Cholinesterase Inhibitors for the Treatment of Alzheimer's Disease:

Abstract

Background: Cholinesterase (ChE) inhibitors currently used in the treatment of Alzheimer's disease (AD) are the acetylcholinesterase (AChE)-selective inhibitors, donepezil and galantamine, and the dual AChE and butyrylcholinesterase (BuChE) inhibitor, rivastigmine. In addition to differences in selectivity for AChE and BuChE, ChE inhibitors also differ in pharmacokinetic and pharmacodynamic properties, and these differences could significantly impact on safety, tolerability, and efficacy.

Objective: The aim of this article was to provide an overview of the ChE inhibitors widely used in AD, focusing on key pharmacologic differences among agents and how these may translate into important differences in safety, tolerability, and efficacy in clinical practice.

Methods: Using published literature collected over time by the author, a review was conducted, focusing on the pharmacology and clinical data of donepezil, galantamine, and rivastigmine.

Results: All ChE inhibitors have the potential to induce centrally mediated cholinergic adverse events (AEs), such as nausea and vomiting, if the dose is increased too rapidly or in increments that are too large. These AEs, which are most likely to occur during the “getting on,” or dose-escalation, phase of treatment, may result in patients discontinuing treatment early without achieving optimum therapeutic benefit. To reduce the incidence of these AEs, a slow dose-escalation schedule has been established in clinical practice, consisting of a “start low, go slow” procedure with a minimum of 4 weeks between dose increases. After “getting on” treatment, maintaining treatment in the long term, or “staying on,” may be achieved with good safety, tolerability, and sustained symptomatic efficacy across the key symptom domains (activities of daily living, behavior, and cognition).

Conclusions: ChE inhibitors provide symptomatic benefit in AD across key symptom domains. Factors influencing the safety, tolerability, and efficacy of these agents in clinical practice include ChE enzymes inhibited, brain and brain-region ChE selectivity, and metabolism route. Class-specific cholinergic AEs can be minimized using slow, flexible dose escalation.

INTRODUCTION

Alzheimer's disease (AD) is a chronic, progressive, neurodegenerative disorder of the brain characterized clinically by deterioration in the key symptoms of activities of daily living (ADLs), behavior, and cognition. The cholinergic hypothesis states that the cognitive decline in AD is secondary to deficits in central cholinergic neurotransmission¹ resulting from a loss of acetylcholine (ACh). Cholinesterase (ChE) inhibitors enhance central cholinergic function by inhibiting the enzymes that degrade ACh, thereby increasing the availability of ACh to stimulate nicotinic and muscarinic receptors within the brain. Since their introduction into clinical practice, ChE inhibitors have been, and remain, the standard approach to the symptomatic treatment of AD. In the United States, these agents are the only approved pharmacologic approach shown to be effective in this disease. Although the clinical benefits of these agents may be relatively small, studies have demonstrated substantial effects in terms of reduced caregiver burden and delayed move to institutionalized care. In addition, economic analyses using models from the United States, the United Kingdom, and Canada suggest that ChE inhibitors, when initiated in the early stages of disease, may be effectively cost-neutral as a result of patients remaining in a nonsevere state of disease for a longer time.^2,3

Data from numerous clinical trials of these agents demonstrate improvements across the 3 key symptom domains,^4–8 and results suggest that, in addition to symptomatic benefits, ChE inhibitors also may have disease-modifying properties.⁹ Despite this evidence, however, significant numbers of AD patients are not prescribed ChE inhibitors or terminate treatment prematurely without attaining maximum clinical benefit.

ChE inhibitor treatment comprises 2 key stages: a dose-escalation phase to achieve a clinically effective dose (ie, “getting on” therapy) and a maintenance phase, during which the patient is sustained at an optimal therapeutic dose (ie, “staying on” therapy). Both of these treatment phases may be associated with adverse events (AEs) that result from the pharmacologic profile of the drug.

