The placebo effect in Parkinson s disease

Transcription

1 302 Review The placebo effect in Parkinson s disease Raúl de la Fuente-Fernández and A. Jon Stoessl The biochemical bases of the placebo effect are still incompletely known. We show here that the placebo effect in Parkinson s disease is due, at least in part, to the release of dopamine in the striatum. We propose that the placebo effect might be related to reward mechanisms. The expectation of reward (i.e. clinical benefit) seems to be particularly relevant. According to this theory, brain dopamine release could be a common biochemical substrate for the placebo effect encountered in other medical conditions, such as pain and depression. Other neurotransmitters or neuropeptides, however, are also likely to be involved in mediating the placebo effect (e.g. opioids in pain disorders, serotonin in depression). Raúl de la Fuente- Fernández A. Jon Stoessl* Pacific Parkinson s Research Centre, University of British Columbia, Purdy Pavilion, 2221 Wesbrook Mall, Vancouver, BC, Canada V6T 2B5. * jstoessl@ interchange.ubc.ca The word placebo is Latin for I shall please [1,2]. Types of placebo include pills, injections and surgical procedures. It is assumed that a placebo is devoid of any specific effect for the medical condition being treated, but researchers have often been mesmerized by the positive responses of patients to these inert interventions. This response is known as the placebo effect [3,4]. Stewart Wolf defined this as any effect attributable to a pill, potion, or procedure, but not to its pharmacodynamic or specific properties [5]. This is only one among a great number of proposed definitions [6,7]. At the heart of them all is the concept that a placebo must lack any specific properties. Interestingly, however, the placebo effect can be very specific [8], and this specificity depends upon the information available to the recipient [9]. For example, placebos can have opposite effects on heart rate, or blood pressure, depending on whether they are given as tranquillizers or as stimulants [8]. Hence, we should distinguish carefully between the placebo as an agent and the placebo effect. Following the principle that any treatment that provides benefit to the patient ought to be incorporated into clinical practice, we should learn to maximize the placebo effect inherent in any active treatment [10]. There is evidence to suggest that the placebo effect is related to the patient s expectation. Two alternative theories have traditionally been proposed to explain the placebo effect [11]: (1) the mentalistic theory, according to which the patient s expectation is the primary basis for the placebo effect, and (2) the conditioning theory, which states that the placebo effect is essentially a conditioned response. Models have been developed to support one or other of these theories [8,12 14]. For example, expectation and conditioning could activate different biochemical pathways in placebo analgesia [15,16]. Naturally, both mechanisms could contribute to the placebo effect in any given patient. One potential problem with the conditioning theory of the placebo effect is that placebo responses can occur in subjects without previous drug experience [17]. Importantly, the theories can be reconciled if, as some recent refinements of the Pavlovian conditioning theory suggest, what is learned in Pavlovian conditioning is in fact an expectation [8,18,19]. Many medical conditions are susceptible to the placebo effect [3,4]. Although a recent meta-analysis has called its importance into question [20], new evidence supporting the expectation theory of the placebo effect suggests that the studies included in this analysis might not have been optimal to address the issue (reviewed in Ref. [21]). Pain, depression and Parkinson s disease are among the disorders in which the placebo effect can be prominent [22 24]. This raises the possibility that the effect might be mediated by changes in neurotransmitter or neuropeptide transmission [25]. It is worth noting that patients with Parkinson s disease often experience depression and pain [26]. Indeed, depression can precede the occurrence of motor dysfunction in Parkinson s disease [27], suggesting that it is a primary defect (i.e. endogenous depression) rather than a consequence of chronic motor disability (i.e. reactive depression). Kinesia paradoxica: psychologically driven dopamine release? A remarkable phenomenon sometimes encountered in advanced Parkinson s disease is that a patient who is usually severely impaired can perform very effective movements during a short period of time, under certain unusual circumstances (e.g. a fire). This phenomenon, called kinesia paradoxica (or paradoxical kinesia), has long puzzled neurologists [28]. Two theories have been put forward to explain it. According to one, the motor activity in kinesia paradoxica could be mediated by non-dopamine system(s) (i.e. neural pathways mostly activated by emergency situations or stress). Alternatively, the damaged nigrostriatal dopamine system of patients with Parkinson s disease could be activated in particular situations, either by environmental stimuli or by the psychological response to such stimuli. According to the latter theory, kinesia paradoxica would reflect a substantial, psychologically induced release of endogenous dopamine. In keeping with this notion, animal experiments have shown that, in the presence of arousing stimuli, dopaminergic neurons of /02/$ see front matter 2002 Elsevier Science Ltd. All rights reserved. PII: S (02)

