Prospects for Pharmacological Intervention in Alzheimer Disease

NEUROLOGICAL REVIEW

SECTION EDITOR: DAVID E. PLEASURE, MD Prospects for Pharmacological Intervention in Alzheimer Disease

Ge´rard Emilien, PhD; Konrad Beyreuther, PhD; Colin L. Masters, MD; Jean-Marie Maloteaux, MD, PhD

lzheimer disease (AD) involves neuronal degeneration with impaired cholinergic trans- mission in the cerebral cortex and hippocampus in areas of the brain particularly associated with memory and higher intellectual functioning. Other neurotransmitter deficits also occur, but the mechanisms underlying the widespread impairment of synaptic functionsA remain uncertain. Research on the molecular basis of AD has elucidated a pathogenic pathway from which a range of rational pharmacological interventions has emerged. Although at least 3 cholinesterase inhibitors (tacrine hydrochloride, donepezil, and rivastigmine tartrate) are now available and provide patients with modest relief, the most promising strategy involves approaches to retarding, halting, or preventing the formation or accumulation of ␤-amyloid (A␤) plaques. Estrogen is believed to have antioxidant or other anti-A␤ effects, as hormonal replacement therapy in women with menopause is associated with a reduced risk or delayed onset of AD. The association between nonsteroidal anti-inflammatory drugs and a reduced risk of AD has not yet been confirmed, but these agents may protect the brain from the reactive glial and microglial responses associated with A␤ deposition. Also, recent studies suggested that antioxidants, such as vitamin E taken alone or in combination with selegiline hydrochloride, can delay the progression of AD. Despite these encouraging results, no current therapy has been shown to halt or reverse the underlying disease process. The proof of the principle that anti-A␤ drugs will work in the transgenic models of AD is eagerly awaited with the expectation that they will eventually prove suc- cessful in humans. Arch Neurol. 2000;57:454-459 The accumulation of ␤-amyloid (A␤) effects of cholinergic-boosting strategies. plaques is the pathognomonic feature of In addition to the 3 licensed compounds Alzheimer disease (AD). How does this ac- (tacrine hydrochloride, donepezil, and ri- cumulation relate to the neuronal degen- vastigmine tartrate), there are many drugs eration that manifests as a progressive cog- awaiting approval or undergoing phase 3 nitive impairment with widespread trials (Table 1). While drugs specifi- neurological and neuropsychiatric distur- cally targeting the ␤-amyloidogenic path- bances? The slowly emerging answer is way only now are beginning to emerge in that A␤ induces a variety of neurotoxic a preclinical setting, most other drugs are phenomena, including reactive oxygen directed at the cholinergic system. There species. However, to date only the sec- are many psychotropic agents available to ondary degenerative effects have been ame- treat the behavioral manifestations of nable to therapy, as seen in the beneficial AD, including antipsychotic, agitation- reducing , antidepressant, anxiolytic, and sedative-hypnotic drugs. Interventions in From the Laboratory of Pharmacology (Drs Emilien and Maloteaux) and Department of Neurology (Dr Maloteaux), Universite´ Catholique de Louvain Cliniques AD include treatment of the underlying Universitaires Saint Luc, Brussels, Belgium; the Centre for Molecular Biology, disease process and amelioration of neu- University of Heidelberg, Heidelberg, Germany (Dr Beyreuther); and the Department rochemical deficits produced by the cel- of Pathology, University of Melbourne, and the Mental Health Research Institute, lular changes. This review discusses cur- Parkville, Australia (Dr Masters). rent perspectives in the pharmacotherapy