This article aimed to provide an overview of the ChE inhibitors, focusing on key pharmacologic differences among agents and how these may translate into important differences in safety, tolerability, and efficacy in clinical practice. Using published literature collected over time by the author, a review was conducted, focusing on the pharmacology and clinical data of donepezil, galantamine, and rivastigmine.

CHOLINESTERASE INHIBITORS: AN OVERVIEW

Three ChE inhibitors are commonly prescribed for the symptomatic treatment of AD—the acetylcholinesterase (AChE)-selective inhibitors, donepezil and galantamine, and the dual AChE and butyrylcholinesterase (BuChE)inhibitor, rivastigmine. A fourth agent, tacrine, is no longer routinely prescribed due to a high incidence of hepatotoxicity at therapeutic doses.^10,11 Although all ChE inhibitors share the same basic mode of action (inhibition of AChE), their pharmacokinetic characteristics differ substantially(Table I).^12–24

Table I.

Overview and pharmacokinetic characteristics of cholinesterase (ChE) inhibitors.

	Donepezil∗	Rivastigmine†	Galantamine‡
Chemical class	Piperidine	Phenylcarbamate	Phenanthrene alkaloid
Enzymes inhibited
AChE	Yes	Yes	Yes
BuChE	Negligible	Yes	Negligible
nAChR modulation	Uncertain	Not studied	Yes
Brain vs peripheral selectivity	Uncertain	Yes	No
Selectivity for AChE isoforms	None	G1	None
Sustained inhibition of ChEs during long-term treatment	No	Yes	No
Metabolism by CYP450 isozymes	Yes	Minimal	Yes
Plasma protein binding, %	∼96	∼40	∼18
Recommended dose, mg/d	5 to 10	6–12	16 to 24

AChE = acetylcholinesterase; BuChE = butyrylcholinesterase; nAChR = nicotinic ACh receptor; CYP450 = cytochrome P450.

^∗References 12–20.

^†References 12, 18, 20–22.

^‡References 12, 14, 15, 19, 23, 24.

KEY PHARMACOLOGIC CHARACTERISTICS

Inhibition of Cholinesterases

Cholinergic neurotransmission occurs when ACh released from the presynaptic neuron binds to nicotinic or muscarinic postsynaptic ACh receptors. Mammalian brains contain 2 major forms of ChEs—AChE and BuChE—both of which possess the capacity to hydrolyze ACh. These 2 enzymes differ genetically, structurally, and kinetically.²⁵ AChE is located in the synaptic cleft (soluble form) and synaptic membranes (membrane-bound form),²⁶ and BuChE is mainly associated with glial cells.²⁵

An important feature differentiating ChE inhibitors is specificity for the ChE enzymes inhibited. All 3 agents inhibit AChE, which is the main enzyme responsible for the breakdown of ACh in the normal brain. In contrast, rivastigmine acts differently from donepezil and galantamine in targeting both AChE and BuChE.²² Although BuChE represents only 10% of total ChE activity in the temporal cortex of the healthy human brain,²⁷ recent results using sensitive histochemical techniques indicate that the enzyme is capable of hydrolyzing ACh and plays a greater role in normal cholinergic transmission than previously thought.²⁸ The importance of BuChE in cholinergic neurotransmission is likely to increase in AD because as the disease progresses, AChE activity decreases by up to 45%, and BuChE activity increases by 40% to 90%.^27,29 Results with rivastigmine show that cognitive improvements (measured using the computerized neuropsychologic test battery) correlate independently with the inhibition of AChE and BuChE in the cerebrospinal fluid of AD patients,³⁰ suggesting that inhibition of both enzymes is a highly desirable feature of AD therapy.

