Evaluation of Therapeutic Interchange From Donepezil to Galantamine Extended-Release

, , , , , , ,
The American Journal of Pharmacy Benefits, January/February 2011, Volume 3, Issue 1

Therapeutic interchange from donepezil to galantamine extended-release: a 3-month analysis.

Alzheimer’s disease (AD) is a progressive, neurodegenerative disorder of unknown etiology associated with a loss of acetylcholine. The loss of acetylcholine disrupts cortical functions such as memory, thinking, and orientation, resulting in a decline in cognition, behavior, and daily functioning.1,2 Treatment options include acetylcholinesterase inhibitors (AChEIs), which exhibit clinical benefits through the inhibition of the cholinesterase enzyme. Inhibition of the cholinesterase enzyme enhances cholinergic function, eventually improving, stabilizing, or slowing disease

progression.3-5 Alzheimer’s disease progresses slowly,

taking between 3 and 18 years to advance from the earliest

symptoms to death, with an average duration of 8 years.

AChEIs available in the United States include donepezil,

galantamine immediate-release (IR) and extended-release

(ER), rivastigmine (oral capsule and patch), and tacrine. In

2000, donepezil was the most commonly prescribed AChEI

in the United States and accounted for approximately 50%

of all prescriptions for the treatment of AD; rivastigmine was

the second most commonly prescribed medication, accounting

for about 6% of all prescriptions for AD.6 Tacrine was

rarely used in the United States because of the increased risk

of hepatotoxicity.6 The last new molecular AChEI approved

by the US Food and Drug Administration for AD was galantamine

in 2001, which is now available as an ER capsule for

once-daily dosing.7,8

Unlike other AChEIs that target only the acetylcholinesterase receptors, galantamine competitively and reversibly inhibits the acetylcholinesterase enzyme and modulates the nicotinic acetylcholine receptors. Modulation of nicotinic receptors allows for a stronger response at deteriorating cholinergic neurons. This unique mechanism of action may provide a potential advantage over other AChEIs in patients with AD.6,9,10 As a class, AChEIs have shown superior efficacy compared with placebo for the symptomatic treatment of mild to moderate AD and have been shown to delay the progression of AD by at least 6 months.10,11 Moreover, patients who do not respond to one AChEI may benefit with adifferent AChEI.11,12

There are 5 key reasons for switching therapies in patients with AD: (1) inefficacy or perceived inefficacy, (2) poor tolerability, (3) physician preference, (4) patient or caregiver request, and (5) economic considerations.6,13 Efficacy, safety, and tolerability must be considered when switching patients from one AChEI to another, as well as the optimal washout period for the discontinued medication and the optimal dose-escalation period for the replacement medication. Preliminary data from a trial published by Rasmusen et al showed that patients can tolerate a therapeutic interchange from one AChEI to another using a zero-day washout period.14 The most common side effects experienced were gastrointestinal (GI), including nausea (17.2%) and diarrhea (12.1%), both of which were tolerated and commonly seen with AChEIs.14 No other clinically significant side effects were observed.14

Using a zero-day washout period was necessary to maintain enough inhibition of the acetylcholinesterase enzyme to prevent destabilization. A study evaluating a switch from donepezil to galantamine IR using washout periods of 4 and 7 days reported an increase in adverse effects with increasing washout periods.15 Moreover, there were more discontinuations in the 7-day washout period compared with the 4-day washout period (10.6% and 4%, respectively).15

Ferris et al reported that prior exposure to an AChEI was not predictive of galantamine response, essentially concluding that patients with and without prior AChEI therapy had similar efficacy outcomes and tolerated galantamine well.6 To further strengthen the argument for galantamine’s use in AD, Wilcock et al provided evidence that patients on galantamine would experience improved clinical outcomes compared with patients on donepezil.16 Mini-Mental State Exam (MMSE) scores of patients receiving galantamine at 52 weeks did not differ significantly from baseline, whereas MMSE scores of patients on donepezil deteriorated.16 Furthermore, caregivers of patients receiving galantamine also reported decreased burden compared with donepezil, creating a case from the caregiver perspective.17

Therapeutic interchanges with other medications have proved to be efficacious, safe, and cost saving.18-20 Switching from a brand to generic product has demonstrated economic benefits while maintaining clinical efficacy and safety.21,22 However, for some medications with narrow therapeutic ranges, therapeutic interchange from brand to generic may lead to destabilization23 and reduction in patient satisfaction.24 Although there is a potential for cost savings with therapeutic interchanges, establishing clinical efficacy and safety should take priority over economic benefits.

