Management of cognition and function: new results from the clinical trials programme of Aricept (donepezil HCl)

Transcription

1 International Journal of Neuropsychopharmacology (2000), 3 (Supplement 2), S13 S20. Copyright 2000 CINP Management of cognition and function: new results from the clinical trials programme of Aricept (donepezil HCl) SUPPLEMENT David S. Knopman Department of Neurology, University of Minnesota Medical School, Minneapolis, USA Abstract Ideally, treatment for Alzheimer s disease (AD) should prevent or cure the disease. Unfortunately, these goals appear unobtainable in the foreseeable future. Nevertheless, symptomatic relief is a feasible treatment option for AD patients and is available currently in the form of cholinesterase inhibitors such as tacrine, donepezil, metrifonate and rivastigmine. Donepezil is a second-generation, piperidine-class, selective and reversible acetylcholinesterase inhibitor. Four double-blind, placebo-controlled clinical trials of donepezil, involving over 1900 individuals with mild to moderate AD, have been published recently. In all trials, significant improvements in cognition were observed consistently for both therapeutic doses of donepezil (5 and 10 mg d), relative to placebo. Similar donepezil-associated benefits were reported for global functioning. In addition, in one 24-wk, multinational clinical trial, patients receiving donepezil (10 mg d) performed better than placebo-treated patients in their ability to perform complex daily functioning tasks. Donepezil was well tolerated in all trials, with approx. 79% of all donepezil-treated patients completing the studies compared with approx. 84% of placebo-treated patients. The most common adverse events associated with donepezil were generally cholinergic-induced and gastrointestinal in nature (e.g. nausea, diarrhoea, and vomiting) which were generally mild, transient and tended to occur after the dose was increased to 10 mg d from 5 mg d after 1 wk only. Sleep disturbances also occurred as the clinical trials utilized a bedtime dosing regimen. There was no evidence of organ toxicity or clinically significant treatment-emergent laboratory test abnormalities. Thus, donepezil appears to be a beneficial symptomatic treatment for patients with mild to moderate AD. Received 10 October 1999; Reviewed 26 December 1999; Revised 6 February 2000; Accepted 26 March 2000 Key words: Alzheimer s disease, clinical trials, donepezil, efficacy, safety. Introduction Alzheimer s disease (AD) is a degenerative neurological disorder characterized typically by an insidious onset and a progressive decline in patients cognition and function. Ideally, treatment of AD should aim to prevent or cure the disease. Those goals are not presently attainable but cholinesterase (ChE) inhibitor therapy represents a currently available palliative treatment. Therapy with ChE inhibitors produces benefits that are obvious to many caregivers and patients, even though the effects are often modest. An important advance in ChE inhibitor therapy has been the development of agents that are well Address for correspondence: Dr D. S. Knopman, Department of Neurology, 420 Delaware St SE, University of Minnesota Medical School, Minneapolis, MN 55455, USA. Tel.: Fax: knopman maroon.tc.umn.edu Note: Dr Knopman has served as a paid consultant to Parke Davis, Athena Neuroscience, Eisai Pfizer, Novartis, Janssen, Amgen, Searle and Wyeth Ayerst. He is currently participating or will participate in clinical trials with Pfizer Eisai. tolerated and easy to use. Donepezil HCl (Aricept ), a piperidine-based, selective and reversible acetylcholinesterase (AChE) inhibitor, is such an agent. Donepezil has been available in the USA by prescription since January 1997 and over patients in the USA have received donepezil as of August Evidence for the beneficial clinical effects of donepezil continues to grow. Three US double-blind, placebocontrolled clinical trials of donepezil have been reported in the past few years. One was a 12-wk trial in which the highest dose studied was 5 mg d (Rogers et al., 1996). The other two studies, one of 12 wk (Rogers et al., 1998a) and one of 24 wk (Rogers et al., 1998b) in duration, investigated both 5 and 10 mg d doses. In addition, another 24-wk, double-blind, placebo-controlled trial of donepezil conducted multinationally has also been completed recently (Burns et al., 1999). The results of all four trials are consistent in their demonstrations of the efficacy, safety and tolerability of donepezil. In addition, preliminary data on long-term use of donepezil have also appeared recently (Rogers and Friedhoff, 1998). This

