The in vivo potency of mazindol for binding to striatal dopamine transporters (DAT) was assessed by [123I]beta-CIT ([123I]2beta-carbomethoxy-3beta-(4-iodophenyl)tropane) single photon emission computed tomography (SPECT). Cocaine-dependent subjects (n = 12) underwent three SPECT scans; one before, between, and after subchronic (1 week) administration of 2 mg/day and 4 mg/day mazindol. For each scan, subjects were injected with [123I]beta-CIT and imaged 24 h later under equilibrium conditions. Results showed a statistically significant main effect of mazindol dose (df = 2, F = 10.30, P < 0.001, repeated measures ANOVA) in reducing the specific to non-displaceable equilibrium partition coefficient, V3'' (a measure proportional to DAT binding potential). Regression analysis of the logit transformed data enabled estimation of the 50% displacement dose of mazindol (ED50 = 30mg/day). These data suggest that low doses of mazindol (i.e., 2-4 mg) occupy a small percentage (i.e., < 25%) of DAT in human cocaine abusers and that much higher, potentially intolerable doses (i.e., > or = 30 mg/day) may be required to antagonize significantly cocaine binding in vivo.