Generic galantamine ER may potentially reduce some of the economic burden on the healthcare system. Therapeutic interchange from donepezil to galantamine ER may be cost saving while maintaining clinical stability. In addition, galantamine ER is dosed once a day, much like donepezil, which makes the therapeutic interchange convenient for patients and caregivers. This has the potential to reduce medication errors resulting from patients having to take galantamine IR, which is dosed twice a day. Moreover, in a disease state such as AD, it is important to reduce the frequency of doses in order to enhance patient and caregiver administration adherence.


In 2008, the Veterans Affairs San Diego Healthcare System (VASDHS) developed a strategy to perform a therapeutic interchange from donepezil to galantamine ER upon refill because of generic galantamine ER availability and potential for cost savings. The therapeutic interchange was approved by the pharmacy and therapeutics committee, the Veterans Integrated Service Network 22 Pharmacy Benefits Management group, and local geriatric-psychiatry and neurology experts. The purpose of this study was to evaluate the safety and tolerability of therapeutic interchange from donepezil to galantamine ER at our facility and to determine the number of patients who switched back to donepezil. To date, there is a gap in the literature evaluating the clinical efficacy and safety of switching AD patients from donepezil to galantamine ER in a Veterans Affairs setting.



This study was conducted at the VASDHS, a 238-bed medical center with specialized clinics servicing the veteran population in areas such as cardiology, endocrinology,gastroenterology, hemodialysis, infectious disease, neurology,ophthalmology, psychiatry, and rheumatology. TheVASDHS uses the Computerized Patient Records Systemto document notes, write prescriptions, order laboratorytests, view images, and perform medication reconciliation.

Therapeutic Interchange Process

Patients with active prescriptions for donepezil underwent a therapeutic interchange to galantamine ER between April 2008 and July 2008, after which patients were followed for an additional 3 months to evaluate tolerability and safety (

Figure 1

). When a patient initiated a prescription refill request for donepezil, the medication was switched to galantamine ER 8 mg and a 30-day supply was dispensed to begin the galantamine ER titration schedule. A separate 1-month supply of galantamine ER 16 mg capsules was automatically processed to arrive at the patient’s address 30 days later by mail from the centralized mail order pharmacy. Afterwards, patients would resume ordering refills via the refill phone line. The date that patients proactively requested their refill of galantamineER 16 mg was identified as the index date (Figure 1). Inclusion and exclusion criteria were applied at the index date to determine which patients to include into the study. All patients received a letter in advance that educated them about the upcoming change in their medication and were provided a phone number to speak with a clinical pharmacist should they have questions.

Posttherapeutic Interchange Analysis

Once patients initiated their galantamine ER 16 mg prescription, they were followed for 3 months after the index date. A successful therapeutic interchange was defined as occurring when a patient was converted from donepezil to galantamine ER and remained on galantamine ER at 3 months after the index date. This 3-month postindex period was selected to allow sufficient time for patients to exhaust any remaining supply of donepezil, complete the galantamine ER titration schedule, and begin the maintenance dose of 16 mg. At 3 months after the index date, patients were separated into 2 groups: those who remained on galantamine ER and those who did not. Primary end points were the number of patients who were successful with the therapeutic interchange and the number of patients who had switched back at 3 months after the index date. Secondary end points were tolerance and MMSE scores. Caregivers’ characteristics were measured for the different groups in order to determine their influence on therapeutic interchange success.


Patients were included into the study if they had an active prescription for donepezil with >1 refill available and if they had been dispensed 1-month supplies of galantamine ER 8 mg and galantamine ER 16 mg, and 1 refill of the galantamine ER 16-mg capsules. Patients were excluded from the study if they did not complete the full therapeutic interchange process or had a previous history of galantamine use. The study was approved by the local institutional review board and ethics committee at the VASDHS.

Statistical Analysis

Descriptive information was presented as mean, median, standard deviation, frequency, and percentage. The Shapiro-Wilks test was used to test for normality. Comparison was made between patients who successfully or unsuccessfully underwent therapeutic interchange from donepezil to galantamine ER at 3 months after the index date. Continuous data were compared using the Student t test or Mann-Whitney U test where appropriate (adjusted for unequal variances). Discrete data were compared using Pearson’s c2 test or Fisher’s exact test where appropriate. Pre-post analysis was performed on clinical outcome (ie, MMSE) using the paired t test or Wilcoxon signed rank test where appropriate. Statistical significance was defined a priori as P <.05 using a 2-tailed test. Statistical analysis was performed using SPSS version 15 (SPSS Inc, Chicago, IL).


