In a clinical trial of tacrine in patients with Alzheimer's disease, improvement, measured by changes on both total (ADAST) and the cognitive sub-component (ADASC) of the Alzheimer's Disease Assessment Scale (ADAS) over a 6-week period, correlated significantly with plasma levels of tacrine. Change on the ADAS (both total and the cognitive subcomponent) also correlated significantly with three monohydroxylated metabolites of tacrine 1, hydroxy-tacrine (1-OH-THA), 2, hydroxy-tacrine (2-OH-THA), and 4, hydroxy-tacrine (4-OH-THA). However, changes of the relatively small non-cognitive component of the ADAS were not significantly correlated with plasma levels of tacrine. Multiple correlational analysis revealed that the combined influences of these metabolites were no greater than the effects of tacrine alone in ameliorating the symptoms of Alzheimer's disease. An alternative measure of cognitive performance, the Mini Mental State Exam (MMSE) did not correlate significantly with plasma concentrations of tacrine or its metabolites. Tacrine and these metabolites are bound to plasma proteins. In 10 patients with Alzheimer's disease receiving tacrine, the percentage of tacrine, 1-OH-THA, 2-OH-THA, and 4-OH-THA, that was free in plasma was found to be 19.7 ± 3.3, 51.3 ± 7.7, 40.7 ± 6.9, and 42.4 ± 6.3 (mean ± SD), respectively. © Wiley-Liss, Inc.