2 S14 D. S. Knopman article reviews some of the highlights from these clinical trials with donepezil. Brief overview of the donepezil clinical trials All of the double-blind clinical trials of donepezil used standard inclusion exclusion criteria for the diagnosis of probable AD in the enrolment of subjects according to the National Institute of Neurological and Communicative Disorders and Stroke Alzheimer s Disease and Related Disorders Association (NINCDS-ADRDA) guidelines (McKhann et al., 1984) and Diagnostic and Statistical Manual of Mental Disorders (DSM-III-R; American Psychiatric Association, 1987). Participating patients had mild to moderate AD but were otherwise in good general health and were on no regular psychoactive medications. The mean Mini-Mental State Examination score at baseline for these patients was approx. 19 (with a range of 8 27), where scores of between and represent the ranges for mild and moderate dementia, respectively (Folstein et al., 1975). At study entry, 79% of patients had a Dementia Rating (Hughes et al., 1982) score of 1 (mild dementia) while 20% had a score of 2 (moderate dementia). The average age of patients entering these trials was 72 yr, with a slightly higher proportion of women than men. The first study to be published was a 12-wk, placebocontrolled, double-blind, dose-ranging study using a maximum dose of 5 mg d donepezil in the active treatment group. A 2-wk placebo washout followed the end of the 12-wk double-blind period (Rogers et al., 1996). Of the 161 patients entered into this trial, 141 completed the initial 12 wk of treatment. The second study was also a 12-wk randomized, placebo-controlled, double-blind study, but examining both 5 and 10 mg d doses of donepezil and the single-blind, placebo washout period was 3 wk in duration. Of the 468 patients who entered into this trial, 412 completed (Rogers et al., 1998a). The first of the longer studies of donepezil was a 24-wk, placebo-controlled, double-blind trial that used 10 mg d donepezil as a maximum dose. A 6-wk, singleblind, placebo washout period was conducted at the end of the trial. In this study, 473 patients were randomized to placebo, 5 mg d donepezil or 10 mg d donepezil groups (Rogers et al., 1998b). A second 24-wk study with a 6-wk, single-blind placebo washout period, enrolled 818 patients in 9 countries, the results of which have been reported recently (Burns et al., 1999). Although the 12-wk studies were useful in demonstrating the short-term efficacy of donepezil, it is the longer studies that provide more clinically meaningful data for practitioners. Most of the summary points to be made in this article are taken from observations from the two 24-wk trials. For both of these studies, assessments were carried out every 6 wk, but only the final ontreatment assessments at week 24 will be discussed in this article. Effects on primary outcome measures Cognitive assessment is one of the bases for the determination of efficacy for anti-ad agents (Leber, 1985; EAEMP, 1997). The other important and essential determinant of a drug s efficacy is the overall global impression of the patient, as judged by the treating clinician. The key findings from all of the clinical trials of donepezil were that both primary outcome measures the Alzheimer s Disease Assessment Scale cognitive subscale (ADAS-cog; Rosen et al., 1984) and the Clinician s Interview-Based Impression of Change scale (CIBIC-plus; Schneider et al., 1997) were both significantly improved in the group of patients receiving donepezil compared with patients administered placebo. Cognition The ADAS-cog is a 70-point mental status examination that is the principal cognitive assessment tool for AD clinical trials (Mohs and Cohen, 1988; Stern et al., 1994). Although simple change from baseline scores can give an indication of the effect of treatment, another representation of response is to consider the effect size of the treatment. Effect size refers to the difference in the change scores (i.e. the difference between baseline and the end of study assessments) between the drug-treated and placebotreated groups. It is difficult to translate a change in score on the ADAS-cog or any mental status examination into simple terms. In patients in placebo-treated groups over 6 months, a mild AD patient achieving a score of between 20 and 30 on the ADAS-cog declines about 2 points. Extrapolating that rough estimate to 1 yr, one could say that a 4-point decline on the ADAS-cog corresponds to 12 months decline. For the 10 mg d dose of donepezil, the effect size was approx. 2.9 points, as measured by ADAS-cog, at the end of the 24-wk treatment periods in both the US (Rogers et al., 1998b), and multinational studies (Burns et al., 1999) (Figure 1). Although the studies were not statistically powered to detect differences between the 5 and 10 mg d doses, a tendency for greater effect was demonstrated for the 10 mg d dose which reached statistical significance in the multinational study. In the US study, the effect size for the 5mg d donepezil group was 2.5 points, compared with 2.9 for the 10 mg d donepezil group. In the multinational study, the 10 mg d dose was also superior to the 5 mg d