Selectivity for Different Molecular Forms of Cholinesterase

The ChE inhibitors also differ in selectivity for different forms of ChE. Both AChE and BuChE exist in a series of globular forms, including the G1 and G4 forms.³¹ In the AD brain, a selective reduction in the level of the G4 form occurs, whereas the level of the G1 form remains unchanged or increases.^29,32 This effect is most pronounced in the cortex and hippocampus, which are severely affected in AD. Therefore, a preferential affinity for G1 AChE, as seen with rivastigmine, may contribute to greater activity in the brain areas most affected in AD.³³ A lower affinity for G4 AChE may lead to improved safety over the course of AD, as the G4 form of AChE predominates in the peripheral nervous system.^34,35 In addition, levels of the G1 form of AChE and BuChE are particularly high in neuritic plaques, a characteristic neuropathologic feature of AD.²⁹

Allosteric Nicotinic Acetylcholinesterase Receptor Modulation

Studies have shown that galantamine binds to the nicotinic ACh receptor (nAChR) at an additional site to that associated with binding of the natural agonist, ACh.^{13–15,23,24} Therefore, it has been proposed that cobinding of galantamine and ACh leads to allosteric modulation of nicotinic receptor function.²³ In vitro studies have shown that physostigmine, donepezil, and tacrine may also allosterically modulate the nAChR.^{13–15,24,36} It is thought that donepezil interacts solely with an allosteric activator site on the nAChR and that tacrine interacts with an allosteric activator site and the ACh binding site.¹³ To date, it is not known whether rivastigmine also possesses allosteric modulatory activity at the nAChR. Although there is much speculation as to the clinical relevance of such an effect, any additional clinical benefit of this pharmacologic property has yet to be elucidated.

GETTING ON TREATMENT

ChE inhibitors have the potential to induce centrally mediated cholinergic AEs (eg, nausea and vomiting) if the dose is increased too rapidly or in increments that are too large. In these circumstances, the initial phase of treatment (ie, “getting on” treatment) may be associated with acute cholinergic AEs,¹⁰ the development of which often can be a decisive factor in whether the patient and/or physician chooses to continue treatment. Clinical trials with the different ChE inhibitors have shown varying incidences of acute cholinergic AEs. For example, in the rivastigmine pivotal clinical trials,^4,5 the frequency of nausea and vomiting was relatively high, largely due to the rapid, forced titration to maximum tolerated dose employed under the clinical trial protocols.

It is thought that the occurrence of these AEs reflects the rapid increase in central ACh levels after oral intake.³⁷ An open-label pilot study³⁸ with 4 pharmacologically distinct antiemetic drugs showed that only centrally acting agents were efficacious in preventing the nausea and vomiting associated with rivastigmine treatment. This information indicates that these AEs are centrally mediated due to elevation of brain ACh levels. The ability of rivastigmine to inhibit both AChE and BuChE might make it more effective in increasing central ACh levels,³⁹ thereby having greater potential to induce cholinergic AEs if the dose is increased too rapidly.

The potential of ChE inhibitors to induce gastrointestinal AEs is a class effect. However, certain patient characteristics, and the way the drug is administered, can affect the acute tolerability of ChE inhibitors (Table II).

Table II.

Factors that may influence the acute tolerability of cholinesterase (ChE) inhibitors. (Adapted with permission.³⁵)

Improve Tolerability	Worsen Tolerability
Slow dose escalation	Rapid dose escalation
Small dose increments	Large dose increments
Administration with a full meal (morning and evening meals)	Administration on an empty stomach
Male sex	Female sex
Previous ChE inhibitor therapy	No previous ChE inhibitor therapy
Concomitant use of centrally acting antiemetics	Concomitant use of inhibitors of hepatic drug metabolism (eg, paroxetine)

