Background: The current study aimed to compare the effects of different cholinesterase inhibitors on acetylcholinesterase (AChE) and butyrylcholinesterase (BuChE) activities and protein levels, in the cerebrospinal fluid (CSF) of Alzheimer disease (AD) patients. Methods and Findings: AD patients aged 50 – 85 years were randomized to open-label treatment with oral rivastigmine, donepezil or galantamine for 13 weeks. AChE and BuChE activities were assayed by Ellmans colorimetric method. Protein levels were assessed by enzyme-linked immunosorbent assay (ELISA). Primary analyses were based on the Completer population (randomized patients who completed Week 13 assessments). 63 patients were randomized to treatment. Rivastigmine was associated with decreased AChE activity by 42.6% and decreased AChE protein levels by 9.3%, and decreased BuChE activity by 45.6% and decreased BuChE protein levels by 21.8%. Galantamine decreased AChE activity by 2.1% and BuChE activity by 0.5%, but increased AChE protein levels by 51.2% and BuChE protein levels by 10.5%. Donepezil increased AChE and BuChE activities by 11.8% and 2.8%, respectively. Donepezil caused a 215.2% increase in AChE and 0.4% increase in BuChE protein levels. Changes in mean AChE-Readthrough/Synaptic ratios, which might reflect underlying neurodegenerative processes, were 1.4, 0.6, and 0.4 for rivastigmine, donepezil and galantamine, respectively. Conclusion: The findings suggest pharmacologically-induced differences between rivastigmine, donepezil and galantamine. Rivastigmine provides sustained inhibition of AChE and BuChE, while donepezil and galantamine do not inhibit BuChE and are associated with increases in CSF AChE protein levels. The clinical implications require evaluation.