Special Report For reprint orders, please contact
[email protected]
Can nicotine be used medicinally in Parkinson’s disease? Expert Rev. Clin. Pharmacol. 4(4), 429–436 (2011)
Claire Thiriez1, Gabriel Villafane1, Frédérique Grapin1, Gilles Fenelon1,2, Philippe Remy1,3 and Pierre Cesaro†1,2 Department of Neurology, University Hospital Henri Mondor and Paris 12 University (UPEC), Créteil, France 2 INSERM U 955, Créteil, France 3 CEA, I2BM, Service Hospitalier Frédéric Joliot, Orsay, France † Author for correspondence: Tel.: +33 149 812 303 Fax: +33 149 812 326
[email protected] 1
www.expert-reviews.com
The risk of Parkinson’s disease is reduced by cigarette smoking, which raises some unanswered questions. Nicotine, a major component of tobacco smoke, could exert either nonreceptormediated biological effects or, more importantly, act on the different subtypes of nicotinic brain receptors, in particular those associated with the nigrostriatal dopaminergic pathway. There is now robust experimental evidence for a neuroprotective effect of nicotine upon dopaminergic neurons. By contrast, in animal models of Parkinson’s disease, nicotine alone has slight or no motor effects. However, nicotine may modulate dopamine transmission and has clear motor effects when associated with l-DOPA, reducing l-DOPA-induced dyskinesias. Clinical trials have yielded inconclusive results thus far and are hampered by different designs and small cohorts. Ongoing studies address either symptomatic motor or nonmotor symptoms, or neuroprotection. There is still no agreement on the daily dosage of nicotine or the method of administration. Together, these data suggest that nicotine or nicotinic receptor drugs have therapeutic potential for Parkinson’s disease, although the specific treatment regimens remain to be determined. Keywords : environment • neuroprotection • nicotine • Parkinson’s disease • smoking • transdermal • treatment
Although many symptoms of Parkinson’s disease (PD) are related to the neurodegeneration of the nigrostriatal dopaminergic pathway, other neurotransmitter systems are involved in PD, such as the nicotinic cholinergic system. The exact role of the latter system in the pathophysiology of PD is not fully understood; however, some major results should be highlighted. First, numerous studies converge to support a role of smoking in protection against PD [1] . Nicotine might be responsible for such an effect and experimental data suggest that it has a neuroprotective action [2] . Second, there is an interaction between the stimulation of nicotinic receptors and the release of dopamine in the striatum [3,4] . Interestingly, post-mortem and recent imaging studies show that the density of nicotinic receptors is reduced in the brain and especially in the nigrostriatal dopaminergic system of PD patients [5] and experimental models of PD [4] . Finally, despite controversial results, the question has been raised of a possible clinical benefit of nicotine in PD patients. The goal of the present article is to answer, to the best of our current knowledge, the actual question: ‘What is the evidence that medicinal use of nicotine in PD is clinically useful?’, using 10.1586/ECP.11.27
epidemiological, preclinical and clinical data. We searched the literature (PubMed) with several keywords: ‘nicotine and Parkinson’ yielded 154 results, and ‘smoking and Parkinson’ yielded 655. The Cochrane database reported no review, but nine clinical trials. We added papers from our personal files and older papers quoted among the reviews. Among search results, we mainly selected reviews for experimental data and original papers for all clinical trials. Ongoing trials were searched on the NIH website; more data were collected on the websites of Neuraltus Inc. and the Michael J Fox foundation website [101,102] . Epidemiology
Physicians working on PD have found nicotine of particular interest since it has been shown that there was less PD among smokers. It was firstly noticed in 1959 [6] and more than 50 epidemio logical studies have confirmed this observation [2,4] . Some suggested that this association could be due either to a selective mortality of smokers [7] (as PD affects more middle-aged and elderly people), to the disease affecting smoking habits or to a reporting bias. However, a case–control study [8] with 350 PD patients
© 2011 Expert Reviews Ltd
ISSN 1751-2433
429
Special Report
Thiriez, Villafane, Grapin, Fenelon, Remy & Cesaro
and 350 matched controls showed a relative risk of 0.52, with an intermediate risk for people who had quit. Furthermore, huge prospective cohort studies [9,10] brought even more credibility to the hypothesis of a protective role of smoking in PD. A pooled analysis confirmed the negative association between PD and smoking that was more pronounced for current smokers than for former ones [11] ; it also showed that there was a dose–response, with the ratio decreasing when the pack-years increase, and that other forms of tobacco also had this effect. Even passive smoking seems to be less frequent in PD than in controls [12] . Duration of smoking also seems to be more important than intensity [13] . A study in twin pairs in which at least one twin had PD found the same negative correlation: the twins without PD smoked significantly more than their brothers, even when smoking dose was calculated only until 10 years before PD onset [14] ; this argued against a genetic and