Brain Opioid Receptor Binding in Early Abstinence from Opioid

BRITISH JOURNAL OF PSYCHIATRY (2007), 191, 63^69. doi: 10.1192/bjp.bp.106.031120

AUTHOR’ S P ROOF

Brain opioid receptor binding in early abstinence in-patient detoxification regime from methadone and were free of opioids or from opioid dependence any other drug of misuse as confirmed by urinalysis at the time of scanning. They were typically long-standing opioid users Positron emission tomography study and had been on methadone for a mean of 6.9 years (s.d.¼5.2, range 2–17 years); TIM M. WILLIAMS, MARK R. C. DAGLISH, ANNE LINGFORD-HUGHES, the mean dose at the start of detoxification LINDSAYG.LINDSAY G. TAYLOR, ALEXANDER HAMMERS, DAVID J. BROOKS, was 31 mg/day (s.d.¼12, range 15–50). The PAUL GRASBY, JUDITH S. MYLESand DAVID J. NUTT patients had a history of using a variety of other drugs prior to admission (Table 1); all but one participant continued to use Background Although opioid There is little research detailing the changes heroin in addition to their prescribed receptor function in humans is clearly to the human brain as a result of chronic methadone in the month prior to detoxifi- opioid use and in early abstinence. We have cation, half were using ‘crack’ cocaine prior reduced during opioid dependence, what previously used [1111C]-diprenorphine positron to detoxification and all were tobacco smo- happens to the receptorin early emission tomography (PET) to measure kers. All participants underwent the stand- abstinence is not understood. levels of available opioid receptors in the ard detoxification regimen under the care brain of patients on methadone mainte- of the clinical team, independent of the Aims This study soughtto examine nance, but found no detectable occupancy investigators. changeschangesin in opioid receptor availabilityavailabilityin in by methadone (Melichar et aletal, 2005). Other Twenty healthy people (18 male, 2 fe- early abstinence from opioid dependence. PET studies have reported that increased male; mean age 34.8 years, s.d.¼8.3, range8.3,range binding of [1111C]-carfentanil in withdrawal 25–48) with no history of dependence on MethodMethod Ten people with opioid and abstinence in cocaine and alcohol de- any drug except nicotine (all current non- dependence who had completed in- pendence is associated with craving (Zubie- smokers) were recruited as controls. Con- tata et aletal, 1996; Gorelick et aletal, 2005; Heinz etet trols had no history of serious psychiatric patient detoxification and 20 healthy alal, 2005). A preliminarystudy also re- or medical disorder as determined by clini- 11 controls underwent [ C]-diprenorphine ported an increase in [1111C]-C]-carfentanilcarfentanil cal interview. They were recruited for this positron emission tomography.Clinical binding in people with opioid dependence and other studies to form a common pool, variables were assessed with structured who were briefly maintained on buprenor- to avoid unnecessary duplication and radia- questionnaires.Opioid receptor binding phine (Zubieta et aletal, 2000). In this study tion exposure. Therefore, only 8 of the 20 we present data on the binding of the controls completed all the same question- was characterised as the volume of opioid receptor PET tracer [1111C]-diprenor- naires as participants with opioid depen- 11 distribution of [ C]-diprenorphine using a phine, which labels mm,, kk andand dd opioid recep- dence. The control group for this study template of predefined brain volumes and tors, in people with opioid dependence was selected to match the age range of an exploratory voxel-by-voxel analysis. during early abstinence. We measured participants with opioid dependence. [[1111C]-diprenorphine binding in brain areas Local research ethics committees and ResultsResults Compared with controls, implicated in dependence and its relation- the UK Administration of Radioactive Sub- participants with opioid dependence had ship to clinical variables. Our hypothesis stances Advisory Committee approved all was that in people with opioid addiction, procedures and experimental protocols. increased [11C]-diprenorphine binding in opioid receptor availability would be in- After full explanation of the study proce- the whole brain and inin15 15 of the 21aprioriapriori creased in early abstinence and that this dures, volunteers gave written informed regions studied. would be related to craving. consent.consent.

