Efficacy of rivastigmine in subjects with moderately severe Alzheimer’s disease
Burns A, Spiegel R and Quarg P;
Commented by , 20 Apr 2004
Background
Cholinesterase (ChE) inhibitors are primarily used in the treatment of mild-moderate Alzheimer’s disease (AD), based on evidence from extensive drug trials in this patient group. Operational criteria for defining "mild", "moderate" and "severe" are unsophisticated.
Aims and Objectives
To evaluate retrospectively, using pooled data from three large placebo-controlled trials, the efficacy and safety of rivastigmine in more advanced AD.
Methods
A retrospective statistical analysis was performed on pooled data from 3 six-month, multicentre, double-blind, placebo-controlled parallel-group studies; these similar studies had previously been published in peer-reviewed journals in 1998 and 1999. Subjects for this analysis were selected on baseline disease severity according to Mini Mental State Examination (MMSE) score of 10-12. This corresponded to a Global Deterioration Scale (GDS) score of 4 and above (“moderately severe to severe dementia”) in the majority of cases.
Efficacy Assessment: cognitive performance had been measured with the ADAS-Cog Scale; activities of daily living with the Progressive Deterioration Scale (PDS); behavioural problems with the BEHAVE-AD scale.
Safety/Tolerability Assessments: stratified according to the severity of dementia at baseline and treatment administered; the relative risk of discontinuing from the study due to adverse events was calculated.
Statistics: raw data on the pooled data set were analysed, using observed cases and intent-to-treat, last-observation-carried-forward populations. Analysis of mean change from baseline in ADAS-Cog and PDS scores used two-way ANOVA.
Results
117 subjects were identified, of whom 62 received rivastigmine 6 – 12 mg per day, 55 received placebo. The demographic and clinical features of the groups were well-matched at baseline, and all three studies contributed similar subject numbers.
After 26 weeks, there was a small improvement (0.2 points) in ADAS-Cog score in the treatment group relative to baseline, compared to a large mean drop of 6.3 points in subjects receiving placebo. Similar trends were found in PDS and BEHAVE-AD scores.
Rivastigmine was well tolerated. Gastrointestinal problems were the most common adverse events. Relative risk analysers suggested that dropping out due to adverse events was related to the level of baseline cognitive impairment.
Discussion
This is a retrospective study based on small numbers, on a selected population that may have not been representative of clinical practice. However, the treatment effect of rivastigmine was consistent across the three studies, and striking in its magnitude; 46 % of rivastigmine-treated subjects either improved or showed no deterioration on ADAS-Cog, compared to 9 % of placebo subjects.
In early trials, subjects with baseline MMSE scores <10 were excluded, though it is possible that cognitive enhancers are equally or more effective in this group. It is not clear whether this is a class effect or specific to rivastigmine.
Rivastigmine seems to be similarly well tolerated in subjects with advanced dementia as it is in subjects with milder disease.