Importance of Slow Dose Escalation

It is thought that slow dose escalation minimizes the incidence of centrally mediated cholinergic AEs via desensitization of dopamine receptors in the hypothalamic area postrema, the putative “vomiting center” of the brain. Clinical evidence^7,40,41 suggests that allowing a minimum interval of 4 weeks between dose increases significantly reduces the occurrence of acute gastrointestinal AEs with all ChE inhibitors. For example, slow dose escalation of donepezil resulted in a reduction in reported acute cholinergic AEs to close to placebo levels.⁴⁰ Similar findings have been noted for galantamine: the dropout rate due to AEs fell from 23% with 1-week titration steps⁴² to 10% with 4-week steps.⁷ With rivastigmine treatment it is particularly important to ensure a 4-week interval between dose increases due to the ability of this agent to potentially increase central ACh levels.³⁹ This dose-escalation schedule led to an AE dropout rate of just 2% with rivastigmine treatment⁴¹ compared with 14% and 9% using 1- and 2-week gaps, respectively.⁴³ With rivastigmine and galantamine treatment, it is also possible to improve acute tolerability and dosing flexibility by using the oral solution to increase the dose in smaller increments.

Administration with Food

In addition to slow dose escalation, administration of some ChE inhibitors with a full meal also may reduce the incidence of centrally mediated cholinergic AEs.^12,41 Administration of these agents with a meal delays drug absorption and lowers the peak plasma and brain concentrations, which reduces the likelihood that the patient will experience acute AEs. For rivastigmine and galantamine, it is recommended that the drug be taken with morning and evening meals.¹²

Achieving a Usual Therapeutic Dose

In clinical practice, a 1-step dose-escalation schedule to a usual therapeutic dose is used for donepezil, galantamine, and rivastigmine. For donepezil, an initial dose of 5 mg once daily is escalated to 10 mg once daily, the maximum recommended dose, after 4 to 6 weeks.¹² For galantamine, a starting dose of 4 mg BID is escalated after 4 weeks to 8 mg BID, the usual therapeutic dose.¹² This agent also provides an additional dose of 12 mg BID, although some clinical studies have failed to demonstrate a significantly greater benefit for this dose over that of 8 mg BID.^7,12 For rivastigmine, the initial dose is 1.5 mg BID, which is escalated to 3.0 mg BID (a usual therapeutic dose) after a minimum of 4 weeks. This dosing level offers a clinically significant benefit, with many patients requiring no further increase.^44–46 A broad therapeutic dosing range and consistent dose-efficacy relationship^5,44,47,48 provide the opportunity for flexible dose increases at 4-week intervals up to a maximum of 6 mg BID if symptomatic deterioration occurs.

STAYING ON TREATMENT

For AD patients treated with ChE inhibitors, it is important that treatment not only starts early but also continues long term. The likelihood that patients will remain on treatment long term (ie, “staying on”) depends on a number of factors, including the occurrence of AEs during maintenance treatment and the clinical efficacy profile across the key symptom domains. Unlike the acute, cholinergically mediated AEs, which occur as a class effect during the dose-escalation phase of treatment, the safety, tolerability, and symptomatic efficacy observed during maintenance treatment may differ significantly among the available ChE inhibitors. These differences occur as a direct result of the underlying pharmacodynamic and pharmacokinetic properties of ChE inhibitors, as outlined in Table I.^12–24

Safety

Drug-Drug Interactions

Donepezil and galantamine are both metabolized via the hepatic cytochrome P450 (CYP450) enzyme system.¹² Therefore, caution should be exercised when coprescribing other drugs metabolized via these pathways.¹² Rivastigmine is metabolized primarily by its target enzymes, AChE and BuChE, with minimal involvement of the hepatic CYP450 system and, therefore, is unlikely to interact with drugs that are metabolized by CYP450 isozymes.^12,47 Unlike donepezil, which has a plasma half-life of ∼70 hours, both rivastigmine and galantamine have short plasma half-lives (∼1–2 and 6 hours, respectively) with a low potential to accumulate, even with long-term use.^12,47,49 In addition, low plasma protein binding (∼18% for galantamine¹² and 40% for rivastigmine¹²) means that these agents do not displace other highly protein-bound drugs during coadministration. In contrast, donepezil shows high plasma protein binding (96%).¹²

Pooled data from placebo-controlled clinical trials⁵⁰ in 2459 individuals (1696 rivastigmine, 763 placebo) demonstrated no clinically significant drug interactions between rivastigmine and 22 classes of concomitant medications (eg, anti-inflammatory, cardiovascular, and antidiabetic drugs) commonly prescribed to the elderly.