2 Review 303 the substantia nigra increase their firing rate and release dopamine in the striatum [29 31]. Moreover, when compared to control subjects, patients with Parkinson s disease show verbal kinesia paradoxica an increase in word counts and discourse duration when talking about emotional experiences [32]. However, animal models of kinesia paradoxica seem to favour non-dopamine-mediated mechanisms [33,34]. For example, rats rendered akinetic by intraventricular injections of 6-hydroxydopamine and treated with a combination of dopamine D1 and D2 receptor antagonists were still able to swim and escape from an ice bath [33]. It remains unclear, however, whether these animal models are appropriate to study the kinesia paradoxica encountered in patients with Parkinson s disease [25]. The placebo effect in Parkinson s disease: results of clinical trials Although some clinical trials have shown only a modest placebo effect in Parkinson s disease (e.g. see Ref. [35]), there is ample evidence for a strong placebo effect in this disorder. Thus, in a double-blind trial of pergolide as a treatment for Parkinson s disease, Diamond and colleagues [36] found a significant improvement with respect to baseline in both the pergolide-treated group (17% improvement after four weeks and 30% after 24 weeks) and the placebo group (16% improvement after four weeks and 23% after 24 weeks). In this study, there was no significant difference between the pergolide-treated and placebo groups. Shetty and colleagues [37] have recently reviewed 98 relevant articles that were published between January 1969 and April 1996, finally selecting 36 that satisfied their inclusion criteria. Twelve of these articles reported an improvement following placebo treatment in Parkinson s disease. The magnitude of the improvement was 9 59% of that seen in the active drug group. Retrospective analysis of the Deprenyl and Tocopherol Antioxidative Therapy of Parkinsonism (DATATOP) study found a slightly lower percentage of patients experiencing improvement during placebo therapy (21%). More recently, Goetz and colleagues [38] reported that 14% of the patients enrolled in a six-month, randomized, multicentre, placebo-controlled clinical trial of ropinirole monotherapy achieved a 50% improvement in motor function while on placebo treatment. In this study, all domains of parkinsonian disability were subject to the placebo effect, but bradykinesia and rigidity tended to be more susceptible than tremor, gait or balance. The importance of including a placebo group when investigating the efficacy of surgical treatments (e.g. transplantation) for Parkinson s disease has been emphasized by several authors [24]. Although the effect of an imitation operation was modest in some studies [39], others reported placebo improvement of the same magnitude as that seen in the transplant group [40]. Several factors could explain the differences in the magnitude of the placebo response among different trials. These include differences in the information given to patients and differences in group characteristics or surgical procedures [21]. Importantly, it should be emphasized that the only two studies to have focused on the placebo effect in Parkinson s disease both found the effect to be strong [37,38]. Dopamine, expectation and reward The dopaminergic system has long been implicated in reward mechanisms [41,42]. Animal experiments have shown that the midbrain dopaminergic cell groups A8, A9 and A10 are activated by primary rewards, reward-predicting stimuli and novel stimuli [43,44]. Although three major dopamine-containing pathways (the nigrostriatal, mesolimbic and mesocortical pathways) participate in these responses, the projection to the nucleus accumbens has received the greatest attention [41,42]. Release of dopamine in the nucleus accumbens occurs during intracranial self-stimulation, a paradigm in which an animal repeatedly presses a lever to electrically stimulate its own dopaminergic projections. This behaviour has been associated with the rewarding properties of dopamine release. Most drugs of abuse (including cocaine, amphetamine, opioids, alcohol and nicotine) are characterized by their ability to increase dopamine levels in the nucleus accumbens. Indeed, the basis of drug dependence seems to be related to the rewarding experience of such dopamine release [45]. Interestingly, there is an emerging concept that several human behavioural disorders, including drug-seeking behaviour, might be related to what has been called the reward deficiency syndrome [46 48]. Individuals with this syndrome are thought to search for unnatural rewards to compensate for an intrinsic reward deficit; that is, they seek unnatural stimulation of dopamine release to compensate for hypoactive dopamine-mediated signalling in their nucleus accumbens. A crucial aspect of the reward theory that remains unclear is whether dopamine release occurs in relation to the reward itself (i.e. to the perception of the reward) or to the expectation of the reward. In an elegant experiment using the intracranial selfstimulation paradigm, Garris and colleagues [49] provided evidence that it is the expectation of the reward that triggers dopamine release in the nucleus accumbens. This is in keeping with previous experiments showing that dopaminergic neurons are activated after stimuli that predict a reward [44]. We hypothesized that, if the expectation of a reward triggers dopamine release, not only in the nucleus accumbens but also in the nigrostriatal pathway, the placebo effect in patients with Parkinson s disease could be related to the expectation of clinical benefit, and could be mediated by the release of dopamine in the striatum [50].