ARCH NEUROL / VOL 57, APR 2000 WWW.ARCHNEUROL.COM 454

©2000 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/23/2021 of AD and examines how the different disciplinary approaches are being incorporated into clinical research for Table 1. Relevant Drugs for Alzheimer Disease Awaiting effective drug treatments. Attention is drawn to new com- Approval or Undergoing Phase 3 Trials* pounds with novel mechanisms of action that could have a tremendous impact in the future treatment of AD. Drug Action Awaiting Approval Trichlorfon MODULATION OF THE CHOLINERGIC SYSTEM AChE inhibitor Physostigmine salicylate Different strategies have been developed to boost the cho- Idebenone Antioxidant linergic system, including increased acetylcholine pro- Nebracetam m1 Muscarinic receptor agonist duction with cholinergic precursors (choline and leci- Nefiracetam thin), prevention of synaptic acetylcholine destruction Propentofylline ACh agonist, calcium channel opener, and with acetylcholinesterase (AChE) inhibitors, such as phosphodiesterase inhibitor tacrine (9-amino-1, 2, 3, 4-tetrahydroacridine, Cognex; Undergoing Phase 3 Trials Parke-Davis, Morris Plains, NJ), donepezil (developed un- Amiridin Eptastigmine AChE inhibitor der the code E2020, Aricept; Pfizer Inc, New York, NY), Galantamine rivastigmine (developed under the code SDZ ENA 713, Cevimeline m1 Muscarinic receptor agonist Exelon; Novartis Pharmaceuticals, East Hanover, NJ), phy- hydrochloride sostigmine salicylate, and galantamine, or direct stimu- Talsaclidine lation of postsynaptic muscarinic receptors with recep- Dehydroepiandrosterone Neurosteroid tor agonists. However, tacrine and donepezil are the only Montirelin hydrate ACh release stimulator, protirelin agonist NS-105 Nootropic agent and ACh and drugs that have been approved by the Food and Drug Ad- GABA modulator ministration (FDA). Selegiline hydrochloride Monoamine oxidase B inhibitor Evidence now indicates that some AChE inhibitors Taltirelin hydrate Protirelin agonist also may provide neuroprotective effects, perhaps through the activation of nicotinic receptors, and may even en- *AChE indicates acetylcholinesterase; ACh, acetylcholine; and GABA, hance neurotrophic regeneration. Other possible ac- ␥-aminobutyric acid. tions include the effect of cholinergic agonists on the processing and secretion of the amyloid precursor protein Donepezil and A␤.1,2 In November 1996, donepezil (a piperidine-based AChE Tacrine inhibitor with specificity for AchE) was approved by the FDA as a symptomatic therapy for mild to moderate AD. After the initial positive and overly optimistic reports in The bioavailability of donepezil is approximately 100%, 1986 on the efficacy of tacrine, it was subsequently noted with peak plasma concentrations occurring between 2 and to be an even stronger inhibitor of the butyrylcholines- 4 hours after an oral dose. Food appears to have no sig- terase family of enzymes. More recently, apart from AChE nificant effect on the drug absorption. Donepezil has a mean inhibition, tacrine has been shown to possess a much elimination half-life of 70 hours, with significant interin- broader pharmacological profile, such as blockage of po- dividual variation; a daily dose is recommended. Done- tassium channels, inhibition of the neuronal mono- pezil is bound highly (93%-96%) to the proteins albumin amine uptake processes, and inhibition of monoamine and ␣1-acid glycoprotein. The drug is metabolized in the oxidase.3 The heightened efficacy of tacrine in alleviat- liver by CYP2D6 and CYP3A3/4 and by glucuronidation. ing some of the behavioral symptoms of AD compared A dosage of 5 mg/d yields steady-state AChE inhibition with other AChE inhibitors might be related to these other of approximately 64% as determined by cholinesterase in- pharmacological actions. The purported cognitive- hibition in human red blood cell samples.10 In a 30-week enhancing effects of tacrine and the AChE inhibitors are phase 3 clinical trial of donepezil, both the 5- and 10-mg often difficult to disentangle from their nonspecific arousal treatment groups had ADAS-Cog (Alzheimer’s Disease As- and behavioral effects, which can be expected from all sessment Scale-Cognitive subscale) scores superior to the classes of cholinergic stimulants. Serious adverse effects placebo group throughout the 6-month trial. Moreover, of tacrine, including hepatotoxic effects,4 have weak- more than 80% of the patients in the treatment group ened its position as a drug of choice. showed either improvement or no decline during the Tacrine has a mean bioavailability of 17%, with in- 6-month trial. The long-term efficacy of donepezil treat- terindividual variability from 2% to 36% (Table 2).10 ment has not been evaluated yet. However, the efficacy for This low bioavailability is thought to be secondary to large up to 2 years was evaluated in patients who completed the presystemic clearance. Food appears to decrease the rate 30-week phase 3 trial and who underwent a long-term, but not the extent of absorption. Tacrine hydrochloride open-label study with donepezil. For a mean of 40 weeks, is rapidly metabolized, with mean half-lives of 1.6 and patients maintained performance levels better than their 2.1 hours after single doses of 25 mg and 50 mg, respec- original baseline scores. The ADAS-Cog scores collected tively, which must be taken 4 times a day. Tacrine me- for more than 2 years suggested that patients receiving tabolism appears to be mediated through cytochrome donepezil maintained the same magnitude of benefit as in P-450 1A 2 isoenzyme. Clinical dosages of 80 to 160 mg/d the beginning of the study, indicating that long-term use usually achieve approximately 30% AChE inhibition. of donepezil may be beneficial.11