Central and Peripheral Nervous System Adverse Events

Agents that lack central selectivity and brain region selectivity (Table I) may be associated with certain AEs during maintenance treatment. These AEs may include sleep disturbances, muscle cramps, generalized weakness, and increased extrapyramidal symptoms (EPSs). Rivastigmine does not appear to be significantly associated with any of these AEs, and this may be due to its preferential affinity for the G1 rather than the G4 molecular form of AChE.^22,51 As previously discussed, G4 AChE is the predominant form in the periphery and particularly at the presynaptic membrane of the skeletal neuromuscular junction, which may explain why rivastigmine appears to be associated with a lower incidence of muscle cramps or weakness than, for example, donepezil.^6,22,52,53

Sleep disturbances and changes in sleep-wake cycles are common clinical features of dementia.⁵⁴ These disturbances can be a significant source of caregiver physical and psychologic burden and can contribute to the decision to institutionalize a patient.⁵⁵ Therefore, it is important that these symptoms not be further exacerbated by the treatment the patient is receiving. Neurophysiologic studies suggest that both donepezil and galantamine treatment may reduce rapid eye movement (REM) latency,^56,57 leading to decreased slow-wave sleep.⁵⁸ In clinical trials^53,59 with donepezil, insomnia was observed 2- to 3-fold more frequently in patients treated with donepezil than placebo. Direct studies of REM latency and density have provided evidence supporting the targeted activity of rivastigmine in the brain. Rivastigmine increases REM density, indicating activity in the hippocampus without any corresponding change in REM latency, suggesting minimal disturbance of brain stem function.^51,60 Further evidence of the brain region selectivity of rivastigmine comes from the improvement in or lack of sleep AEs in patients with AD, Lewy body dementia (LBD), and Parkinson's disease dementia (PDD) receiving this agent.^61–65

EPSs include muscle stiffness, slowness of movement, postural instability, body restlessness, gait imbalance, and tremor. These symptoms can occur in patients with AD, LBD, and PDD, and may be exacerbated by neuroleptic drug use.^66–68 Rivastigmine has a low potential for causing EPSs due to a relative lack of G4 AChE inhibition in the striatum, which has been supported by various studies^65,69–71 in patients with LBD and PDD exhibiting a high level of parkinsonian symptoms at baseline. In contrast, EPSs have been reported more commonly in LBD patients treated with donepezil.^72,73 Severe EPSs also have been observed when donepezil was administered in combination with risperidone or tiapride.^74,75

Safety and Tolerability of Cholinesterase Inhibitors in Clinical Practice

A direct comparison of reported AEs occurring in clinical practice in the United States during the first 6 months after the introduction of donepezil and rivastigmine revealed that the total number of adverse drug reactions and serious AEs per 1000 prescriptions was significantly higher with donepezil treatment than with rivastigmine (adverse drug reactions, 0.64 vs 0.55, respectively [P<0.05]; serious AEs, 0.22 vs 0.14, respectively [P<0.01]).⁷⁶ No published data are yet available for galantamine.

Efficacy

Sustained Cholinesterase Inhibition During Long-Term Treatment

As previously discussed, during AD progression, AChE activity decreases and BuChE activity increases.^27,29 The ratio of BuChE:AChE may change from 0.6 to as high as 11.0,⁷⁷ which may alter the normally supportive role of BuChE in the hydrolysis of excess ACh. As a dual inhibitor of both AChE and BuChE, rivastigmine may provide maximum preservation of ACh levels independent of the changing contributions of AChE and BuChE to ACh hydrolysis as the disease progresses. This occurrence is in contrast to the AChE-selective inhibitors, galantamine and donepezil, the inhibitory capacities of which are likely to decline as BuChE becomes more important for ACh hydrolysis.

