, only 55% of participants were more than 90% compliant, and only 66% were more than 70% compliant. In 39 patients the primary outcome could not be determined because of insufficient study participation.
With these shortcomings in mind, only seven of 92 vigabatrin participants met the primary endpoint of abstinence during the last two weeks of the 12-week treatment phase, compared with five of 94 placebo participants (7.6% vs 5.3%; p=0.67).
There were no significant differences between the groups in weekly fraction of cocaine use days, percentage of drug-free urine samples during weeks one through 13, or improvements in craving or global function.
“In a post hoc measure of vigabatrin in the urines of some participants it was estimated that the actual adherence may have been around 50%,” Dr. Somoza said. “Nevertheless, an analysis of just the adherent patients revealed that the medication was still not efficacious.”
Adverse event rates were similar in the two groups, except that placebo participants reported significantly more headaches.
Changes in visual acuity have been reported in association with vigabatrin use, but there were no clinically significant decreases in visual acuity during the study.
These results contrast with findings in a Mexican placebo-controlled trial in which 28% (n=14) of vigabatrin participants maintained full abstinence during the last three weeks of treatment, compared with 7.5% (n=4) of placebo participants (p=0.009).
The researchers suggest that their disappointing results may have been due to “weak efficacy of the drug” or significant nonadherence.
They also note that a third vigabatrin trial, funded by the National Institute on Drug Abuse, was designed “to more adequately address medication adherence (e.g., with once-daily dosage, observed dosing three times per week, and riboflavin level monitoring to measure compliance).”
Dr. Somoza said, “I know that this trial ended several months ago, but I do not know when the official results will be announced.” He said the results might be presented at the annual College on Problems of Drug Dependence (CPDD) meeting in San Diego in June.
“Cocaine is very different from other addictive agents, particularly because it lacks the intense withdrawal symptoms of other substances like opioids, alcohol, and benzodiazepines,” Dr. Somoza explained. “Thus, the cocaine withdrawal symptoms are so mild, non-specific, and often non-existent that they cannot be used to detect cocaine withdrawal. Note that in spite of this, DSM5 continues to assign as much importance to cocaine withdrawal symptoms as to the other addictive agents.”
“My feeling is that the scientific community and NIDA need to put a stronger emphasis on finding a better outcome variable for use in cocaine clinical trials to replace what is currently being used,” Dr. Somoza continued. “At the present time most clinical trials on cocaine employ only self-report measures of cocaine use. They generally claim that these self-reports are ‘validated’ by measuring benzoylecgonine (BE) levels in urine. However these BE measures are generally done qualitatively (and infrequently). Also, the ‘validation’ procedures tend to be quite feeble and seldom described, thereby making self-reports the de facto primary outcome variable.”
JAMA Psychiatry 2013.