ARCH NEUROL / VOL 57, APR 2000 WWW.ARCHNEUROL.COM 455

Rivastigmine is a neuronal selective AChE inhibitor that Galantamine, a naturally occurring amarylidacea alka- is still under clinical investigation. Results of phase 2 tri- loid, is a long-acting, selective, reversible, and competi- als showed that patients with AD tolerated up to 12 mg/ tive AChE inhibitor. Patients with AD who took galan- d.12 Adverse effects did not include hepatotoxic effects. tamine had improved performance on memory tests, and The results of a meta-analysis of 3 phase 3 trials demon- it was well tolerated.14 Galantamine use was approved in strated significant beneficial effects on measures of cog- Austria. nition using the ADAS-Cog scale, global functioning, and activities of daily living. The Swiss regulatory authority Nicotinic Cholinergic Strategies approved rivastigmine in August 1997 for the treatment of patients with mild to moderate AD. In May 1998, ri- It has been suggested that there is an inverse relation- vastigmine received marketing approval from the Euro- ship between smoking and AD.15 Dose dependency has pean Medicine Evaluation Agency, London, England, and not yet been shown for this protective action. Subcuta- is currently awaiting approval from the FDA. neous administration of nicotine has been claimed to improve attention and information processing in patients Metrifonate with AD. Interestingly, these effects were more evident than memory improvement. Further clinical investiga- Metrifonate is an AChE inhibitor that acts as a prodrug tions are clearly required to confirm these results. for the direct, long-acting inhibitor DDVP (2,2-dimethyl- dichlorovinyl phosphate). In blood samples, metrifo- MODULATION OF OTHER nate has a mean half-life of 2.3 hours and DDVP has a NEUROTRANSMITTER SYSTEMS half-life of 3.8 hours.7 Thus, prolonged elevation of acetylcholine levels can be achieved.13 Estimates of the half- There is a growing body of evidence that disturbances life for cholinesterase recovery vary depending on study of glutamatergic neurotransmission may underlie a mecha- methods, with a mean ± SD of 26.6 ± 15.2 hours. Re- nism of neurotoxic excitatory amino acids contributing cently, after some patients in clinical trials experienced to cognitive deficits in patients with AD.16 Age-related muscle weakness, the request for approval for metrifo- changes in NMDA (N-methyl D-aspartate) receptors have nate in Europe was withdrawn. been found in cortical areas and in the hippocampus in

Table 2. Pharmacokinetic and Pharmacodynamic Profiles of Some Drugs Used or in Clinical Development for the Treatment of Alzheimer Disease*