Inhibition of both ChE enzymes by rivastigmine is sustained during long-term treatment,²¹ which is important for the maintenance of ACh levels and thus for clinical efficacy. Although the mechanism behind this long-term effect is as yet unknown, it may be a result of slow dissociation from target enzymes. After 12 months of treatment with rivastigmine (mean dose at 12 months, 11.8 mg/d), AChE and BuChE activities decreased from baseline by 46% and 65%, respectively.²¹ In comparison, 6 to 12 months of treatment with donepezil or galanta-mine resulted in increased AChE activity.^18–20 The effects of long-term donepeziland rivastigmine administration on ChE activities are depicted in Figure 1.^18,21 Although these findings clearly warrant further investigation, the increase in ACh activity seen with donepezil and galantamine may have negative implications in terms of disease progression and may explain why a sudden worsening of symptoms, or “crash effect,” has been observed in some patients following discontinuation of donepezil.⁷⁸

Figure 1.

(A) Long-term effects on cerebrospinal fluid (CSF) cholinesterase inhibition by donepezil and rivastigmine. ^∗P<0.02; ^†P<0.03. AChE = acetylcholinesterase. (Adapted with permission.¹⁸) (B) Long-term effects of high-dose rivastigmine on CSF AChE and butyrylcholinesterase (BuChE) activity. (Adapted with permission.²¹)

Long-Term Symptomatic Efficacy

Data on the long-term use of donepezil comprise open-label extension studies of randomized, controlled trials (RCTs)^79–81 and placebo-controlled data for use up to 1 year.^82,83 Doody et al⁷⁹ presented results from 763 patients enrolled in a 144-week, open-label extension of 2 preceding RCTs. The deterioration in the mean change from baseline of the Alzheimer's Disease Assessment Scale⁸⁴ cognitive subscale (ADAS-Cog: 0–70 points; higher scores = impairment) was 7.4 to 9.1 points for the group that received 5 mg/d and 7.9 to 10.0 points for the group that received 10 mg/d, depending on the study. For patients who had received placebo during the RCTs, the decline was between 11.3 to 12.9 points, depending on the study.⁷⁹ In 2 articles^80,81 describing the same sample of 133 patients receiving donepezil in an open-label extension of a preceding RCT, 46 patients (34.6%) were still receiving the drug after 98 weeks. The deterioration in ADAS-Cog mean change from baseline was 10.2 points compared with a predicted decline of ∼18.0 points in untreated patients. These studies suggest that the beneficial effects of donepezil treatment are maintained during long-term treatment, although the patient numbers involved are relatively small. Two placebo-controlled trials^82,83 assessed the efficacy of donepezil over a period of 1 year. In the study by Winblad and colleagues⁸² (N = 286), donepezil failed to show significant cognitive benefits over placebo at study end point in the primary analysis (intent-to-treat, last observation carried forward) on the primary efficacy measure, the Gottfries-Bråne-Steen Scale,⁸⁵ or any significant improvement in behavior (Neuropsychiatric Inventory⁸⁶). Improvement in ADLs (as measured by the Progressive Deterioration Scale⁸⁷) was seen at 52 weeks but not at earlier time points, and cognitive improvement was observed in a secondary measure (Mini-Mental State Examination^®88 [MMSE]: scale: 0–30 points; lower scores = impairment). In the study by Mohs et al,⁸³ which enrolled 431 patients with moderate AD, a higher proportion of patients receiving placebo (56%) compared with those receiving donepezil (41%) met criteria for clinically evident functional decline, although this difference was not significant. In addition, no difference in the mean functional status in patients completing the study was found between the donepezil and placebo groups. However, the time to evident functional decline significantly favored the donepezil group compared with placebo (P = 0.02).