3 trials of Aricept S Least squares mean change from baseline ADAS-cog score (±S.E.) Burns et al. (1999) Placebo Donepezil 5 mg/d 10 mg/d * p 0.05 ** p Rogers et al. (1998b) Placebo Donepezil 5 mg/d 10 mg/d Endpoint 30 Study week Placebo washout improvement decline Figure 1. Mean ( S.E.) change from baseline in ADAS-cog scores for the placebo and donepezil (5 or 10 mg d) groups during two, 24-wk controlled clinical trials. Data from the US (Rogers et al., 1998b) and the multinational study (Burns et al., 1999) are shown. Data-points for weeks 6, 12, 18 and 30 represent observed cases, while the data-points for endpoint at week 24 represent observed data from week 24 or last observations during double-blind treatment carried forward. dose over 24 wk, with effect sizes of 1.5 and 2.9 points for the 5 mg d and 10 mg d donepezil groups at study endpoint, respectively (Figure 1). The clinician s impression of change The CIBIC-plus is a clinician s global assessment scale that incorporates input from the primary caregiver (Schneider et al., 1997). It is performed by an experienced clinician independently of and blinded to other assessments. It measures patients global performance in terms of cognition, general function, behaviour and activities of daily living (ADLs) through the examination of 15 separate domains. Using the baseline interview as the sole source for comparison, patients are re-examined at each study visit to determine whether their condition has altered. This change from baseline is rated using a 7-point Likerttype rating scale, where 1 marked improvement, 4 no change and 7 marked worsening. The CIBIC-plus ratings are purposely insensitive estimators of drug effects, but the scale does convey an appreciation of how many individual patients can be expected to exhibit treatment benefits. In the multinational study, only 14% of placebo patients were rated as improved (scores of 3, 2 or 1) at study endpoint (which was usually the 24-wk point) compared with 21 and 25% of the 5 and 10 mg d donepezil-treated patients (Burns et al., 1999). The CIBIC-plus ratings can be compared to the ADAS-cog effects to gain an appreciation of how clinicians are able to detect changes seen on a cognitive test. These results are almost identical to the values observed in the US study (Rogers et al., 1998b). Effects of washout In both of these studies, examination of the outcome data from weeks 24 and 30 (the end of the placebo 6-wk washout period) showed that withdrawal of donepezil

4 S16 D. S. Knopman was associated with a decline in cognitive function with scores on both the ADAS-cog and CIBIC-plus declining to values that were not statistically different from placebo. Effects on function Subsequent to the initiation of the 24-wk US trial, it became clear that clinicians and healthcare policy-makers required further information on the effects of antidementia drugs, including donepezil, on patients daily functioning. Although experts in the field assumed that others could make the extrapolation from cognitive changes to effects on ADLs, the experience with both tacrine and then donepezil was otherwise and both practitioners and policy-makers requested further evidence from clinical trials. As a consequence, the effects of donepezil on daily function were assessed in the multinational study using a modified version of a scale known as the Interview for Deterioration in Daily living activities in Dementia (IDDD; Teunisse et al., 1991). There are 33 items in the IDDD assessment; 17 rate instrumental or complex tasks (IDDD-CT), e.g. shopping, initiating conversations, ability to find one s way in the neighbourhood or using a telephone. Six of these items specifically address the degree of initiative that the patient demonstrates while performing the activities. The other 16 items relate to self care (IDDD-SC), which include more basic ADLs such as bathing, dressing and toileting. LS mean change from baseline (±S.E.) p = p = Placebo Donepezil 5 mg/d 10 mg/d p = p = Both initiation of tasks and their performance by the patient are quantified during a structured interview of the caregiver. In this trial, severity of disability at baseline was scored for each item of the IDDD. As the IDDD was originally designed as a severity scale, it was modified at subsequent visits to assess change from baseline, where the evaluator rated improvement, no change or deterioration, in comparison to the baseline performance level. In assessment of change, a 7-point Likert-type scale was used, where 1 marked improvement, 4 no change from baseline and 7 marked deterioration from baseline. Hence, a patient exhibiting no change in complex task functioning, for example, would receive a total score of 68 (17 4) on the IDDD-CT. Thus, a score below 68 represents improvement and a score above 68 denotes deterioration. Patients who received the 10 mg d dose of donepezil performed better than placebo-treated patients on the IDDD-CT subscore at study endpoint (p 0.007; Figure 2). While the placebo group declined by slightly more than 3 rating points over 24 wk, the donepezil 10 mg d group declined by just over 1 point. Indeed, patients receiving the 10 mg d dose of donepezil performed better than placebo-treated patients on the IDDD-CT at all time points (p 0.01; Figure 2). There was no drug effect on the scale that assessed self-care (IDDD-SC). However, this was expected given that the patients enrolled into the study were mildly impaired at baseline p = Endpoint 30 Study week Placebo washout improvement decline Figure 2. Mean ( S.E.) change from baseline in IDDD (Complex Tasks), for 5 and 10 mg d donepezil and placebo patients with mild to moderately severe Alzheimer s disease. Data from the 24-wk multinational clinical trial are shown (Burns et al., 1999). Data-points for weeks 6, 12, 18 and 30 represent observed cases, while the data-points for endpoint at week 24 represent observed data from week 24 or last observations during double-blind treatment carried forward. (Reproduced with the kind permission of Karger, Switzerland.)

5 trials of Aricept S17 and therefore performing near the ceiling on self-care activities. Such patients would not be expected to deteriorate on such functions as bathing and toileting skills over a 24-wk study. In the 24-wk US study, a retrospective analysis was undertaken in which ratings from three subsections of the Dementia Rating Sum of the Boxes (CDR-SB; Berg et al., 1992; Hughes et al., 1982) that reflect ADLs, namely community affairs, home and hobbies, and personal self care, were used as a measure of daily function (Friedhoff and Rogers, 1997). Out of a possible total of 9 CDR-SB points, the mean combined rating of patients at baseline was Substantial deterioration from baseline was defined as a decline in 1 rating point. By week 24, the donepezil 10 mg d group had a significantly lower proportion of patients who met the criteria for deterioration than did the placebo group (p 0.05). Duration of the effect of donepezil An interim analysis on the open-label, long-term use of donepezil in subjects who had previously participated in the first 12-wk study (Rogers et al., 1996) has been reported recently (Rogers and Friedhoff, 1998). After successfully completing the initial 12-wk treatment and 2- wk washout portions of the study, patients were escalated to a maximum dose of 10 mg d donepezil (following a protocol amendment), or their highest tolerated dose, for up to a further 192 wk. In the final analysis of the study (Friedhoff et al., 1999), the majority ( 70%) of the 133 Mean change from the baseline (±S.E.) n = patients who began the open-label, long-term extension trial were taking 10 mg d donepezil by week 60. Although the number of patients from whom data were available after 1 yr was relatively small, attrition over the first 52 wk was low in this study (Figure 3). Of the 133 patients who began the open-label, long-term extension study, 100 were still receiving medication at the end of the first year. The mean ADAS-cog scores for the group did not decline to below baseline levels until approx. 38 wk into therapy. This duration of a positive effect of donepezil might have been predicted from studies that have followed the natural history of the symptoms of AD on the basis of the treatment effect size at the end of 24 wk (Kramer-Ginsberg et al., 1988; Stern et al., 1994). After that point, the rate of decline appeared similar to that according to the expected natural history rates of decline, but it should be noted that the magnitude of initial treatment effect appears to be maintained for up to 2 yr of treatment. However, because there was no placebo comparison, these data must be interpreted with caution, although it is obviously difficult, both ethically and logistically, to perform long-term, placebo-controlled studies. In the same extension trial, the mean score of the CDR- SB in patients treated with donepezil declined to below baseline at approx. week 12 but remained close to 0 (i.e. a negligible difference from baseline) until week 26, before declining slowly. By comparison, placebo-treated patients tended to decline immediately, or within 6 wk, after initiation in the double-blind studies. The group effect on improvement decline Cumulative weeks of donepezil treatment Figure 3. Maintenance of treatment effect over the long term. LS mean change ( S.E.) from baseline ADAS-cog score over the first 98 wk of a long-term, open-label, extension study. Data-points represent observed cases, with the number of subjects contributing to each measurement displayed.