Conclusions This study suggests that opioid receptor binding is increased METHOD Clinical measures throughoutthe braininbrain in early abstinence Subjective measures included the Adjective Participants from dependent opioid use.These data Checklist (Jasinski, 1997), which measures We recruited ten people with opioid depen- effects of opioid agonists (16 items) and complementcomplementthe the findingsfindingsin in cocaine and dence (8 male, 2 female; mean age 31.7 opioid withdrawal symptoms (21 items), alcohol dependence. years, s.d.years,s.d.¼6.3, range 25–45) undergoing and the 49-item short form of the Addiction in-patient detoxification at Bristol Specia- Research Centre Inventory (ARCI), which Declaration of interest None. list Drug Service (Bristol, UK). All fulfilled is scored for euphoria, dysphoria and seda- Funding detailed in Acknowledgements. DSM–IV criteria for opioid dependence tion scales (Haertzen, 1970). (American Psychiatric Association, 1994) Objective evidence of withdrawal (Opi- prior to detoxification but were excluded ate Withdrawal Scale; OWS) was measured if they fulfilled diagnostic criteria for an- using an adaptation of the Kolb and other current Axis I disorder (excluding Himmelsbach point system as previously nicotine dependence). The participants were described (Law et aletal, 1997). Drug craving in their 10th day of a lofexidine-assisted was assessed using two tools; the 45-item

6363

Downloaded from https://www.cambridge.org/core. 26 Sep 2021 at 22:39:25, subject to the Cambridge Core terms of use. WILLIAMS ET AL

AUTHOR’SAUTHOR’ S PROOFP ROOF

Ta b l e 1 Drug and alcohol use in the 30 days prior to scanning previous studies were selected for comparison. These areas were the orbito-frontal Ever usedusedEver Any use in past Days used in past 30 days cortex, anterior cingulate, ventral striatum (including the nucleus accumbens), dorsal ((nn¼10)10) 30 days (nn¼10)10) striatum (including the caudate nucleus DrugMean (s.d.)Median Range and putamen), thalamus, amygdala and periaqueductal grey matter. All these re- Heroin 10 914.7 (4.7)17 6^19 gions have been shown to have a role in Intravenous 10 712.7(6.4)7 12.7 (6.4)16 3^19 the addiction process. The anterior cingu- Smoked 10 4174 17.5.5(3.9) (3.9) 17.517.510^18 late cortex, orbitofrontal cortex, nucleus Cannabis 10 810.3(7.7)8 10.3 (7.7)10.5 1^20 accumbens and amygdala are key in reward Crack 959 5 8.2 (5.4)(5.4)8.2 63 6 3^15^15and motivation during drug-using from the Benzodiazepines 10 415.0(6.0)4 15.0 (6.0)16 8^20 evaluation of stimuli to reward-based deci- Amphetamines 909 0^^^^^^ sion-making and learning. The periaque- MDMAMDMA 808 0^^^^^^ ductal grey matter is an important element LSDLSD 10 0^0 ^^^^^ of the endogenous opioid system and is in- AlcoholAlcohol 10 5 8.2 (6.3)52 5 2^16^16volved in conditioned processes in depen- NicotineNicotine 10 10 30 3030All 30 dent drug use. Similarly the thalamus, caudate and putamen form part of the emo- MDMA, methylenedioxymethamphetamine; LSD, lysergic acid diethylamide. tional reward neurocircuitry which has an important role in motivational factors and Heroin Craving Questionnaire (HCQ; continuously online in accordance with a links to motor pathways, possibly being a WeinsteinWeinstein et aletal 1997) and the Obsessive standard protocol and discrete blood route for the development of locomotor Compulsive Drinking Scale (OCDS), samples were taken every 5–10 min for sensitisation with continued drug use adapted to measure opioid compulsive be- assay of radiolabelled metabolites in (Kalivas & Volkow, 2005; Nutt et al, 2006).,2006). haviour and obsessive thoughts, and allow- plasma (Ranicar et aletal, 1991).,1991). We used statistical parametric mapping ing for mode of drug delivery (Anton et aletal to transfer 73 standardised volumes of 1996). Two experienced addiction clini- Image processing and statistical interest derived