Protein Common Adverse Effects Mechanism Half-life, Starting Bioavailability, Binding, Leading to Treatment Development Drug of Action h Oral Dosage % Food Effect % Discontinuation Status Tacrine Cholinesterase 1.60-2.14 10 mg, 4 times 2.4-36.0 Decreases 75 Nausea, vomiting, Registered with hydrochloride inhibition a day effectiveness by increased salivation, the FDA (since 30%-40%, take sweating, lacrimation, 1993) and the 1 h before meals and elevated EMEA transaminase levels Donepezil5 Cholinesterase 50-70 5 mg/d 100 None 93-96 Nausea, diarrhea, Registered with inhibition insomnia, and the FDA (since vomiting 1996) and the EMEA Rivastigmine Cholinesterase NA 2 mg/d, slow Approximately NA NA Nausea, vomiting, Awaiting FDA tartrate6 inhibition titration to 35 diarrhea, dizziness, approval and 6-12 mg, 2 and headaches were approved by or 3 times a evident at high doses the EMEA day (May 1998) Trichlorfon7 Cholinesterase 2.3 Daily, possibility Low Decreases NA Gastrointestinal Filed with the FDA inhibition of weekly maximum disturbances (November plasma 1997) and concentration; the EMEA prolonged time to (October 1997) reach maximum concentrations Galantamine8 Cholinesterase 5.7 20-50 mg/d 90-100 NA Negligible Nausea, vomiting, Approved in inhibition agitation, and sleep Austria disturbances Propentofylline9 Adenosine NA 300 mg, 3 Approximately NA NA Gastrointestinal Awaiting approval reuptake times a day 32 disturbances, by the FDA inhibition dizziness, headaches, and heartburn

*FDA indicates Food and Drug Administration; EMEA, European Medical Evaluation Agency; NA, data not available.

ARCH NEUROL / VOL 57, APR 2000 WWW.ARCHNEUROL.COM 456

©2000 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/23/2021 Table 3. Classes of Drugs in Preclinical or Early Clinical Development for the Treatment of Alzheimer Disease (AD)*

Mechanism of Action Drug Comment Antiexcitotoxins L-701252 Being developed for treatment of AD, epilepsy, and cerebrovascular ischemia NMDA antagonist LY-235959 Competitive antagonist; potential for use in AD and other CNS diseases WIN-63480-2 Uncompetitive antagonist; does not produce phencyclidine-like effects in animals Neurotrophic AIT-082 Undergoing phase 2 clinical trials AK-30-NGF Monoclonal antibody; NGF delivery system NGF agonist NBI-106 Potent immune stimulation and memory-enhancing properties rhNGF Recombinant protein; also undergoing phase 3 clinical trials for peripheral neuropathy therapy Hormonal ABPI-124 Specific for CNS; does not interact with other tissues Estrogen Neurestrol An estrogen agonist developed by Endocon Inc, South Walpole, Mass, for treatment of women with AD Anti-inflammatory Anti-inflammatory agent SC-110 Undergoing phase 1 clinical trials Cox 2 inhibitor GR-253035 IC50 to COX-1 of more than 100 µmol/L and IC50 to COX-2 of 0.14 µmol/L; entering phase 1 clinical trials for treatment of AD Cytokine modulator NBI-117 Reported to bind and activate newly discovered receptors of cytokine activin Antioxidants ARL-16556 Spin-trapping effects that scavenge free radicals and the ability to modulate the effects of nitric oxide Antioxidant indicate that it may have advantages over existing compounds; undergoing phase 1 clinical trials MDL-74180DA Undergoing preclinical trials; analog of vitamin E that inhibits in vitro and ex vivo lipid oxidation and protects mice against CNS trauma Antiamyloid MDL-28170 Dipeptide aldehyde (Z-Val-Phe-H); cysteine protease inhibitor Calpeptin Dipeptide aldehyde (Z-Leu-Nle-H); cysteine protease inhibitor ␥-Secretase inhibitor† MG-132 This tripeptide aldehyde (Z-Leu-Leu-Leu-H) and others at low concentrations inhibit cysteine proteases and the degradation of the cytoplasmic domains of amyloid precursor protein; at higher concentrations they inhibit ␥-secretase and proteasome MW-167 Substrate-based difluoroketone, aspartyl protease inhibitor Antifibrillogenic SKF-74652 In a model of A␤ aggregation, an IC50 of 28 µmol/L

