Deregulation of APP metabolism is considered to be a key pathogenic event in Alzheimer's disease. Data from cell cultures indicate that the secretion of A beta(1-42) might be inhibited by cholinesterase inhibitors, possibly via M1 receptors stimulation. Treatment with tacrine, a dual acetyl- and butyrylcholinesterase inhibitor, had no significant effect on mean plasma A beta species concentrations. However, a correlation was observed between higher drug concentrations and lower A beta levels that might indicate an effect on APP metabolism with an increased alpha-cleavage. A beta(1-40) and A beta(1-42) levels were measured in the plasma of 28 AD subjects by means of a commercially available ELISA before rivastigmine treatment and at week 2 after the first dose of the drug (3 mg/day) had been administered. Treatment with rivastigmine exhibited a significant effect on mean plasma concentrations of A beta(1-42) (mean difference 7.8 +/- 8.4, t=-4.9, pmean difference 7.8 +/- 8.4, t=-4.9, p<0.001) with a negative correlation with the patients age (Pearson R=-0.40, p=0.035). No significant effect on plasma A(beta 1-40) was observed. The observed increase of mean levels of plasma A beta(1-42) after rivastigmine treatment might indicate an effect of the drug on A beta metabolism, mobilization of A beta(1-42) from deposits in the affected brain areas and a consecutive A beta(1-42) brain-to-plasma efflux. The negative correlation between A beta(1-42) plasma levels changes and age may be a sign of impairment of this process in the older patients. A large individual variation of the observed response, however excludes drawing definite conclusions. Whether those subjects who respond to rivastigmine in terms of A beta(1-42) plasma levels changes also respond clinically needs to be established.