behavioral hypothesis, as a link between genetical propensity to smoke and susceptibility to PD. A recent review on nicotine addiction suggested a potential interaction between multiple small-effect alleles and the environment [15] . The risk of developing PD in a cohort was 27% lower when both parents were smokers [16] ; but when the research was restricted to never-smokers, the correlation was no longer seen. The effect of parental smoking on susceptibility of PD was probably due to the higher proportion of smokers in offspring when parents smoke. As a study showed that this negative correlation was not found in the elderly [17] , the hypothesis was that smoking could delay the onset of PD. However, age at onset seemed to be the same in both populations of smokers and nonsmokers [17,18] . In addition, a study found no significant differences in progression of Parkinsonism, cognitive impairment and mood between smoking and nonsmoking PD patients [19] . The demonstration that there is a negative correlation between smoking and PD is of great interest as it may point to potential protective strategies for PD. However, there are numerous components in cigarette smoke that may contribute to this apparent neuroprotective effect. Interestingly, evidence using Parkinsonian animal models suggest that the nicotine in tobacco may play a role. Long-term studies in PD patients are now required to address this issue. Experimental data Motor effects
Administration of nicotine alone or in combination with an agonist of D2 receptors has slight or no motor effect in animals with nigrostriatal damage. By contrast, it can increase the motor effect when coadministered with l-DOPA methyl ester [20] . Moreover, motor effects can be reversed by nicotine receptor antagonists [2,21] . The presynaptic nicotinic receptors, which regulate dopamine release, seem to be involved in these actions [21,22] . In more recent experiments using a blinded motor evaluation, nicotine alone did not improve motor impairment, nor influence the motor effects of l-DOPA in Parkinsonian rats or in Parkinsonian monkeys [23,24] . However, in the same experimental series, administration of nicotine in the drinking water reduced the occurrence of dyskinesias by 50% [23,24] . 430
Neuroprotection
Besides in vitro studies, numerous results indicate a neuro protective effect of nicotine against various lesions (mechanical lesions, 6-hydroxydopamine and 1-methyl-4-phenyl-1,2,3,6tetrahydropyridine [MPTP]) [3,20] . Of particular interest is the neuroprotective effect against a slow MPTP-induced neuro degenerative lesion in monkeys [25] . However, when nicotine is administered in animals with a completed lesion, no curative effect is observed, neither in rodents nor in monkeys [26] . The mechanisms of this neuroprotective effect remain unknown: some are reversed by nicotine receptor antagonists or may not appear in knockout mice lacking α4β2 receptor subtypes [27] . A potential increase of trophic factors may involve nicotinic receptors, whereas antioxidant or other neuroprotective mechanisms are not mediated by receptors [2,3] . A direct action on brain mitochondria could involve a reduction of the synthesis of reactive oxygen species by complex 1 [28] . It has also recently been shown that nicotine and hydroquinone inhibit a-synuclein aggregation [29] . The molecular mechanism of neuroprotection may involve phosphorylation of Akt and upregulation of Bcl2 and Bclx [30] . Thus, experimental data indicate that chronic nicotine administration has a motor effect – particularly on dyskinesias – and a neuroprotective effect in Parkinsonian models. Studies in humans
Preclinical studies mentioned above suggest that nicotine might have either a neuroprotective or a symptomatic effect in PD. In the following secttion, we will show that the nicotinic receptors are affected in PD and describe the first clinical studies of nicotine in humans. Loss of nicotinic receptors in PD
Several convergent reports have shown that the density of nicotinic receptors is reduced in patients with PD and that this reduction mainly involves the nigrostriatal system. First, numerous post-mortem studies reported a reduction of such receptors in the striatum of deceased PD patients [2,31–36] . The amplitude of this reduction is marked, ranging between -30 and -50% of control values [2,35,36] . In addition, the substantia nigra was examined once [33] and a significant decrease of nicotinic receptors was also reported. This may result from a cell loss or a primary pathogenic disturbance in receptor number or subtype. In line with these post-mortem data, recent in vivo studies using PET or SPECT radiotracers that bind to the nicotinic receptors have shown a reduction of these latter receptors, and especially the a4b2 subtype, in the brain of PD patients. For example, two SPECT studies reported a global reduction of the density of these receptors in the cortex of patients with PD [37,38] . More precisely, a PET study confirmed the more specific reduction of a4b2 nicotinic receptors in the striatum (-10%) and the substantia nigra (-14.9%) of nondemented PD patients [5] . Although significant, this mild reduction seems to be unrelated to the severity of PD measured either with clinical scores or with 6-[18F] fluoro-L-3,4-dihydroxyphenylalanine (18F-DOPA) PET [5] . Expert Rev. Clin. Pharmacol. 4(4), (2011)
Can nicotine be used medicinally in Parkinson’s disease?