*NMDA indicates N-methyl D-aspartate; CNS, central nervous system; NGF, nerve growth factor; rhNGF, recombinant human NGF; Cox, cyclooxygenase; IC50, the inhibition concentration at 50%; and A␤, ␤-amyloid. †␥-Secretase may be one mechanism involved, but further study is needed.

many species. Based on these findings, several strategies utility of NGF as a neuroprotective treatment for basal have been developed to improve cognition by the use of forebrain cells in AD.18 An innovation in AD therapy may NMDA antagonists as neuroprotective agents to slow the come from NGF-mimetic drugs. Neotrofin or AIT-082 progression of dementia. These antagonists include dex- (an analog of hypoxanthine) (NeoTherapeutics, Irvine, tromethorphan hydrobromide, memantine (a congener Calif) is an orally active compound that is claimed to of amantadine hydrochloride),17 and nitroglycerine. Pre- enhance the levels of various neurotrophic factors, such liminary clinical evidence suggests that memantine use as NGF, ciliary neurotrophic factor, and neurotrophin-3, may improve symptoms of dementia. Functional im- and also potentiate the effects of NGF. provement and reduced care dependence were observed in patients with severe dementia who used meman- DECREASING THE CELLULAR REACTION tine. Beneficial effects also were noted in other behavioral TO NEURODEGENERATION patterns. Current drugs in preclinical development include L-701252, LY-235959, and WIN-63480-2 Microglial cells, closely related to the macrophage series (Table 3). of cells in the periphery, increase in size and number in the brain with AD. From this observation and the pres- NEUROTROPHIC GROWTH FACTORS ence of complement in amyloid plaques, the concept of AD as an inflammatory disease has emerged. It has been Nerve growth factor (NGF) as the prototypic neuro- reported that individuals taking anti-inflammatory drugs, trophic growth factor is intimately related to the main- such as nonsteroidal anti-inflammatory drugs (NSAIDs), tenance of function of the cholinergic basal forebrain sys- have fewer cerebral microglia19 and are less likely to de- tem. Forebrain cholinergic neurons are the only cells in velop AD, with a fairly consistent risk reduction of about the adult brain that express high amounts of the low- 50%.20 The greatest protection is observed in those indi- affinity p75 receptors for NGF. The NGF increases hip- viduals with late onset (Ͼ70 years of age) and a strong fam- pocampal acetylcholine and prevents cholinergic cell loss ily history of AD. In one randomized, placebo-controlled and atrophy after fornix lesions, indicating the potential study21 with indomethacin sodium in 44 patients with mild