3 304 Review Placebo Fig. 1. [ 11 C]raclopride (RAC) PET scans of a patient with Parkinson s disease at open baseline and after placebo administration (i.e. saline injected; b). The placeboinduced decrease in striatal radioactivity (indicated by the less intense red colour) is thought to reflect an increase in the synaptic level of dopamine, which inhibits RAC from binding to dopamine D2 receptor sites. After placebo PET study: dopamine release in response to placebo treatment in Parkinson s disease Using PET with [ 11 C]raclopride (RAC), we found that patients with Parkinson s disease release substantial amounts of dopamine in the dorsal striatum (i.e. caudate and putamen) in response to subcutaneous injection of saline (Fig. 1) [50]. In this paradigm, changes in RAC binding between baseline and post-activation states (in this case, in response to placebo injection) represent a change in synaptic dopamine levels, reflecting the release of endogenous dopamine. Placebo-induced changes in striatal RAC binding potential (BP) (17% in the caudate nucleus, 19% in the putamen) (Fig. 2) were of similar magnitude to those obtained after therapeutic doses of levodopa or apomorphine [50,51]. Indeed, the magnitude of the placebo-induced change in RAC BP in Parkinson s disease patients was comparable to that obtained in healthy volunteers after amphetamine administration [52]. The effect showed a slight increasing rostrocaudal gradient, in contrast to the magnitude of dopamine loss in Parkinson s disease, which is maximal in the posterior putamen. All patients showed a biochemical placebo effect (i.e. changes in RAC BP) but only half reported clinical benefit in motor function after placebo administration. Interestingly, the estimated amount of dopamine release was greater in those patients who perceived a placebo effect than in those who did not (22% and 12% decreases in RAC BP in the caudate nucleus, respectively; 24% and 14% decreases in the putamen, respectively) (Fig. 2). Based on the known relationship between dopamine levels in the putamen and motor function, we concluded that the placebo effect in Parkinson s disease was triggered by the expectation of reward (i.e. the expectation of clinical benefit). This idea was confirmed when we analysed placebo-induced changes in the ventral striatum (nucleus accumbens). This region displayed a decline in RAC binding similar to that seen in the caudate and putamen. However, in contrast to results from the dorsal striatum, the magnitude of placebo-induced changes was not significantly different between patients who experienced clinical 3.5 Placebo Fig. 2. Individual [ 11 C]raclopride (RAC) binding potential values at open baseline (without placebo intervention) and after subcutaneous administration of placebo (saline), for patients who either did (solid squares, solid lines) or did not (open squares, broken lines) perceive a placebo-induced clinical benefit. Placebo-induced RAC binding potential changes were highly significant for both the caudate nucleus and putamen (P <05 for both regions, two-tailed paired t-test). This indicates that there is a strong placebo effect in Parkinson s disease and that this is mediated by the release of dopamine in the striatum. It is assumed that the clinical effect is related to increased dopamine levels in the putamen. benefit and those who did not (R. de la Fuente- Fernandez et al., unpublished). Because the perception of clinical benefit must be rewarding, this finding suggests that dopamine is released within the ventral striatum in relation to the expectation of reward (i.e. the expectation of clinical benefit) rather than the reward itself (i.e. the perception of clinical benefit). Controlling for confounding factors As noted earlier, dopaminergic neurons are activated not only in relation to reward, but also in response to novel stimuli [43,44]. Hence, one could argue that our results reflect the novelty of the test situation. To address this, and other, potential issues, we studied a second group of patients with Parkinson s disease (matched with the first for age and severity of parkinsonism) in an open fashion (i.e. without placebo intervention), as a control group. We found no differences in placebo-free, open baseline RAC BP between the patients described above and this second control group (Fig. 3). It could also be argued that the placebo-induced changes in RAC BP might be secondary to stress associated with the subcutaneous injection itself. If this were so, both placebo-tested