6 S18 D. S. Knopman the CDR-SB tends to exhibit decline in such trials as the scale includes functional effectiveness among the rated domains. Thus, while the duration of the plateau on the CDR-SB for the donepezil-treated patients did not appear to be as long as with the ADAS-cog scale, the same pattern of delaying symptomatic deterioration was observed with this functionally sensitive outcome measure. Tolerability In studies, there was the low rate of subject attrition, even at the 10 mg d dose. In the US 24-wk study (Rogers et al., 1998b), 85 and 68% of the patients assigned to the 5 and 10 mg d donepezil groups, respectively, completed the trial, compared with 80% of patients assigned to placebo. In the multinational study, 78 and 74% of the 5 and 10 mg d donepezil-treated patients, respectively, completed 24 wk of therapy, compared with 80% of the placebo group (Burns et al., 1999). When titrated slowly over several weeks, the 10 mg d dose of donepezil has been well tolerated with an incidence of side-effects marginally higher than that of placebo and with no evidence of organ toxicity. The sideeffects that did occur were mainly cholinergic-induced and gastrointestinal in nature (e.g. nausea, diarrhoea, vomiting and loss of appetite). Sleep disturbances also occurred, probably because the clinical trials required dosing at bedtime. However, these were usually mild and transient, and generally resolved during continued donepezil administration. Indeed, as Figure 4 demonstrates, when 5 mg d donepezil is administered for 4 6 wk, rather than for just 1 wk, prior to the initiation of Percentage of patients the 10 mg d dose, the incidence of side-effects attributable to the 10 mg d dose decreased to that reported for 5mg d donepezil- and placebo-treated patients (Rogers and Friedhoff, 1997). Conclusions The results of the clinical trials of donepezil (5 and 10 mg d) completed to date are highly consistent across studies, regardless of the patients nationality, the investigators or local medical practice. All studies demonstrate that donepezil is an effective, well-tolerated and easily administered medication for the symptomatic treatment of mild to moderately severe AD patients. Most patients with mild to moderate AD should, at least, be offered a trial of donepezil or another cholinesterase inhibitor of comparable safety, tolerability and efficacy. Patients with serious, life-shortening medical illnesses and those who live alone who consequently are unable to comply with medication safely are not good candidates for ChE inhibitor therapy. There should also be a clear understanding among the caregivers and the physician that the therapy is intended to be used for several years. At the present time, use of donepezil in patients with chronic illnesses and a number of commonly used medications in the elderly appears to be safe (McRae et al., 1998). One of the unique contributions of the multinational study was to demonstrate the functional benefits of donepezil in AD patients compared with placebo on a measure of daily function. It had been assumed previously that functional improvement would be demonstrated in Donepezil 10 mg/d after 1 wk (n = 315) Donepezil 10 mg/d after 4 wk (n = 1034) Donepezil 10 mg/d after 6 wk (n = 269) Placebo (n = 315) 2 0 Nausea Diarrhoea Insomnia Fatigue Vomiting Muscle cramps Adverse event Figure 4. Incidence of adverse events with donepezil 10 mg d as a function of time to dose increase compared with placebo (results from controlled Phase II and III trials; Reichman et al., 1998).