from a probabilistic atlas cians (T.W. and M.D.) independently rated analysisanalysis of brain images (Hammers et aletal, 2003) onto each patient for the amount of opioids used The dynamic PET scans were analysed to individual scans by inverting the defor- in the month and year prior to scanning, as produce parametric images of ligand mations used to spatially normalise the well as lifetime use, using a structured volume of distribution using spectral analy- images. The volume of distribution maps rating scheme taking into account the com- sis with in-house receptor parametric map- were sampled using the individualised atlas bination of opioids used, length and route ping software implemented in Matlab for every participant to generate mean of use. We also looked for associations with (Mathworks Inc., Natick, Massachusetts) volume of distribution values for each published post-mortem data reporting (Gunn(Gunn et aletal, 2002). Spectral analysis with volume of interest. These values were then regional densities of mm opioid receptors individual metabolite corrected plasma in- compared between groups using a two- (Pfeiffer(Pfeiffer et aletal, 1982).,1982). put function takes account of any difference tailed non-paired tt-test – unequal variances Participants also completed the Spiel- in tracer delivery between individuals or were assumed. Pearson’s correlation statis- berger State–Trait Anxiety Inventory (SSAI, groups. The volume of distribution is the tics were used to assess the association of STAI; Spielberger 1983), the 36-item short ratio of total free and bound tissue to free clinical variables with opioid receptor form of the General Health Survey (SF-36; plasma ligand concentration at equilibrium binding.binding. McHorneyMcHorney et aletal, 1994), the revised and provides an index of receptor binding. In addition, [1111C]-diprenorphine volume Eysenck Personality Questionnaire (EPQ–R; We used Statistical Parametric Mapping of distribution images were analysed on a Eysenck & Eysenck, 1975) and the Eysenck (SPM2, Wellcome Department of Imaging voxel-by-voxel basis using SPM2. Spatially Impulsiveness Questionnaire (IVE; Eysenck Neuroscience, Institute of Neurology, Uni- normalised parametric images were et aletal, 1985).,1985). versity College London, UK) running in smoothed with a 12 mm kernel at full width Matlab to transform the volume of distri- half maximum. Mean differences between PET scansscansPET bution images into a standard space as groups were interrogated using non-pairednon-paired All participants underwent [1111C]-diprenor- defined by the MontreMontreal´al Neurological tt-tests, and correlations between clinicalclinical phine PET using a CTI/Siemens (Munich, Institute (Evans et aletal, 1993) using a variables and [1111C]-diprenorphine binding Germany) ECAT 953b brain camera in weighted-mean [11C]-diprenorphine template were explored using linear regression with- high-sensitivity three-dimensional mode. A created inhouse from the PET scans of in the general linear model in SPM2. bolus of 370 MBq [1111C]-diprenorphine seven healthy volunteers. Proportional scaling was used to normalise was given intravenously over 30 s. Dynamic We performed two types of analysis, global differences throughout. For regions emission data were acquired over 90 min, first using predefined volumes of interest where there existed an a prioriapriori hypothesis,hypothesis, in 18 time frames (27 frames for 9 controls) and then using SPM2 for an exploratory results are reported as significant at a and reconstructed into 31 contiguous hori- analysis. Twenty-one regions for which we threshold of PP550.05 uncorrected. For all zontal image planes (Jones et aletal, 1994).,1994). had anhadan a prioriapriori hypothesis for increased other areas familywise error correction Radioactivity in arterial blood was assayed opioid receptor availability based on was used.wasused.