Overall, post-mortem and in vivo studies confirm the reduction of nicotinic receptors in the brain of PD patients, especially in the nigrostriatal pathway. The main consequence on motor symptoms or disease progression might depend on the different subtypes involved in such processes. Clinical effect of nicotine
With such epidemiological and fundamental data, studies in humans have logically been performed searching for a beneficial effect of nicotine in PD. Despite some benefit observed in case reports [39,40] , more systematic studies of the clinical efficacy of nicotine on motor symptoms yielded controversial results (Tables 1 & 2) . Such discrepancies may result from differences in the clinical stages of PD patients included in these trials, in smoking habits of patients, the method of nicotine administration, dosages and treatment duration (from
Special Report
acute to 29 weeks exposition). In one trial side effects limited the intake of active drug and therefore the evaluation of the efficacy of nicotine [41,42] ; however, the overall tolerance was acceptable in most studies. Open-label studies obtain mostly positive results. One of the seven open-label studies [42] found no effect at all, and, more worryingly three out of four controlled studies found no benefit [43–45] (to our knowledge, only four placebo-controlled studies of nicotine on PD symptoms have been reported) (Table 2) [43–46] . This mismatch could result from several bias, as for instance, placebo effect or believing in the efficacy of the raters. When analyzing all these studies, it is obvious that their methodologies have major differences that can explain such varied results. The included patients were different, with atypical parkinsonism in some cases [46,47] , early-onset PD in two studies [48,49] and mild-to-moderate nondemented PD in most studies. According
Table 1. Open-label clinical trials of the clinical efficacy of nicotine on motor symptoms†. Patients (n)
Product (dose)
Duration (total/max dose) ‡
Post-encephalitic parkinsonism§ (13)
Hypodermic injections 2–3 weeks (1/5–1/10 grain × 3/day)
Efficacy’s evaluation
Results (efficacy)
Tolerance (AE)
Ref.
Tone of the muscles, ‘shortening and lengthening reactions’
+ Change in pulse rate, Improvement in 9 vomiting
[47]
Tremor > 1 year Nicotine hydrogen Parkinsonian tremor (7) tartrate 3–6 mg iv. (15) (ET, rubral, cerebellar (1–2 mg nicotine) tremor associated with anxiety)
2 min
Recording of tremor
+ Reduction in 6 at 6 mg, increasing in 1 at 3 mg
‘Often’ Tingling, numbness, pallor, sick-feeling, perspiration
[51]
Smokers Early-onset IPD (6)
Smoking, gum
Acute
Tremor, rigidity, bradykinesia, gait
+ Smoking > gum; first > second
Good tolerance
[48]
IPD: clinical criteria for preliminary PD H&Y = 1–3; MMS >27, Mattis >123 (15)
Nicotine iv. (1.25 μg/kg/min x 30 min) Transdermal (14 mg/day)
15 days, 30 min (iv.)
Cognitive +++ Arms’ bradykinesia and stand– walk–sit test
+ (improvement persists 1 month later)
Good tolerance (no precision)
[59]
Early-onset parkinsonism (4/8 smokers) (8)
Gum (2 mg)
Acute
UPDRS, auditory event-related potentials
+ in smokers, 0 in nonsmokers
Nondemented IPD; H&Y = 1–3 (22 + 10 controls)
Transdermal (21 mg/day)
39/3 days
Bradykinesia, 0 UPDRS-III; (9 completed) cognitive tests
10 withdrawal due to AE
[42]
Nondemented IPD H&Y = 3 (6)
Transdermal nicotine (105 mg/day)
29/4 weeks
UPDRS total + and III, l-DOPA dose
Vomiting: 4/6; insomnia: 3; orthostatic hypotension: 6
[52]
[49]
Among these studies, doses given to patients have been highly variable from 1 to 105 mg/day and duration of treatment before evaluation vary from 1 min to 29 weeks, which may account at least partially for the differences of results. ‡ Duration (total/max dose) means: total duration of administration of nicotine/duration of administration at maximal dose. § Parkinsonism: parkinsonism not related to PD as postencephalitic parkinsonism [46] or when there is not enough precision in the article to know if patients included have IPD or other parkinsonism. +: Efficacy found; 0: No efficacy found. AE: Adverse event; ET: Essential tremor; H&Y: Hoehn and Yahr; IPD: Idiopatic Parkinson’s disease; iv.: Intravenous; UPDRS: Unified Parkinson’s Disease Rating Scale. †
www.expert-reviews.com
431
Special Report
Thiriez, Villafane, Grapin, Fenelon, Remy & Cesaro
Table 2. Controlled studies of the clinical efficacy of nicotine on motor symptoms†. Type of study
Patients (n treated/ n placebo)
Product (dose) Study’s Primary Results Tolerance (AE) duration/ efficacy criteria (efficacy) duration of treatment‡
Two independent, placebocontrolled, dose-reversal, double-blind
1 dystonic parkinsonism (little DOPA sensitivity), 1 akinetic former smoker PD(2/2)
Polacrilex gum (4 mg/patch 15/35 mg)
26/19 days Self-rating of (P1) tremor, rigidity, 27/25 weeks sleep, thinking (P2)
+
Good tolerance, transient increasing of awakenings at night in half
[46]
Placebocontrolled, double-blind
48 mild-tomoderate IPD; no on–off, no dementia (25/23)
Polacrilex gum (2 mg × 3)
4 h 30/2 × 30 min
UPDRS-III Tolerance
0
2 (T) withdrawal