ARCH NEUROL / VOL 57, APR 2000 WWW.ARCHNEUROL.COM 457

©2000 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/23/2021 AD a slight improvement in cognitive function from base- also remains uncertain, placing some restriction on our line was noted in the treatment group after 6 months of ability to predict the unintended adverse effects of APP treatment. down-regulation. New products with anti-inflammatory properties in Most therapeutic research has been devoted to de- clinical development for AD include SC-110 and GR- veloping inhibitors of the ␤- and ␥-secretases, which are 253035, a cyclooxygenase 2 (COX-2) inhibitor. A DNA- responsible for the proteolytic cleavage events that gen- binding protein, NF-␬B, has an important role in driv- erate A␤. While the identity of these enzymes remains ing transcription from inflammation-related genes, such unknown, several pharmaceutical companies have de- as in the production of COX-2, that operate in stressed veloped compounds that are efficient inhibitors of ␥-secre- tissues like the brain with AD. Blocking intervention at tase (Table 3). Most are still undergoing preclinical de- this stage may be a potential approach for treatment. velopment, although registration of some compounds may Propentofylline (3-methyl-1-[5-oxohexyl]-7-propyl- occur in 2000. Studies26,27 that implicate the presenilins xanthine), a xanthine derivative and an inhibitor of the in the ␥-secretase pathway also are introducing new thera- reuptake of adenosine, was reported to inhibit glutamate peutic strategies, although the involvement of presenil- release, to increase cerebral blood flow, and to act mainly ins in Notch signaling has caused some concern about on astrocytes and microglia.9 It appears to delay the pro- the potential adverse effects of ␥-secretase inhibitors. gression of mild to moderate dementia in addition to pro- Compounds directed at inhibiting the toxic effects viding symptomatic relief. In a 12-month, randomized, of A␤ or stabilizing the aggregated forms of A␤ to pro- placebo-controlled study22 with propentofylline in patients mote its clearance from the brain are now undergoing with AD and vascular dementia, the total patient popu- active development. Further insight is required to un- lation showed statistically significant treatment differences derstand the roles of other proteins or lipids (eg, cho- in favor of propentofylline for the global measures of de- lesterol) that interact with A␤ (such as apolipoprotein E mentia and cognitive evaluations. Propentofylline is await- and ␣2-macroglobulin) or with APP as it travels through ing registration in the European Union and Canada for use the cell toward its biogenesis of A␤. Recently, a remark- in mild to moderate AD and vascular dementia. able approach was described in which transgenic mice immunized with human A␤ showed attenuation of amy- DRUGS THAT REDUCE OXIDATIVE STRESS loid plaque formation.28 This attenuation may represent a novel mechanism for promoting the clearance of A␤ There is increasing evidence that the brain with AD is from the brain, as the rates of A␤ production were not under severe oxidative stress, as a result of either A␤- altered. The prospect of large-scale human immuniza- mediated generated oxyradicals or perturbed ionic cal- tions with potential autoantigens raises considerable cium balances within neurons and their mitochondria. challenges. Drugs, such as vitamin E,23 idebenone,24 or estrogen, that have strong antioxidant properties are showing variable THE EMERGING FIELD degrees of efficacy. Newer antioxidants undergoing clini- OF PHARMACOGENETICS cal development include the free radical scavenger ARL- 16556, an ␣-phenyl-t-butylnitrone derivative with spin- As in all complex diseases, many genetic elements are re- trapping effects. sponsible for the clinical phenotype. Predicting who, in Estrogen replacement therapy in healthy postmeno- a mixed population, will respond best to any given thera- pausal women favorably affects mood and may have mod- peutic compound is a challenge for pharmacogeneti- est effects on cognitive function. Many clinical trials are cists. There are already some indicators that the apoli- now in progress. In addition to its antioxidant properties, poprotein E allotype may affect responses to AChE estrogen may promote growth of cholinergic neurons, re- inhibitor therapy.29 duce plasma levels of apolipoprotein E, modify inflammatory responses, or even directly reduce A␤ generation.25 CONCLUSIONS New products undergoing preclinical development include an estrogen-based drug (sustained-release formu- Much has been learned from the first few years of lation Neurestrol; Endocon Inc, South Walpole, Mass) and specifically targeted therapy for AD. A comparison of ABPI-124, a compound without the adverse effects of femi- tacrine with other second-generation AChE inhibitors in nization.Theeffectsofraloxifenehydrochlorideandtamoxi- clinical studies shows that despite these drugs having mod- fene citrate (nonsteroidal selective estrogen receptor modu- est clinical efficacy, their main differences are in the fre- lators) in AD are yet to be determined. quency of adverse effects, number of dropouts, and per- centage of patients whose conditions improve. Although THERAPEUTIC STRATEGIES efficacy may be similar between the AChE inhibitors at IN THE AMYLOID PRECURSOR PROTEIN effective doses, peripheral cholinergic adverse effects, tol- AND ␤-AMYLOIDOGENIC PATHWAY erability, and hepatotoxic effects are severe limitations. The controlled studies using AChE inhibitors have gen- The gene dosage effect of trisomy 21 syndrome has shown erally been short-term, from 12 weeks to 6 months, and that down-regulation of amyloid precursor protein (APP) use similar kinds of cognitive outcome measures. There- expression is a theoretical therapeutic strategy, but our fore, long-term (Ͼ1 year) controlled studies need to be understanding of the factors that regulate APP transcrip- evaluated. Furthermore, reliable controlled data on mean- tion are still too rudimentary. The normal function of APP ingful outcomes, such as dependency and institutional-