Downloaded from https://www.cambridge.org/core. 26 Sep 2021 at 22:39:25, subject to the Cambridge Core terms of use. BRBRAIN AIN OPIOID RECEPTOR BINDING IN EARLY ABSTINENCE

AUTHOR’ S P ROOF

RESULTSRESULTS Ta b l e 22Tab Clinical measures Clinical measures Test and measure Opioid dependent (nn¼10)Controls (nn¼8)8) PP At the time of scanning, there was no clinically significant opioid withdrawal as mean (s.d.) measured by the OWS (mean score 1.2, s.d.s.d.¼1.2, range 0–3). However the subjec- Withdrawal tive ARCI and Adjective Checklist showed Observer Withdrawal Scale 1.2 (1.2)(1.2)1.2 0.1 (0.4)0.024 higher measures of withdrawal in partici- Adjective Checklist^ withdrawal 20.8 (9.9)(9.9)20.8 5.0 (5.3) 550.000.0011 pants with opioid dependence than in con- Adjective Checklist ^ agonist effects22.5 (9.4) 20.5 (5.1)(5.1)20.5 0.595 trols. They showed significantly higher ARCI state, but not trait, anxiety than controls Withdrawal effects 23.0 (9.9) 5.0 (5.4) 550.000.0011 on the day of their scan. Participants with Euphoric effects 7.8 (3.4)(3.4)7.8 4.3 (2.8)0.030 opioid dependence scored significantly low- Agonist effects 22.5 (9.3)(9.3)22.5 20.5 (5.1)(5.1)20.5 0.595 er than controls on some measures of health (Table 2). On the personality question- Spielberger naires participants with opioid dependence SSAI ^ statestateSSAI 43.7 (8.6)(8.6)43.7 32.9 (9.1)(9.1)32.9 11 0.00.01414 scored significantly higher for psychoti- STAI ^ trait 44.5 (11.5)(11.5)44.5 37.5 (10.6)(10.6)37.5 11 0.172 cism, extraversion, impulsiveness and ven- SF^36SF^36 turesomeness than controls, but not for General health 43.6 (19.5)(19.5)43.6 82.5 (14.2)(14.2)82.5 550.000.0011 neuroticism or empathy. Physical function 80.2 (10.6)(10.6)80.2 96.9 (5.3)(5.3)96.9 0.000.0011 All participants with opioid dependence Physical role limitation 45.0 (36.9)(36.9)45.0 96.9 (8.8)(8.8)96.9 0.000.0011 reported craving on the HCQ (mean score Bodily painpainBodily 39.7 (17.9) 83.4 (19.7) 550.000.0011 15.7, s.d.15.7,s.d.¼6.0) and modified OCDS (mean Emotional role limitation 43.3 (41.7) 87.5 (35.4)(35.4)87.5 0.030 score 21.67, s.d.¼10.6), and the scores Social functioning 52.5 (23.4)(23.4)52.5 87.6 (17.7)0.003 were highly correlated (rr¼0.76,0.76, Vitality 39.0 (16.5) 49.6 (16.5)0.193 PP550.018). Craving scores were comparable with our previous study of the same Mental health 43.8 (11.8) 45.8 (9.9)(9.9)45.8 0.699 stage of detoxification where craving was EPQ^REPQ^R elicited using an imagery-based procedure Psychoticism 11.0 (2.3) 5.2 (2.9)(2.9)5.2 22 550.000.0011 (Weinstein(Weinstein et aletal, 1997).,1997). Extraversion/intraversion 18.9 (4.4) 13.7 (3.0)22 0.008 Neuroticism15.3 (5.1)11.9 (8.2)22 0.2850.285 Image analysis IVE Participants with opioid dependence Impulsiveness13.5 (4.5)4.9 (2.9) 550.000.0011 showed a significantly higher level of Venturesomeness 12.3 (2.8) 8.0 (3.5) 0.0100.010 opioid receptor availability, as measured Empathy 13.0 (4.0) 14.3 (3.1) 0.4770.477 by global [1111C]-diprenorphine volume of distribution, when compared with controls 1. nn¼10. 2.2. nn¼9. (19.3(19.3 v.v. 17.1, 11.4% increase, 95% CI ARC1, Addiction Research Centre Inventory; SF^36, 36-item short form of the General Health Survey; EPQEPQ^R, ^R, 2.1–20.7,2.1–20.7, PP¼0.019). In 15 of the 21 a prioriapriori Eysenck Personality Questionnaire ^ Revised; IVE, Eysenck Impulsiveness Questionnaire. regions studied there was a significant increase in volume of distribution in people with opioid dependence when compared with controls (PP550.05 uncorrected). These were the brain-stem, right amygdala, left medial orbital cortex and bilateral anterior cingulate, putamen, thalamus, and anterior, lateral and posterior orbitofrontal cortex. Only the left lateral orbital area remained significant if these areas are considered independent and the overly conservative Bonferroni correction is applied (PP¼0.042,0.042, corrected). There was no significant association between age and global [1111C]-dipre-C]-diprenorphine volume of distribution for the whole group (nn¼30,30, rr¼770.30,0.30, PP¼0.105)0.105) or when the control (nn¼20,20, rr¼770.31,0.31, PP¼0.184) and opiod-dependent groups ((nn¼10,10, rr¼770.02,0.02, PP¼0.962) were analysed Fig. 11Fig. Global andandGlobal aprioriregional [11C]-diprenorphine binding for people with opioid dependence (&&)and separately.separately. controls ((controls &&). *).*PP550.05 unpaired tt-test.-test.

6565

Downloaded from https://www.cambridge.org/core. 26 Sep 2021 at 22:39:25, subject to the Cambridge Core terms of use. WILLIAMS ET AL