4 Review 305 Fig. 3. Mean [ 11 C]raclopride (RAC) binding potentials in the placebo-free, open baseline state for control (i.e. no previous placebo exposure; open bars, n = 6) and previously placebo-tested (solid bars, n = 6) groups. The two groups were matched for age and clinical severity of parkinsonism. There were no differences between the open baseline RAC binding potentials of the control and previously placebo-tested groups, for either the caudate nucleus (P = 9) or the putamen (P = 0.84). This suggests that our observations do not simply reflect the novelty of the test situation. Error bars, ± 1 SE. Acknowledgements Our work is supported by the Canadian Institutes of Health Research, the National Parkinson Foundation (Miami, FL, USA), the British Columbia Health Research Foundation (Canada) (R.F-F.), the Pacific Parkinson s Research Institute (Vancouver, BC, Canada) (R.F-F.), and the Canada Research Chairs program (A.J.S.). We would also like to acknowledge the contributions of Tom Ruth and the UBC-TRIUMF PET team. Control group Placebo group and placebo-free (control) groups should have a similar degree of change induced by subcutaneous injections of apomorphine. However, this was not the case, and RAC BP values were lower in the placebo group than in the open, placebo-free control group for each dose of apomorphine (Fig. 4). These observations strengthen the view that the changes in RAC BP were a true reflection of the placebo effect and not a manifestation of either novelty or stress-induced dopamine release. Conclusions and implications The power of placebos to alleviate a great number of medical conditions has long been recognized [3,4]. Parkinson s disease is among the disorders in which the placebo effect can be substantial [24,36 38,40], and PET studies have indicated that the placebo effect in Parkinson s disease is related to the release of dopamine in the striatum [50]. Other factors can also contribute to increased dopamine-mediated neurotransmission in response to placebo treatment. For example, Guttman and colleagues [53] have recently shown placebo-induced downregulation of the dopamine transporter. Interestingly, the nucleus accumbens is also susceptible to placebo-induced dopamine release in Parkinson s disease, supporting the notion that placebos can activate reward mechanisms. Dopamine, which is sometimes known as the pleasure molecule [47], could potentially be involved in the placebo effect encountered in other medical conditions, such as pain and depression [21]. For example, there is structural, electrophysiological and biochemical evidence for opioid dopamine interactions in the mesolimbic and mesocortical pathways [54]. This provides a morphological substrate to explain the opioidinduced release of dopamine in the nucleus accumbens. Indeed, enhanced activity of dopamine in the nucleus accumbens seems to play a role in analgesia [55]. Alternatively, the release of endogenous endorphins in placebo analgesia [56] could be modulated by dopamine-mediated mechanisms [57]. Control group Placebo group APO 0 APO 1 APO 2 APO 0 APO 1 APO 2 Fig. 4. and putamen mean [ 11 C]raclopride (RAC) binding potentials for control (open bars) and placebo-tested (green bars) groups, both before (APO 0) and after subcutaneous administration of particular doses of apomorphine (APO 1 = 3 mg kg 1 ; APO 2 = 6 mg kg 1 ). Note that values for APO 0 represent true baseline values for the control group, and placebo-related values for the placebo-tested group (these patients had received subcutaneous injections of saline, in a blind fashion). In response to apomorphine, patients in the placebo group had significantly lower RAC binding potential values than patients in the control group. This eliminated other potential confounding factors (e.g. the stress associated with the subcutaneous injection of saline) from our observations of a powerful biochemical placebo effect in Parkinson s disease. Error bars, ± 1 SE. Finally, placebos could also be efficacious in longterm substitution programmes for the treatment of drug addiction [5]. Most drugs of abuse (including cocaine, amphetamine, heroin, alcohol and nicotine, to name but a few) release dopamine in the nucleus accumbens, and such dopamine-mediated activation seems to be related to the dependence properties of these drugs [58,59]. We have seen that placebos can induce release of dopamine in the nucleus accumbens. Indeed, there is evidence to suggest that placebos can be addictive, causing withdrawal symptoms when treatment is discontinued [2,5]. There are reports of successful saline substitution for the active drug in morphine addicts [60], and methadone substitution programmes for heroin addicts can also benefit from the use of placebos [61]. Although there is the potential for placebo addiction, a reduction in the use of illicit drugs would represent a major achievement in terms of health and social safety.