Data on the long-term use of galantamine (24 or 32 mg/d) comprise a 26-week, placebo-controlled study followed by an additional 26 weeks of open-label treatment (24 mg/d), which yielded inconsistent results.⁴² Patients who received galantamine 24 mg/d for the full 52 weeks maintained cognitive function and ADL performance compared with patients who switched from placebo to galantamine at 26 weeks. However, by the end of the study, the results from patients receiving 32 mg/d during the placebo-controlled arm followed by24 mg/d during open-label treatment were not significantly different from those who were switched from placebo to active treatment after 26 weeks. More recently, data have emerged suggesting that the efficacy of galantamine may be sustained for up to 3 years.⁸⁹ Of 327 patients entering the extension period, 55.7% completed 3 years of treatment. Twenty-five percent of patients showed an improvement of ≥4.0 points on the ADAS-Cog after 12 months. By month 36, ∼7% of patients still maintained this improvement.

Long-term data on rivastigmine from a historic-controlled meta-analysis⁹⁰ of 2010 patients showed that following 1 and 2 years of rivastigmine treatment, ADAS-Cog scores had deteriorated by 4.0 and 5.0 points less than those predicted for untreated patients (Figure 2). MMSE scores for cognition and global rating scales supported these results. The model used in this study projected the natural course of dementia in untreated patients using the Stern equation.⁹¹ In this analysis, the actual baseline values of the patients were entered into the equation to obtain a more accurate comparator group. In a separate open-label extension study⁹² (N = 34), 2 years of rivastigmine treatment had a favorable effect on mood symptoms, activity disturbances, hallucinations, and paranoid behaviors.

Figure 2.

Scores on the Alzheimer's Disease Assessment Scale⁸⁴ cognitive subscale (ADAS-Cog; 0–70 points; higher scores = impairment) during 2 years of rivastigmine treatment and projected scores in untreated patients (mean [95% CI]). (Adapted with permission.⁹⁰)

Hypothetically, the continued long-term efficacy of rivastigmine may be a result of sustained AChE and BuChE inhibition along the continuum of disease severity. A lack of BuChE inhibition, combined with increased AChE activity, suggests that neither donepezil nor galantamine would provide the same level of long-term efficacy as rivastigmine. The results of a large, ongoing, head-to-head study (EXCEED [Exelon^® Comparison of Efficacy versus Donepezil]), assessing the safety, tolerability, and efficacy of rivastigmine and donepezil over 2 years will provide direct evidence of the comparative long-term effects of donepezil and rivastigmine.

Potential Effects on Disease Progression

The efficacy of ChE inhibitors is well established in terms of providing symptomatic benefits to patients with AD. However, interest in the potential disease-modifying effects of these agents is growing. At present, only limited data⁹ are available to examine this hypothesis. For example, delayed-start clinical trials in patients with mild to moderately severe AD (baseline MMSE scores, 10–26) indicate that any delay in the initiation of treatment with ChE inhibitors may result in irretrievable loss of benefit compared with patients receiving treatment earlier in the disease course. Two 52-week studies^9,93 were conducted, with an initial 26-week placebo-controlled phase followed by a 26-week, open-label extension study during which all patients received rivastigmine. In addition to providing symptomatic benefit in patients with mild to moderate AD⁹ and more severe disease,⁹³ the results confirm that rivastigmine treatment may modify the natural progression of the disease. Patients initially treated with placebo failed to “catch up” with those individuals who received the drug for the full 52 weeks (Figure 3).

Figure 3.