Patient Demographics and Clinical Characteristics

A total of 212 patients who had active prescriptions for donepezil 10 mg with at least 1 refill of galantamine ER were identified. The final study sample that met inclusion and exclusion criteria was 109 patients. The median age of patients in the study was 81 years (mean 79.25 years; SD 9.23 years), with a median duration of disease of 2 years (mean 2.28 years; SD 2.05 years). The majority of the patients were male (98%, n = 107), and roughly half had a history of either substance abuse or smoking (59%, n = 64). Baseline median MMSE was 22.5 (mean 21.56; SD 5.63). Of the 109 patients included in this study, 100 remained on the galantamine ER maintenance dose at 3 months after the index date.

Patients in both groups did not differ significantly according to baseline demographic comparison or comorbidities (

Table 1

). Patients in the group that remained on galantamine ER were slightly older than those who did not remain on galantamine ER (median age of 81.5 years vs 79 years; P = .19). Patients who remained on galantamine ER at 3 months after the index date compared with those who discontinued treatment were mostly male (99% [n = 99] vs 89% [n = 8]; P = .159), had a shorter duration of disease (median 2 years vs 3 years; P = .214), and were less likely to have a history of alcohol use (29% [n = 29] vs 33% [n = 3]; P = .720) and smoking (28% [n = 28] vs 44% [n = 4]; P = .444). Patients who remained on the galantamine ER maintenance dose had similar baseline MMSE scores (median 23 vs 19.5; P = .579).

Patients who remained on the galantamine ER maintenance dose at 3 months after the index date had higher rates of baseline comorbid disease states such as endocrine (40% vs 33.3%; P >0.99), psychiatric (43% vs 33.3%; P = .731), respiratory (12% vs 0%; P = .593), hematologic (6% vs 0%; P >0.99), nephrologic (10% vs 0%; P >0.99), musculoskeletal (29% vs 0%; P = .109), and genitourinary (28% vs 22.2%; P >0.99) disorders compared with those who did not remain on galantamine ER 16 mg at 3 months after the index date, but these differences were not statistically significant (

Table 2

). Patients who did not remain on the galantamine ER maintenance dose had higher baseline rates of neurologic (44.4% vs 20%; P = .105) and ophthalmic (55.6% vs 26%; P = .115) disorders, as well as a slightly higher rate of cancer (22.2% vs 19%, P = .683); however, these differences were not statistically significant (Table 2).

Primary Analyses

Of 109 patients, 100 (91.7%) were successfully switched to galantamine ER. Patients who remained on the galantamine ER maintenance dose at 3 months after the index date reported adverse drug events of diarrhea, fall, weight loss, and difficulty swallowing, whereas patients who did not remain on galantamine ER reported experiencing agitation, excessive sleepiness, and dry mouth. No statistically significant difference was seen between the 2 groups with respect to adverse events. Among patients who underwent the therapeutic interchange, 9 stopped taking galantamine ER within the postindex period, 4 (3.7%) stopped taking galantamine ER and switched back to donepezil, 1 (0.9%) stopped taking galantamine ER and received donepezil from an outside provider, and 4 (3.7%) were defined as nonpersistent or nonadherent with their prescription of galantamine ER (

Table 3


Secondary Analyses

A larger number of patients who remained on galantamine ER maintenance dose at 3 months after the index date had a caregiver available to assist them with activities of daily living compared with those who did not remain on the galantamine ER maintenance dose (86% [n = 81] vs 75% [n = 6]; P = .437). Of the patients who remained on galantamine ER and had a caregiver available, 67 (79.8%) had family member assistance, 15 (17.9%) received assistance from a facility, and 2 (2.4%) received assistance from a friend. A total of 12 caregivers (14.6%) lived outside the patient’s home, 68 caregivers (82.9%) lived at the patient’s home, and the residence of 2 caregivers (2.4%) was unknown; 5 patients (6.3%) lived alone. Of the patients who did not remain on the galantamine ER maintenance dose and had a caregiver available, 5 (83.33%) had assistance from family members, and 1 (16.7%) received assistance from an outside facility. A total of 2 caregivers (33.3%) lived outside the patient’s home, 3 caregivers (50%) lived at the patient’s home, and the residence of 1 caregiver (16.7%) was unknown; 2 patients (33.3%) lived alone. No statistically significant differences were seen between the 2 groups.

Available mean MMSE scores for 3 months after the index date were also compared between the groups (

Figure 2

). The MMSE scores of patients who remained on the galantamine ER maintenance dose versus those of patients who discontinued the treatment were not significantly different. The average MMSE score for 19 patients on the galantamine ER maintenance dose was 20.64 (SD 4.93) compared with a mean MMSE of 20.33 (SD 8.51) for the 3 patients that did not remain on the galantamine ER maintenance dose at 3 months after the index date, with P = .644. In the pre-post analysis, there were no statistically significant differences between baseline MMSE scores and scores 3 months after the index date for patients who remained on the galantamine ER maintenance dose and those who did not.