AUTHOR’ S P ROOF

The re-detoxification daily methadone using [using[1111C]-diprenorphine PET in people controls (Gabilondo et aletal, 1994; Ferrer-Alcon total dose (mg/kg) and duration of metha- with opioid dependence undergoing recent et aletal, 2004). These studies suggest that done use showed no association with either detoxification. We found a significant in- chronic opioid exposure might not alter global or regional [1111C]-diprenorphine crease in [1111C]-diprenorphine volume of opioid receptor availability and import- binding. Furthermore, no relationship with distribution in the majority of a prioriapriori antly not increase receptor availability. [[1111C]-diprenorphine binding was found regions of interest, although notably not when the participants with opioid depen- in the nucleus accumbens or caudate nu- 11 dence were divided into two groups: cleus. Exploratory statistical parametric Increased [ C]-diprenorphine short-term users (use for under 10 years) mapping at a threshold for significance ap- bindingbinding and long-term users (use for more than 10 plying a full correction for multiple com- The increase in [1111C]-diprenorphine binding years). We found no significant effect for parisons identified the right fusiform and reflects an increase in availability of opioid either alcohol or ‘crack’ cocaine on [1111C]-C]- right parahippocampal gyri as areas of receptors to this PET tracer. Increased re- diprenorphine binding by comparing those significantly increased binding. At a lower ceptor affinity for the tracer could account that had used in the previous 30 days with threshold, the mapping confirmed the find- for this increased availability, but there is those that had not. Of interest, although it ings of the volume of interest analysis when no preclinical evidence that chronic opioid did not achieve significance, was a trend to- applied to regions specified a prioriapriori.. administration alters affinity. Therefore, wards a negative correlation between level of the increase in [1111C]-diprenorphine binding opioid use in the previous month, year or might be due to: (a) an increase in opioid lifetime and global [11C]-diprenorphine bind- Opioid receptor availability receptors during early abstinence from ing. There was no correlation between [11C]-C]- in dependence opioid drugs; (b) an increase in opioid diprenorphine volume of distribution and There is limited preclinical research that receptor number that develops with the craving, subjective opioid effects or with- helps to interpret the results of this study. chronic use of an opioid agonist; (c) a re- drawal, or any of the personality variables. Although it is clear that chronic opioid duction in competition from endogenous Regional densities of mm opioid recep- exposure leads to reduced opioid receptor opioids. We believe that it is most likely tors, as reported in post-mortem tissue function (tolerance), the mechanisms that our findings reflect a significant in- (Pfeiffer(Pfeiffer et aletal, 1982), correlated with through which this is achieved are not crease in opioid receptor number immedi- [[1111C]-diprenorphine binding in each region certain and may include receptor internalisa- ately following detoxification from for the whole group (rr¼0.54,0.54, PP¼0.006),0.006), tion or reduced receptor–effector coupling opioids. We know that withdrawal and and when control and patient groups were (Williams(Williams et aletal, 2001).,2001). In vitroInvitro studies have early abstinence is a time when the brain analysed separately (rr¼0.51,0.51, PP¼0.010 and0.010and shown that chronic exposure to an opioid is under stress, and that such an increase rr¼0.53,0.53, PP¼0.007 respectively). There was agonist can lead to a downregulation in might represent a neuroadaptive response. no correlation between [1111C]-diprenorphine opioid receptors (Goodman et aletal, 1996).,1996). This would explain the similar findings volume of distribution and kk opioid recep- However, this is not a consistent pattern after cocaine and alcohol dependence tor (combined group rr¼0.05,0.05, PP¼0.803) or0.803)or inin in vivoinvivo studies, which have reported in- (Zubieta(Zubieta et al ,,etal 1996; Gorelick et aletal, 2005;,2005; dd opioid receptor regional densities (com- creases, decreases and no change in opioid HeinzHeinz et aletal, 2005).,2005). bined group rr¼0.31,0.31, PP¼0.141). No differ- receptors, depending on the paradigm used Increased [1111C]-diprenorphine binding ences were apparent between the groups (Zadina(Zadina et aletal, 1995). In humans, tolerance could also reflect increased opioid receptor in these correlations. to opioid agonists is well characterised but availability as a result of suppression of An exploratory comparison of people there are virtually no data on brain opioid endogenous opioid release. Preclinical evi- with opioid dependence and controls using receptor imaging. We have previously dence shows that chronic treatment with statistical parametric mapping showed a demonstrated a dose-related reduction in methadone does not alter the concentration significant increase in [1111C]-diprenorphine opioid receptor function in people with or function of endogenous opioids, binding only in the right fusiform and para- opioid dependence who are on methadone although later studies with other opioids hippocampal gyri (MNI coordinates maintenance by showing that they are less and other drugs of misuse suggest that xx¼38 mm,mm,38 yy¼7726 mm, zz¼7732 mm, clus- sensitive to the effects of an opioid agonist, endogenous opioids play a role in craving tersize 594 voxels, peak T¼5.15,5.15, hydromorphone (Melichar et aletal, 2003).,2003). or drug-seeking behaviour (for a review PP¼0.016 familywise error corrected). No However, in a parallel [1111C]-diprenorphine see GerritsseeGerrits et aletal, 2003). Activation of the significant correlations were found using PET study, we found no difference in bind- endogenous opioid system is associated statistical parametric mapping between ing compared with a healthy control group, with the regulation of emotions, physical [[1111C]-diprenorphine volume of distribution suggesting limited occupancy and no signif- and emotional pain (Ribeiro et aletal, 2005).,2005). and any clinical variables. Using a statisti- icant changes in receptor number (Melichar A possibility is that the exogenous opioids cal threshold of PP440.05 uncorrected to in- et aletal, 2005). This complements a study used to alter emotions by people with vestigate the a prioriapriori areas confirmed the using [using[1818F]-cyclofoxy PET that also sug- opioid dependence might lead to suppres- findings of the region of interest analysis. gested that methadone requires only very sion of the endogenous opioid system and low levels of opioid receptor occupancy consequently a compensatory upregulation for efficacy (Kling et aletal, 2000). Lastly, of receptors. This would leave more DISCUSSION post-mortem studies of people with heroin receptors available for occupancy by dependence have shown inconsistent changes [11C]-diprenorphine in early abstinence. This is the first study, to our knowledge, (reduction or no difference) in mm opioidopioid We are not aware of any human studies that has studied opioid receptor binding receptor density compared with healthy describing the impact of chronic opioid