ARCH NEUROL / VOL 57, APR 2000 WWW.ARCHNEUROL.COM 458

©2000 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/23/2021 ization or other aspects of long-term efficacy, are ur- 9. Mielke R, Moller HJ, Erkinjuntti T, Rosenkranz B, Rother M, Kittner B. Propen- gently needed. tofylline in the treatment of vascular dementia and Alzheimer-type dementia: over- view of phase I and phase II clinical trials. Alzheimer Dis Assoc Disord. 1998; In contrast to the AChE inhibitors, the beneficial ef- 12(suppl 2):S29-S35. 30 fect of estrogen therapy may delay the progression of AD. 10. Parnetti L. Clinical pharmacokinetics of drugs for Alzheimer’s disease. Clin Phar- Since combination therapies may be crucial, it will be in- macokinet. 1995;29:110-129. teresting to perform trials with combinations of drugs that 11. Rogers SL, Perdomo C, Friedhoff LT. Clinical benefits are maintained during long- possess different mechanisms of action; for example, term treatment of Alzheimer’s disease with the acetylcholinesterase inhibitor E2020. Eur Neuropsychopharmacol. 1995;5:386-387. AChE inhibitors and estrogen. Clinical studies are nec- 12. Gottwald MD, Rozauski RI. Rivastigmine: a brain-region selective acetylcholin- essary to assess the efficacy and interactive effects of these esterase inhibitor for treating Alzheimer’s disease. Expert Opin Invest Drugs. 1999; approaches. With regard to the use of NSAIDs in AD, el- 8:1643-1682. derly people with AD are more susceptible to the ad- 13. Itoh A, Nitta A, Katono Y, et al. Effects of metrifonate on memory impairment verse effects of NSAIDs; therefore, these drugs should be and cholinergic dysfunction in rats. Eur J Pharmacol. 1997;322:11-19. 14. Thomsen T, Bickel U, Fischer JP, Kewitz H. Galantamine hydrobromide in a long- used with caution. The development of COX-2 and leu- term treatment of Alzheimer’s disease. Dementia. 1990;1:46-51. kotriene inhibitors might be very important. The use of 15. Ulrich J, Johannson-Locher G, Seiler WO, Sta¨helin HB. Does smoking protect antioxidants, such as vitamin E, is worth considering in from Alzheimer’s disease? Alzheimer-type changes in 301 unselected brains from patients with AD, since they can be obtained over the patients with known smoking history. Acta Neuropathol (Berl). 1997;94:450- 454. counter and are relatively nontoxic and inexpensive. 16. Mu¨ller WE, Mutschler E, Riederer P. Noncompetitive NMDA receptor antago- In summary, the knowledge gained to date has served nists with fast open-channel blocking kinetics and strong voltage-dependency to set the standards by which all future therapies for AD as potential therapeutic agents for Alzheimer’s dementia. Pharmacopsychiatry. will be measured. Progress in the development of drugs 1995;28:113-124. and clinical trials for AD has been remarkable, with ev- 17. Winblad B, Poritis N. Memantine in severe dementia: results of the M-9 BEST Study (Benefit and Efficacy in Severely Demented Patients During Treatment With ery prospect that more effective strategies will emerge in Memantine). Int J Geriatr Psychiatry. 1999;14:135-146. the near future. 18. Holtzman DM, Li Y, Chen K, Gage FH, Epstein CJ, Mobley WC. Nerve growth factor reverses neuronal atrophy in a Down syndrome model of age-related neuro- Accepted for publication September 20, 1999. degeneration. Neurology. 1993;43:2668-2673. Reprints: Ge´rard Emilien, PhD, 127 Rue Henri Prou, 19. Mackenzie IR, Munoz DG. Nonsteroidal anti-inflammatory drug use and Alzheimer- type pathology in aging. Neurology. 