7 trials of Aricept S19 patients treated with donepezil due to the relatively strong association between cognitive and functional performance and from the beneficial results of studies with some of the other ChE inhibitors. However, as no direct or clinical evidence for functional efficacy had been available for donepezil previously, the completion of the multinational study provided the required verification requested by healthcare policy-makers and practitioners (Burns et al., 1999). Unfortunately, the magnitude of the effect is difficult to convey since there is not much experience with the particular assessment instrument employed, the IDDD. The nature of the effectiveness of donepezil on functional compared with cognitive outcomes is of importance in understanding the drug s benefits. Averaged across all subjects receiving donepezil, some overall improvement is observed on cognitive measures such as the ADAS-cog. In fact, the average improvement is sustained up to approx. 38 wk while placebo-treated patients decline to below baseline after only 6 12 wk. After 38 wk, the rate of decline in donepezil-treated patients appeared similar to that which would be experienced by untreated patients. In contrast, with functional measures, some worsening in the mean change over time is observed. The key point from these observations is that, with donepezil, deterioration (where subjects became worse than placebo) was delayed compared with placebotreated patients by approx. 6 9 months. The relevance of these observations lie in the caregivers perceptions of the drug s effects. With treatments that delay symptomatic progression as opposed to producing dramatic improvements, there is very little opportunity for caregivers to notice the benefits of therapy. Indeed, in clinical practice, patients have requested discontinuation of medication after a few months simply because no improvement has been observed by the family. Such early discontinuation is unfortunate as the potential benefits of donepezil may only be noticed retrospectively from where the patient would have been without treatment. This concept of such an inferential comparison is not an easy one to illustrate to caregivers, but it is a fundamental feature of the relative effectiveness of donepezil. The duration of therapy is an issue that the clinical trials have not addressed adequately to date. Nevertheless, some insights can be provided. Both 24-wk studies of donepezil (Burns et al., 1999, Rogers et al., 1998b) suggest that the beneficial effects of donepezil are equivalent to a delay in the progression of symptoms for approx. 6 9 months. The single-blind, open-label extension study of the initial 12-wk study of donepezil also demonstrated a duration of symptom delay that was in that same range (Rogers and Friedhoff, 1998). Perhaps more importantly, the effects of the single-blind placebo, washout phases of the 24-wk studies demonstrate clearly the negative consequences of abrupt withdrawal of donepezil. The data show that patients after withdrawal of donepezil experience an abrupt decline in function and cognition. This rapid decline after sudden donepezil withdrawal is disconcerting to patients and families. It is imprudent, therefore, simply to implement donepezil therapy for 3 6 months and discontinue use. Rather, it is more reasonable to administer donepezil treatment until patients develop severe dementia or until a point is reached at which discontinuation of donepezil will produce no noticeable decline in symptoms, particularly function. As yet there are no observations on nursing-home placement with donepezil as there are with tacrine (Knopman et al., 1996), but all of the other similarities in efficacy suggest that donepezil will have a comparable effect on such an endpoint. In addition, one controlled study of donepezil s potential effects on behavioural symptoms experienced in AD has been reported (Cummings et al., 1999). References American Psychiatric Association (1987). Diagnostic and Statistical Manual of Mental Disorders (3rd edn, revised). Washington, DC: American Psychiatric Association. Berg L, Miller JP, Baty J, Rubin EH, Morris JC, Figiel G (1992). Mild senile dementia of the Alzheimer type: 4 evaluations of intervention. Annals of Neurology 31, Burns A, Rossor M, Hecker J, Gauthier S, Petit H, Mo ller H-J, Rogers SL, Friedhoff LT, and the International Donepezil Study Group (1999). The effects of donepezil in Alzheimer s disease results from a multinational trial. Dementia and Geriatric Cognitive Disorders 10 (3), Cummings JL, Katz IR, Tariot P, Perdomo CA, Whalen E, Schwamm EM and the Donepezil Nursing Home Study Group (1999). Donepezil in the treatment of Alzheimer s disease in a nursing home population. Neurology 52 (Suppl. 