6666

Downloaded from https://www.cambridge.org/core. 26 Sep 2021 at 22:39:25, subject to the Cambridge Core terms of use. BRBRAIN AIN OPIOID RECEPTOR BINDING IN EARLY ABSTINENCE

AUTHOR’ S P ROOF

agonist use on levels of endogenous and memory was highlighted by statistical In other studies reporting a relationship opioids.opioids. parametric mapping. We found no differ- between craving and opioid receptor levels, ence in [[encein 1111C]-diprenorphine binding be- [[1111C]-carfentanil, a mm-selective tracer was tween people with opioid dependence and used (Zubieta et aletal, 1996, 2000; Gorelick Opioid system after abstinence controls in several of the a prioriapriori areas, no-areas,no- et aletal, 2005; Heinz et aletal, 2005). It may be from substances tably the periaqueductal grey matter (in the that since [1111C]-diprenorphine labels mm,, kk In addition to being the primary target for brain-stem), nucleus accumbens and cau- andand dd opioid receptors, mm receptor-related opioid drugs, the opioid neurotransmitter date. The template used for the brain-stem changes were obscured by alterations in system is important in initiating and main- region is not precise enough to isolate the the other subtypes. However we think this taining dependence on a variety of misused periaqueductal grey matter within the unlikely as the [1111C]-diprenorphine signal substances (Herz, 1997; Gerrits et aletal, 2003;,2003; brain-stem region of interest, which may correlated only with the reported mm opioidopioid KreekKreek et aletal, 2004). A number of recent account for the lack of association there. receptor density in each brain region and neuroimaging studies in humans using the However, we are surprised to find no asso- not with the kk andand dd opioid receptor den- mm-selective agonist [1111C]-carfentanil have ciation with the nucleus accumbens and sity. Nevertheless, it would be beneficial reported elevations of tracer binding in caudate in the light of previous findings of to repeat this study using a more selective early abstinence from cocaine and alcohol, increased receptor number in these areas opioid receptor tracer, such as [1111C]-carfen-C]-carfen- which are associated with craving during withdrawal from cocaine and alco- tanil, to determine whether the increase in (Zubieta(Zubieta et aletal, 1996; Gorelick et aletal, 2005;,2005; hol. In the two studies of cocaine depen- opioid receptor binding demonstrated here HeinzHeinz et aletal, 2005). Detoxification from a dence, significant increases were seen in is mainly due to increase in any particular short course of buprenorphine has been the ventral striatum and the anterior cingu- subtype.subtype. shown in a preliminary study to result late, frontal and temporal cortices, caudate Opioid receptor binding levels were not in a significant increase in mm opioidopioid and thalamus (Zubieta et aletal, 1996;,1996; related to withdrawal symptoms as found [[1111C]-carfentanil binding in the inferofron- GorelickGorelick et aletal, 2005), whereas in alcohol in cocaine and alcohol dependence (Zubieta tal cortex and anterior cingulate regions dependence, significant increases were re- et aletal, 1996; Gorelick et aletal, 2005). This is compared with controls (Zubieta et aletal,, stricted to the ventral striatum (Heinz et al,, consistent with the clinical picture where 2000). Therefore, it appears that similar in- 2005). In people with opioid dependence opioid withdrawal can be ameliorated by creases in opioid receptor availability are the changes were much more widespread, non-opioid pharmacotherapy. We did not seen during early abstinence from cocaine perhaps because of the direct pharmaco- find a correlation between age and opioid and alcohol, and preliminary data suggest