5 306 Review References 1 Pepper, O.H.P. (1945) A note on placebo. Am. J. Pharmacol. 117, Brody, H. (1980) Placebos and the Philosophy of Medicine: Clinical, Conceptual, and Ethical Issues, University of Chicago 3 Beecher, H.K. (1955) The powerful placebo. J. Am. Med. Assoc. 159, Beecher, H.K. (1961) Surgery as placebo: a quantitative study of bias. J. Am. Med. Assoc. 176, Wolf, S. (1959) The pharmacology of placebos. Pharmacol. Rev. 11, Grunbaum, A. (1985) Explication and implications of the placebo concept. In Placebo: Theory, Research, and Mechanisms, pp. 9 36, Guilford 7 Brody, H. (1985) Placebo effect: an examination of Grunbaum s definition. In Placebo: Theory, Research, and Mechanisms, pp , Guilford 8 Kirsch, I. (1997) Specifying nonspecifics: psychological mechanisms of placebo effects. In The Placebo Effect: An Interdisciplinary Exploration, pp , Harvard University 9 Flaten, M.A. et al. (1999) Drug-related information generates placebo and nocebo responses that modify the drug response. Psychosom. Med. 61, Brody, H. (1997) The doctor as therapeutic agent: a placebo effect research agenda. In The Placebo Effect: An Interdisciplinary Exploration, pp , Harvard University 11 Byerly, H. (1976) Explaining and exploiting placebo effects. Perspect. Biol. Med. 19, Evans, F.J. (1985) Expectancy, therapeutic instructions, and the placebo response. In Placebo: Theory, Research, and Mechanisms, pp , Guilford 13 Wickramasekera, I. (1985) A conditioned response model of the placebo effect: predictions from the model. In Placebo: Theory, Research, and Mechanisms, pp , The Guilford 14 Ader, R. (1997) The role of conditioning in pharmacotherapy. In The Placebo Effect: An Interdisciplinary Exploration, pp , Harvard University 15 Amanzio, M. and Benedetti, F. (1999) Neuropharmacological dissection of placebo analgesia: expectation-activated opioid systems versus conditioning-activated specific subsystems. J. Neurosci. 19, Benedetti, F. et al. (1999) Somatotopic activation of opioid systems by target-directed expectation of analgesia. J. Neurosci. 19, Pihl, R.O. and Altman, J. (1971) An experimental analysis of the placebo effect. J. Clin. Pharmacol. New Drugs 11, Reiss, S. (1980) Pavlovian conditioning and human fear: an expectancy model. Behav. Ther. 11, Bootzin, R.R. (1985) The role of expectancy in behavior change. In Placebo: Theory, Research, and Mechanisms, pp , Guilford 20 Hrobjartsson, A. and Gotzsche, P.C. (2001) Is the placebo powerless? An analysis of clinical trials comparing placebo with no treatment. New Engl. J. Med. 344, de la Fuente-Fernández, R. et al. The placebo effect in neurological disorders. Lancet Neurology (in press) 22 Turner, J.A. et al. (1994) The importance of placebo effects in pain treatment and research. J. Am. Med. Assoc. 271, Enserink, M. (1999) Can the placebo be the cure? Science 284, Freeman, T.B. et al. (1999) Use of placebo surgery in controlled trials of a cellular-based therapy for Parkinson s disease. New Engl. J. Med. 341, Cooper, J.R. et al. (1996) The Biochemical Basis of Neuropharmacology (7th edn), Oxford University 26 Rajput, A.J. (1994) Clinical features and natural history of Parkinson s disease (special consideration of aging). In Neurodegenerative Diseases, pp , W.B. Saunders 27 Santamaria, J. et al. (1986) Parkinson s disease with depression: a possible subgroup of idiopathic parkinsonism. Neurology 36, Adams, R.D. et al. (1997) Principles of Neurology (6th edn), McGraw-Hill 29 Chiodo, L.A. et al. (1979) Reciprocal influences of activating and immobilizing stimuli on the activity of nigrostriatal dopamine neurons. Brain Res. 176, Keller, R.W., Jr et al. (1983) Environmental stimuli but not homeostatic challenges produce apparent increases in dopaminergic activity in the striatum: an analysis by in vivo voltammetry. Brain Res. 279, Horvitz, J.C. (2000) Mesolimbocortical and nigrostriatal dopamine response to salient nonreward events. Neuroscience 96, Crucian, G.P. et al. (2001) Emotional conversations in Parkinson s disease. Neurology 56, Keefe, K.A. et al. (1989) Paradoxical kinesia in parkinsonism is not caused by dopamine release: studies in an animal model. Arch. Neurol. 