Long-term donepezil treatment in 565 patients with Alzheimer’s disease (AD2000): randomized double-blind trial.
AD2000 Collaborative Group;
Commented by , 28 Jul 2004
Aim of the study
To assess whether donepezil produces worthwhile improvements in disability, dependency, BPSD, carers’ psychological wellbeing, or delay in institutionalization.
Method
Community living patients with mild to moderate (MMSE 10-26) AD as per DSM IV criteria with or without vascular dementia entered a 12-week period on donepezil (5mg/day) or placebo; they were then re-randomized to donepezil (5 to 10mg/day) or placebo up to 206 weeks through which three 4 to 6-week washouts took place.
Primary endpoints were entry to institutional care and progression of disability, defined as loss of 2/4 basic ADL or 6/11 instrumental ADL from the Bristol ADL scale. Secondary outcome measures included Bristol ADL total score, NPI total score, MMSE score, GHQ-30 score, death. An economic evaluation was also done using a caregiver activities scale. Assessments were done by a local nurse coordinator at a clinic or at home, by mail or by phone. Analysis was done by log-rank and multilevel models. Recruitment was planned for 3,000 patients within 2-3 years...
Results
566 patients entered the 12-week run-in and 55 dropped-out (12.8%), 486 entered the re-randomization for 48 weeks and 193 dropped-out (39.7%); after the first washout 194 patients were left for another 48 weeks, and a further 83 dropped-out (42.8%); in the third year of the study there were 51 patients left and 20 completed the year.
On donepezil the MMSE scores were 0.9 better than baseline in the first 12 weeks, then 0.8 points better than placebo (95% CI 0.5-1.2; p<0.0001) and Basic ADL scores 1 point better (0.5-1.6; p<0.0001) over the first 2 years. No significant difference was found in institutionalization (42% vs 44% at 3 years; p=0.4) or progression to disability (58% vs 59% at 3 years; p=0.4). There were no differences in the secondary clinical or economic outcomes. Discussion
This is a pragmatic study as defined by Tunis et al (JAMA 2003; 290: 1624-1632) which aims at developing high-quality scientific evidence to support clinical and health policy choices. Unfortunately this study fails in its design with an unexplainable set of washouts and an excessive drop out rate which renders any conclusion beyond 2 years unreliable, which is unfortunate because most of the institutionalization would be expected in the third year or beyond.
As highlighted by the accompanying editorial of Dr Lon S Schneider (Lancet 2004; 363: 2100-1), despite being underpowered to start with, it is remarkable that this study demonstrated statistically significant differences in favor of donepezil on both MMSE and Basic ADL in the first 2 years. On the other hand it is appropriate to highlight the lack of impact on institutional rate in this study, which may be the case in the area of England where the study was performed.
Of note is the reporting of this study in the lay press with headlines suggesting that all drugs for AD are useless. As already reported by Molnar et al (CMAJ 1999; 161: 393-395), medical advice columns for elderly readers may contain inappropriate or even potentially dangerous advice. It is hoped that the 8th International Conference on Alzheimer’s disease and Related Disorders in Philadelphia July 17-22 will set the record straight.