logical effect of opioids and possible receptor levels in either the group with a comparable increase in people with changes in the endogenous opioid system. opioid dependence or controls. In a [1111C]-C]- opioid dependence. carfentanil PET study of healthy controls The evidence to date points to eleva- with a wider age range, increasing age tions in opioid binding being an acute effect Opioid receptor availability was associated with higher opioid receptor of early abstinence, and our results in and clinical variables levels in the neocortex (Zubieta et aletal,, opioid dependence complement these find- We found no correlation between craving 1999). Our more limited age range and ings. It is not clear whether these changes and opioid receptor availability, which is younger average age likely contributed to persist or even become additive with pro- at variance with our hypothesis and pre- the absence of such a correlation. All of gressive detoxifications. In cocaine depen- vious findings in alcohol and cocaine de- our group with dependence were tobacco dence, opioid receptor binding in some pendence (Zubieta et aletal, 1996; Gorelick etet smokers and controls were current non- but not all regions returns to control levels alal, 2005; Heinz et aletal, 2005). Our parti- smokers, but there was no correlation be- within 1 week (Gorelick et aletal, 2005). In cipants with opioid dependence demon- tween quantity of cigarettes smoked and alcohol dependence the increase appears strated high scores on two rating scales [[1111C]-diprenorphine binding. Furthermore, more persistent, with no reduction evident for craving, which were comparable with another study of alcoholism reported no after 5 weeks of abstinence (Heinz et aletal,, those in a previous study (Weinstein et aletal,, significant interaction between smoking 2005). It was not possible to scan our 1997) and with individuals maintained on status andstatusand mm opioid receptor availability in participants after a period of abstinence methadone who had withdrawal induced patients and controls (Heinz et aletal, 2005).,2005). owing to high relapse rates and strict resi- by naloxone (Schuster et aletal, 1995) but were dential rehabilitation programmes, but this higher than scores for people maintained on would be valuable in future studies. methadone (Schuster et aletal, 1995). Further- Limitations We found significant increases in [1111C]-C]- more, our participants experienced levels Although this study was appropriately diprenorphine binding in the majority of of and variance in craving scores that were powered to detect the measured effect in a regions analysed using the atlas, and signif- comparable with earlier studies in which PET study of this nature, it may have been icant increases in fusiform/parahippocampal craving was induced and resulting brain underpowered to determine associations gyri using exploratory voxel-based statisti- activation detected (Daglish et aletal, 2001).,2001). with clinical variables, especially craving. cal parametric mapping, although increases Craving measures vary and so comparison The studies reporting an association be- were seen throughout the brain when the with other studies is hampered. However, tween craving and opioid receptor levels threshold for significance was lowered. It in our study we chose two commonly used had dependent groups of 10, 17 and 25 re- is not clear why an area incorporating the scales with a total of seven craving sub- spectively (Zubieta et aletal, 1996; Gorelick etet fusiform/parahippocampal gyri which is scales, so the absence of a correlation here alal, 2005; Heinz et aletal, 2005). However, the involved in processing visual associations is robust.isrobust. participants in our study were craving at