1998;50:986-990. 78340 Les Clayes-Sous-Bois, France (e-mail: gemilien 20. Breitner JC, Welsh KA, Helms MJ, et al. Delayed onset of Alzheimer’s disease @aol.com). with nonsteroidal anti-inflammatory and histamine H2 blocking drugs. Neuro- biol Aging. 1995;16:523-530. 21. Rogers J, Kirby LC, Hempelman SR, et al. Clinical trial of indomethacin in Alz- REFERENCES heimer’s disease. Neurology. 1993;43:1609-1611. 22. Marcusson J, Rother M, Kittener B, et al, for the European Propentofylline Study 1. Jope RS, Song L, Powers RE. Cholinergic activation of phosphoinositide signaling Group. A 12-month, randomized, placebo-controlled trial of propentofylline (HWA is impaired in Alzheimer’s disease brain. Neurobiol Aging. 1997;18:111-120. 285) in patients with dementia according to DSM III-R. Dement Geriatr Cogn Dis- 2. Nitsch RM, Slack BE, Wurtman RJ, Growdon JH. Release of Alzheimer amyloid ord. 1997;8:320-328. precursor derivatives stimulated by activation of muscarinic acetylcholine re- 23. Sano M, Ernesto C, Thomas RG, et al. A controlled trial of selegiline, alpha- ceptors. Science. 1992;258:304-307. tocopherol, or both as treatment for Alzheimer’s disease. N Engl J Med. 1997; 3. Jossan SS, Adem A, Winblad B, Oreland L. Characterisation of dopamine and 336:1216-1222. serotonin uptake inhibitory effects of tetrahydroaminoacridine in rat brain. Phar- 24. Gutzmann H, Hadler D. Sustained efficacy and safety of idebenone in the treat- macol Toxicol. 1992;71(3, pt 1):213-215. ment of Alzheimer’s disease: update on a 2-year double-blind multicentre study. 4. Knapp MJ, Knopman DS, Solomon PR, Pendlebury WW, Davis CS, Gracon SI, J Neural Transm Suppl. 1998;54:301-310. for the Tacrine Study Group. A 30-week randomized controlled trial of high- 25. Xu H, Gouras GK, Greenfield JP, et al. Estrogen reduces neuronal generation of dose tacrine in patients with Alzheimer’s disease. JAMA. 1994;271:985-991. Alzheimer ␤-amyloid peptides. Nat Med. 1998;4:447-451. 5. Tiseo PJ, Rogers SL, Friedhoff LT. Pharmacokinetic and pharmacodynamic pro- 26. Hardy J, Israe¨l A. Alzheimer’s disease: in search of ␥-secretase. Nature. 1999; file of donepezil HCl following evening administration. Br J Clin Pharmacol Suppl. 398:466-467. 1998;46:13-18. 27. Haass C, De Strooper B. The presenilins in Alzheimer’s disease: proteolysis holds 6. Polinsky RJ. Clinical pharmacology of rivastigmine: a new-generation acetylcho- the key. Science. 1999;286:916-919. linesterase inhibitor for the treatment of Alzheimer’s disease. Clin Ther. 1998; 28. Schenck D, Barbour R, Dunn W, et al. Immunization with amyloid-␤ attenuates 20:634-647. Alzheimer-disease-like pathology in the PDAPP mouse. Nature. 1999;400:173- 7. Ringman JM, Cummings JL. Metrifonate (trichlorform): a review of the phar- 177. macology, pharmacokinetics and clinical experience with a new acetylcholin- 29. Richard F, Helbecque N, Neuman E, Guez D, Levy R, Amouyel P. APOE geno- esterase inhibitor for Alzheimer’s disease. Expert Opin Invest Drugs. 1999;8: typing and response to drug treatment in Alzheimer’s disease. Lancet. 1997; 463-471. 349:539. 8. Rainer M. Galanthamine in Alzheimer’s disease: a new alternative to tacrine? CNS 30. Paganini-Hill A, Henderson VW. Estrogen replacement therapy and risk of Alz- Drugs. 1997;7:89-97. heimer disease. Arch Intern Med. 1996;156:2213-2217.

ARCH NEUROL / VOL 57, APR 2000 WWW.ARCHNEUROL.COM 459