An interim analysis of 41 evaluable patients compared gepirone to placebo treatment in a randomized, double-blind, 12-week study of cocaine dependence without opiate abuse. The response to gepirone at a mean dose of 16.25 mg/day did not differ from placebo by measures of time in study, positive urine cocaine screens (greater than 6 weeks), Clinical Global Impressions (CGI) Global Improvements Scale, Cocaine Craving Scale (CCS), Quantitative Cocaine Inventory (QCI), Addiction Severity Index (ASI), Global Assessment Scale (GAS), Hamilton Rating Scale for Depression (HAM-D), and Hamilton Anxiety Scale (HAM-A). Both treatment groups showed similar modest, average improvements during the study in all treatment measures. Adverse events were not treatment limiting. The following demographic and study measures suggested favorable trends for study outcomes: older age, divorced status, higher pre-treatment cocaine use, lower CCS scores, and lower self-reports of cocaine use according to QCI.