The results from our retrospective review suggest that a majority of patients who underwent zero-day washout therapeutic interchange from donepezil to galantamine ER were successfully switched at 3 months. There was 1 report of discontinuation due to GI events, which was expected with AChEIs. Previous studies have demonstrated clinical efficacy with therapeutic interchanges between donepezil and rivastigmine/galantamine IR,15,25 but no published evidence was available for a switch from donepezil to galantamine ER. The potential for overinhibition of the AChEI with a zero-day washout period may lead to increased adverse reactions; however, this effect was not observed in our study.

Mean and median MMSE scores did not significantly differ between the 2 groups at baseline and at 3 months after the index date. However, these scores were not alwaysreported in the chart and may not be a true reflection ofthe population at baseline or at 3 months after the indexdate. In addition, we evaluated the pre-post differencesfor patients who successfully or unsuccessfully underwenttherapeutic interchange to the target medication and foundno statistically significant differences. Adverse drug reactionrates of patients switched from donepezil to galantamineER also were not significantly different. Althoughthe group that remained on galantamine ER at 3 monthsafter the index date reported side effects of diarrhea, fall,weight loss, and difficulty swallowing, the group that didnot persist on galantamine ER at 3 months after the indexdate reported a different constellation of side effects(agitation, excessive sleepiness, and dry mouth). Despite these differences in reported side effects, there was insufficientevidence to support an association between adversereaction and discontinuation of galantamine ER.

Caregivers’ relationship to their patients has been shown to decrease the progression of AD; patients who had close relationships with their caregivers had a slower decline in cognitive and functional capacity.26 The availability of a caregiver could potentially influence the outcome of the therapeutic interchange. In order to investigate this confounding variable, we evaluated success of therapeutic conversion between patients who had (or did not have) a caregiver. In our study, we did not find any statistically significant differences between patients who had a caregiver and those that did not; however, low sample size and missing data limit the generalizability of our results, which should be interpreted with caution.

When considering the results of this study it is also important to address issues such as missing data. Baseline information such as caregiver availability, social history, date of AD diagnosis, and duration of disease were inconsistently available, causing baseline analysis of patient demographics to reflect only those with available data. Chart notes were also inconsistently documented, with missing MMSE scores. The MMSE scores were reported only if they were documented in the Computerized Patient Records System within 3 months of baseline or at 3 months after the index date. The availability of data was a limitation in our study. To precisely capture these data, a formal policy measure such as mandatory MMSE reporting should be applied.

To be considered as successfully converted, patients had to be on a stable dose of donepezil (eg, 10 mg daily) and had to have gone through the titration (galantamine ER 8 mg daily for 30 days, then galantamine ER 16 mg daily for maintenance). Moreover, patients had to proactively refill the first maintenance dose after the titration to demonstrate that they were taking galantamine ER. It was theorized that patients who actively refilled a galantamine ER maintenance dose were responding to and tolerating their new medication. Patients who did not go through the titration were excluded from analysis because it was assumed that these patients were not able to experience the benefits of the medication. Early discontinuations or switches back to donepezil before the index date were not included in this analysis. These patients were not reflective of the actual sample who underwent the full therapeutic interchange, and including them in the analysis would have biased the results toward a negative outcome. We sought to reduce this bias by developing clear requirements for the titration schedule and ensuring that patients had an opportunity to have an adequate trial of galantamine. It is possible that patients did not complete the titration schedule due to adverse drug reactions or lack of efficacy, but those data were not retrievable in our study.

This study has several limitations that we were unable to control for. First, this was a retrospective study with missing data for the clinical outcomes (ie, MMSE scores). The small sample that was used in the pairwise analysis was not sufficiently powered; hence, there was a potential for type II error. The study used evidence from another trial that used a zero-day crossover and assumed that all patients would tolerate this strategy. However, this strategy may have contributed to patients discontinuing galantamine early before completing their titration period. Finally, this study was limited to a small sample within the VASDHS and may not reflect the true population. Larger prospective studies evaluating clinical outcomes with MMSE scores and caregiver availability will need to be initiated to reduce the amount of confounding.


A majority of patients (92%) included in this study remained on galantamine ER at 3 months after the index date; however, there was a 3.7% switchback rate to donepezil. Small sample size and lack of available data (eg, MMSE scores) limit the generalizability of our results. Future prospective trials are needed in order to validate our findings.