Objective: To determine the value of continued donepezil treatment in patients with Alzheimer's disease for whom clinical benefit was initially judged to be uncertain.
Methods: The study consisted of three phases: (i) a 12- to 24-week, pre-randomisation, open-label donepezil-treatment phase; (ii) a 12-week, randomised, double-blind, placebo-controlled phase; and (iii) a 12-week, single-blind (i.e. patient-blind) donepezil-treatment phase. Patients with mild to moderate Alzheimer's disease received open-label treatment with donepezil (5 mg/day for 4 weeks, then 10 mg/day for the remainder of the phase) for 12-24 weeks. Patients who exhibited a decline or no change from baseline on the Mini-Mental State Examination (MMSE) and whose physician was not sufficiently certain of clinical benefit to warrant continued treatment were randomised into the double-blind phase in which patients received 12 weeks of treatment with donepezil (10 mg/day) or placebo. At the end of the double-blind phase, donepezil-treated patients continued to receive donepezil, while placebo-treated patients were rechallenged with donepezil, in a 12-week single-blind phase. Patients were assessed at the start of the double-blind phase and at weeks 6 and 12 of this phase, and at the end of the single-blind phase.
Results: Six hundred and nineteen patients completed the open-label phase; 69% showed clear clinical benefit and 31% showed uncertain benefit. 202 patients were randomised to continued donepezil treatment (n = 99) or placebo (n = 103). Differences in favour of continued donepezil versus placebo were observed in cognition and behaviour. In addition, there was a non-significant trend favouring donepezil in activities of daily living (ADL) [week 12 observed case mean treatment differences: MMSE, 1.13 (p = 0.02); Alzheimer's Disease Assessment Scale - cognitive subscale, 0.57 (p = 0.5); the Neuropsychiatric Inventory, -3.16 (p = 0.02); Disability Assessment for Dementia scale, 3.67 (p = 0.1)].
Conclusion: Most patients showed clear clinical benefit during initial donepezil treatment. Among patients for whom clinical benefit was uncertain, improvement in cognition and behaviour were observed for those who continued donepezil treatment compared with the group switched to placebo. Initial decline or stabilisation does not necessarily indicate a lack of efficacy in Alzheimer's disease, and the decision to discontinue treatment should be based on an evaluation of all domains (cognition, behaviour and ADL) and performed at several timepoints.