46, Carey, R.J. (1992) Pavlovian conditioning of L-dopa induced movement. Psychopharmacology (Berl.) 107, Parkinson study group (1997) Safety and efficacy of pramipexole in early Parkinson s disease. A randomized dose-ranging study. J. Am. Med. Assoc. 278, Diamond, S.G. et al. (1985) Double-blind trial of pergolide for Parkinson s disease. Neurology 35, Shetty, N. et al. (1999) The placebo response in Parkinson s disease. Clin. Neuropharmacol. 22, Goetz, C.G. et al. (2000) Objective changes in motor function during placebo treatment in PD. Neurology 54, Freed, C.R. et al. (2001) Transplantation of embryonic dopamine neurons for severe Parkinson s disease. New Engl. J. Med. 344, Watts, R.L. et al. (2001) A double-blind, randomized, controlled, multicenter clinical trial of the safety and efficacy of stereotaxic intrastriatal implantation of fetal porcine ventral mesencephalic tissue (Neurocell -PD) vs imitation surgery in patients with Parkinson s disease (PD). Parkinsonism Relat. Disord. 7, S87 41 Phillips, A.G. and Fibiger, H.C. (1989) Neuroanatomical bases of intracranial selfstimulation: untangling the Gordian knot. In The Neuropharmacological Basis of Reward (Liebman, J.M. and Cooper, S.J., eds), pp , Clarendon 42 Wise, R.A. and Rompre, P.P. (1989) Brain dopamine and reward. Annu. Rev. Psychol. 40, Rebec, G.V. et al. (1997) Regional and temporal differences in real-time dopamine efflux in the nucleus accumbens during free-choice novelty. Brain Res. 776, Schultz, W. (1998) Predictive reward signal of dopamine neurons. J. Neurophysiol. 80, Robinson, T.E. and Berridge, K.C. (1993) The neural basis of drug craving: an incentivesensitization theory of addiction. Brain Res. Brain Res. Rev. 18, Blum, K. et al. (1996) The D2 dopamine receptor gene as a determinant of reward deficiency syndrome. J. R. Soc. Med. 89, Blum, K. et al. (2000) Reward deficiency syndrome: a biogenetic model for diagnosis and treatment of impulsive, addictive, and compulsive behaviors. J. Psychoact. Drugs 32, Comings, D.E. and Blum, K. (2000) Reward deficiency syndrome: genetic aspects of behavioral disorders. Prog. Brain Res. 126, Garris, P.A. et al. (1999) Dissociation of dopamine release in the nucleus accumbens from intracranial self-stimulation. Nature 398, de la Fuente-Fernández, R. et al. (2001) Expectation and dopamine release: mechanism of the placebo effect in Parkinson s disease. Science 293, de la Fuente-Fernández, R. et al. (2001) Biochemical variations in the synaptic level of dopamine precede motor fluctuations in Parkinson s disease: PET evidence of increased dopamine turnover. Ann. Neurol. 49, Breier, A. et al. (1997) Schizophrenia is associated with elevated amphetamine-induced synaptic dopamine concentrations: evidence from a novel positron emission tomography method. Proc. Natl. Acad. Sci. U. S. A. 94, Guttman, M. et al. (2001) Influence of L-dopa and pramipexole on striatal dopamine transporter in early PD. Neurology 56, Sesack, S.R. and Pickel, V.M. (1992) Dual ultrastructural localization of enkephalin tyrosine hydroxylase immunoreactivity in the rat ventral tegmental area: multiple substrates for opiatedopamine interactions. J. Neurosci. 12, Altier, N. and Stewart, J. (1999) The role of dopamine in the nucleus accumbens in analgesia. Life Sci. 65, Levine, J.D. et al. (1978) The mechanism of placebo analgesia. Lancet 2, Turchan, J. et al. (1998) Effects of repeated psychostimulant administration on the prodynorphin system activity and kappa opioid receptor density in the rat brain. Neuroscience 85, Di Chiara, G. and Imperato, A. (1988) Drugs abused by humans preferentially increase synaptic dopamine concentrations in the mesolimbic system of freely moving rats. Proc. Natl. Acad. Sci. U. S. A. 85, Nestler, E.J. (1994) Molecular neurobiology of drug addiction. Neuropsychopharmacology 11, Leslie, A. (1954) Ethics and practice of placebo therapy. Am. J. Med. 16, Curran, H.V. et al. (1999) Additional methadone increases craving for heroin: a double-blind, placebo-controlled study of chronic opiate users receiving methadone substitution treatment. Addiction 94,