Long-term effects of rivastigmine on cognition. Mean change from baseline in Alzheimer's Disease Assessment Scale⁸⁴ cognitive subscale (0–70 points; higher scores = impairment) score following 1 year of treatment with rivastigmine. Patients with Global Deterioration Scale (1–7 points; higher scores = impairment) score ≥5; observed cases data set. ^∗P<0.05 versus original placebo. (Adapted with permission.⁹³)

High levels of both AChE and BuChE, in particular the G1 forms of these enzymes, are associated with the neurotoxic plaques characteristic of AD.²⁹ While AChE is present in both diffuse and compact plaques,⁹⁴ BuChE becomes associated with the plaques at approximately the same time as β-amyloid protein (Aβ) assumes the neurotoxic compact β-pleated conformation.⁹⁵ The association of both AChE and BuChE with Aβ has led to the suggestion that they are actively involved in the malignant transformation of plaques.^95,96 Supporting this is the observation that AChE colocalizes with Aβ deposits in the AD brain.⁹⁷ AChE may bind to a protease-sensitive nonamyloidogenic conformer of Aβ and induce a conformational transition into a partly protease-resistant amyloidogenic conformer of Aβ, and thus to amyloid fibrils.⁹⁸ Therefore, AChE directly promotes the formation of amyloid fibrils and stable Aβ-AChE complexes.⁹⁹ These complexes can change the pharmacologic and biochemical properties of AChE and increase the neurotoxicity of the Aβ fibrils.^100,101 Similarly, the addition of BuChE to Aβ in tissue culture results in a dramatic increase in neurotoxicity, although the mechanism underlying this effect is unknown.¹⁰² These data suggest that rivastigmine, as a dual ChE inhibitor, may possess a greater disease-modifying potential than the AChE-specific inhibitors donepezil and galantamine, although further studies in this area are warranted. As mentioned earlier, long-term rivastigmine treatment results in sustained ChE inhibition,²¹ whereas long-term administration of both donepezil and galantamine has been shown to increase AChE activity,^18–20 which may exacerbate the rate of disease progression.

DISCUSSION

ChE inhibitors offer an evidence-based approach to delaying the symptomatic decline seen in patients with AD. All ChE inhibitors prevent the hydrolysis of ACh and have demonstrated efficacy, first in clinical trials and more recently in clinical practice. Although all these agents share a common mechanism of action, we know that the drugs differ pharmacologically. These differences can significantly affect the safety, tolerability, and efficacy profiles of each drug and may ultimately determine the success of treatment.

This article has examined the 2 key stages of treatment from the physician's perspective: “getting [a patient] on” treatment and, once a patient is stabilized at a therapeutic dose, “staying on” treatment. After AD has been diagnosed, the patient should receive ChE inhibitor treatment as early as possible. The most common AEs experienced during the initial stages of treatment are class-specific gastrointestinal AEs. Some of these (eg, nausea, vomiting) arise, in part, from a rapid increase in central ACh levels and, therefore, are common to all agents in the ChE inhibitor class. When starting initial treatment, a slow, flexible dose-escalation schedule should be used, allowing at least 4 weeks before any dose increase. This type of schedule serves to substantially decrease the incidence of these acute AEs, thereby improving the likelihood that the patient will continue with treatment. This protocol should be followed with all 3 currently available ChE inhibitors. In addition, for rivastigmine and galantamine, patients should be advised to take their medication with morning and evening meals to further reduce the likelihood of tolerability problems.

After “getting on” treatment, when the patient is stabilized at a usual therapeutic dose, the second and, indeed, final stage of treatment is “staying on,” or maintenance, treatment. In contrast to observations during the initial phase of treatment, the safety, tolerability, and symptomatic efficacy of ChE inhibitors may differ significantly among the available agents during maintenance treatment, largely due to differences in their pharmacologic profiles.

CONCLUSIONS

All ChE inhibitors provide symptomatic benefit across the key symptom domains of AD. Of the 3 currently available agents, rivastigmine appears to show a particularly favorable profile during maintenance therapy. It is likely that specific pharmacologic characteristics, including dual and sustained AChE and BuChE inhibition, brain versus peripheral and brain region selectivity, and CYP450-independent metabolism may contribute to the good safety, tolerability, and efficacy profiles observed with this agent in clinical practice.