Cholinesterase Inhibitors

Donepezil

The FDA approval of donepezil marked the beginning of the era where AD was perceived as a treatable disease. The initial and subsequent clinical trials have demonstrated the safety of this agent, and fail to show any evidence of hepatotoxicity or transaminase elevations. Due to the long serum half-life, donepezil is dosed once daily.

The safety and efficacy of donepezil in mild-moderate AD was examined in several pivotal trials.^10,11In a 30 week, double-blind placebo-controlled, multicenter trial, 473 patients with MMSE scores ranging from 10-26 were randomized to receive placebo (N=162), donepezil 5mg/day (N=154), or donepezil 10mg/day (N=157) for 24 weeks followed by a 6 week washout period to examine duration of effect.¹¹The 10mg treatment arm received 5mg of donepezil per day for one week, which was then increased to 10mg daily.

Effects of Donepezil on Cognition: ADAS-Cog11

The percentages of patients completing the study were placebo, 80%, donepezil 5mg, 85%, and donepezil 10mg, 68%. These results sharply contrasted with the retention rates observed in the tacrine trials. Adverse effects were generally infrequent and the incidence of peripheral cholinergic adverse effects reached statistical significance only in the 10mg group for fatigue, muscle cramps, nausea, vomiting and diarrhea. No data on weight loss was reported. The somewhat higher rates of certain adverse effects in the 10mg group were attributed to the forced titration from 5mg after week one of the study period.

 

An improvement in ADAS-cog scores of 4 points or greater was observed in 26.8% of placebo group, 37.8% of the 5mg group and 53.5% of the 10mg group. Intent to treat analysis showed the mean drug-placebo difference at endpoint in ADAS-Cog to be 2.49 points for the donepezil 5mg/day group and 2.88 points in the 10mg/day group. Mean decreases in the ADAS-Cog at endpoint were 0.67 for donepezil 5mg and 1.06 in the 10mg group. During the same period, the placebo group deteriorated (gained) 1.82 points on the ADAS-Cog. Scores on the CIBIC plus were significantly improved for both the 5 and 10mg groups relative to placebo at 26 weeks.

 

After the 6-week washout period, both treatment arms demonstrated no maintenance of treatment effect, with both groups deteriorating on all outcomes measures to levels near those of the placebo group.

 

The results of the Rogers et al trial previously described and other donepezil trials determined current clinical practice in several ways. First, donepezil therapy is best tolerated when initiated at 5mg/day for 4-6 weeks, then increased to 10mg/day. Secondly, the effects of treatment are not maintained after drug discontinuation - therapy once started, should be uninterrupted if the patient has demonstrated a positive response. Although not reported in the literature, it is the opinion of many clinicians that restarting therapy after the patient deteriorates when a CHE-I is discontinued often does not result in a return to the patient's previous CHE-I treated baseline.

 

Donepezil efficacy beyond 30 weeks:

Clinical efficacy data beyond 6 months is not required by the FDA for drug approval; however, knowledge about the long-term efficacy of the CHE-I is needed. The longest published trial of CHE-I treatment is that of Rogers and Friedhoff.¹²The authors reported on the 98 week interim analysis of an open label 192 week extension trial of donepezil. Patients who had completed a previous 14 week double-blind placebo-controlled trial of donepezil were eligible for participation.

Long-term Effects of Donepezil on Cognition: ADAS-Cog12

 

122 patients completed the first 12 weeks of the study; 29% were continuing in the study at the time of analysis and had been receiving donepezil for more than 3.5 years. The majority of patients received donepezil 10mg/day from week 60-120. As with any long-term study, attrition was associated with a variety of factors, but only 14 patients discontinued treatment because of adverse effects. Analysis of the ADAS-Cog data showed that patients maintained at or above their baseline for the first 38 weeks of open label experience. After week 50, subjects declined relative to baseline, indicating disease progression. At 98 weeks, the mean ADAS-Cog estimated increase of 6.6 points/year was postulated to represent a slower rate of disease progression than if the subjects had been untreated during the period - the ADAS-Cog increase was estimated to be 11.6 points/year for untreated patients.