2), A481. European Agency for the Evaluation of Medicinal Products (1997). Human Medicines Evaluation Unit. Note for Guidance on Medicinal Products in the Treatment of Alzheimer s Disease (September). London: EAEMP. Folstein MF, Folstein SE, McHugh PR (1975). Mini-mental state. A practical method for grading the cognitive state of patients for the clinician. Journal Psychiatry Research 12 (3), Friedhoff LT, Ieni JR, Rogers SL, Pratt RD (1999). Donepezil provides long-term clinical benefits for patients with Alzheimer s disease. International Journal of Neuropsychopharmacology 2 (Suppl. 1), S175 (PW11017). Friedhoff LT, Rogers SL (1997). Donepezil lengthens time to loss of activities of daily living in patients with mild to

8 S20 D. S. Knopman moderate Alzheimer s disease results of a preliminary evaluation. Neurology 48 (3), A100 (P02.026). Hughes CP, Berg L, Danziger WL, Coben LA, Martin RL (1982). A new clinical scale for the staging of dementia. British Journal Psychiatry 140, Knopman D, Schneider L, Davis K, Talwalker S, Smith F, Hoover T, Gracon S (1996). Long-term tacrine (Cognex) treatment: effects on nursing home placement and mortality. Tacrine Study Group. Neurology 47, Kramer-Ginsberg E, Mohs RC, Aryan M, Lobel D, Silverman J, Davidson M, Davis KL (1988). predictors of course for Alzheimer patients in a longitudinal study: a preliminary report. Psychopharmocology Bulletin 24, Leber P (1985). US Food and Drug Administration, Center for Drug Evaluation and Research, Division of Neuropsychopharmacological Drug Products. Form and contents of NDA reviews: strategies for the efficacy analysis (5 November). McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM (1984). diagnosis of Alzheimer s disease. Report of the NINCDS-ADRDA work group under the auspices of the Department of Health and Human Services Task Force on Alzheimer s Disease. Neurology 34, McRae T, Relkin N, Knopman D. (1998). A large-scale, openlabel trial of donepezil in the treatme of Alzheimer s disease. Neurology 50 (Suppl. 4), A90. Mohs RC, Cohen L (1998). Alzheimer s Disease Assessment Scale (ADAS). Psychopharmacology Bulletin 24, Reichman WE, McRae T, and the Donepezil 313 Study Group (1998). Abstract presented at the 151st Annual Meeting of the American Psychiatric Association (APA), Toronto, Canada. Rogers SL, Doody RS, Mohs RC, Friedhoff LT (1998a). Donepezil improves cognition and global function in Alzheimer disease: a 15-week, double-blind, placebocontrolled study. Donepezil Study Group. Archives of Internal Medicine 158, Rogers SL, Farlow MR, Doody RS, Mohs R, Friedhoff LT (1998b). A 24-week, double-blind, placebo-controlled trial of donepezil in patients with Alzheimer s disease. Donepezil Study Group. Neurology 50, Rogers SL, Friedhoff LT (1997). Donepezil is well tolerated at clinically effective doses for the treatment of Alzheimer s disease. European Neuropsychopharmacology 7 (S2), S251. Rogers SL, Friedhoff LT (1998). Long-term efficacy and safety of donepezil in the treatment of Alzheimer s disease: an interim analysis of the results of a US multicentre open label extension study. European Neuropsychopharmacology 8, Rogers SL, Friedhoff LT, and the Donepezil Study Group (1996). The efficacy and safety of donepezil in patients with Alzheimer s disease: results of a US multicentre, randomized, double-blind, placebo-controlled trial. Dementia 7, Rosen WG, Mohs RC, Davis KL (1984). A new rating scale for Alzheimer s disease. American Journal of Psychiatry 141, Schneider LS, Olin JT, Doody RS, Clark CM, Morris JC, Reisberg B, Schmitt FA, Grundman M, Thomas RG, Ferris SH (1997). Validity and reliability of the Alzheimer s Disease Cooperative Study Global Impression of Change. The Alzheimer s Disease Cooperative Study. Alzheimer Disease and Associated Disorders 11 (Suppl. 2), S Stern RG, Mohs RC, Davidson M, Schmeidler J, Silverman J, Kramer-Ginsberg E, Searcey T, Bierer L, Davis KL (1994). A longitudinal study of Alzheimer s disease: measurement, rate, and predictors of cognitive deterioration. American Journal of Psychiatry 151, Teunisse S, Derix MM, Van Crevel H (1991). Assessing the severity of dementia. Patient and caregiver. Archives of Neurology 48,