6767

Downloaded from https://www.cambridge.org/core. 26 Sep 2021 at 22:39:25, subject to the Cambridge Core terms of use. WILLIAMS ET AL

AUTHOR’ S P ROOF

similar levels and with a wide range of TIM M.WILLIAMS,M.WILLIAMS,MBChB,MRCPsych,MARKR.C.DAGLISH MBChB, MRCPsych, MARK R. C. DAGLISH,,MBChB,MRCPsych, MBChB, MRCPsych, craving scores, making it likely that any ANNE LINGFORD-HUGHES,LINGFORD-HUGHES, PhD,PhD,BM BM BCh, MRCPsych, Psychopharmacology Unit,Unit,University University of Bristol association should have been apparent. and MRC Clinical Sciences Centre and Division of Neuroscience, Faculty of Medicine, Imperial College School of Medicine, London,LINDSAYG. TAYLOR,TAYLOR, MSc, DRI, DAppSc, Psychopharmacology Unit,University of Bristol, Implications ALEXANDER HAMMERS,HAMMERS, PhD,PhD,MD,MRC MD, MRC Clinical Sciences Centre and Division of Neuroscience, Faculty of We have reported a significant widespread Medicine, Imperial College School of Medicine, London,DAVID J.BROOKS,J. BROOKS,MD,DSc,FRCP,FMedSci,MRC MD,DSc, FRCP,FMedSci, MRC increase in brain opioid receptor availabil- Clinical Sciences Centre and Division of Neuroscience, Faculty of Medicine, Imperial College School of Medicine, London,PAULGRASBY,MBBS,DCM,MRCPsych,MD,FRCP,FMedSci,MRCClinicalSciencesCentre PAULGRASBY,MBBS,DCM,MRCPsych,MD,FRCP,FMedSci,MRCClinicalSciencesCentre ity in people with opioid dependence dur- and Division of Neuroscience, Faculty of Medicine, Imperial College School of Medicine, London, ing early abstinence from methadone. JUDITH S. MYLES,MYLES, MB BCh, BAO,BAO,FRCPsych, FRCPsych, Bristol Specialist Drug Service, Blackberry Hill Hospital, Bristol, Together with previous evidence, we argue DAVID J. NUTT,DM,NUT T, DM, FRCP,FRCPsych, FMedSci, Psychopharmacology Unit,University of Bristol,Bristol,UK UK that this reflects an increase after cessation of methadone rather than a chronic change. Correspondence:Correspondence: Professor David Nutt,PsychopharmaNutt,Psychopharmacologycology Unit,University of Bristol,Bristol,Dorothy Dorothy If this is the case, it could give us a crucial Hodgkin Building,Whitson Street,Bristol BS13NY.Email: david.j.nutt@@bristol.ac.uk insight into the mechanisms that underlie (First received 18 September 2006, final revision 1February 2007, accepted 19 February 2007) opioid craving. Although clinically we know that substitution treatment is effec- tive we do not know whether prolonged agonist exposure permanently alters brain Ferrer-Alcon, M., La Harpe, R. & Garcia-Sevilla, Jones, A. K., Cunningham,V.J., Ha-Kawa, S. K., et aletal neurochemistry and whether these changes J. A. (2004) Decreased immunodensities of micro- (1994)(19 94) Quantitation of [11C] diprenorphine cerebral hamper recovery. Furthermore, since such opioid receptors, receptor kinases GRK 2/6 and beta- kinetics in man acquired by PETusing presaturation, an increase in opioid receptors has also arrestin-2 in postmortem brains of opiate addicts. BrainBrain pulse-chase and tracer-only protocols. Journal ofofJournal been shown in alcohol and cocaine depen- Research, Molecular Brain Research,, 121,114^122.,114^122. Neuroscience Methods,, 5151,123^134. dence, this argues for a fundamental role Gabilondo, A. M., Meana, J. J., Barturen, F., et aletal Kalivas, P.W. & Volkow, N. D. (2005) The neural basis of addiction: a pathology of motivation and choice. of the opioid system in addiction, or at least (19 94) mu-Opioid receptor and alpha 2-adrenoceptor agonist binding sites in the postmortem brain of heroin American Journal of Psychiatry,, 162162,,1406-1413. 14 0 6 -1413. in the early abstinence syndrome. The con- addicts.addicts. Psychopharmacology (Berlin),, 115115, 135^140.,135^140. Kling, M. A., Carson, R. E., Borg, L., et aletal (2000)(2000) tribution of this to clinical states and Gerrits, M. A., Lesscher,Lesscher,H. H. B. & van Ree, J. M. Opioid receptor imaging with positron emission treatment outcomes has yet to be fully (2003) Drug dependence and the endogenous opioid tomography and [(18)F]cyclofoxy in long-term, characterised. system. European Neuropsychopharmacology,, 1313, 424^,424^ methadone-treated former heroin addicts. Journal of 434. Pharmacology and Experimental Therapeutics,, 295,, 1070^1076.1070^1076. Goodman, C. B., Emilien, B., Becketts, K., et aletal Kreek, M. J., Schlussman, S. D., Bart, G., et aletal (2004)(2004) ACKNOWLEDGEMENTS (19 9 6) Downregulation of mu-opioid binding sites Evolving perspectives on neurobiological research on following chronic administration of neuropeptide FF the addictions: celebration of the 30th anniversary of This work was funded by a programme grant from (NPFF) and morphine. Peptides,, 1717, 389^397.,389^397. NIDA. Neuropharmacology,, 4747 (suppl. 1), 324^344. the Medical Research Council, UK, and by Avon Gorelick, D. A., Kim,Y.K., Bencherif, B., et aletal (2005)(2005) Law, F. D., Bailey, J. E., Allen, D. S., et aletal (19 9 7) TheThe and Wiltshire Partnership NHS Trust research and Imaging brain mu-opioid receptors in abstinent cocaine development funding. Radiotracers and scanners feasibility of abrupt methadone-buprenorphine transfer users: time course and relation to cocaine craving. in British opiate addicts in an outpatient setting. were supplied by Hammersmith Imanet,GE Health- Biological Psychiatry,, 5757,,1573^1582. 1573^1582. Addiction Biology,, 22,191^200.,191^200. care. Gunn, R. N., Gunn, S. R.,R.,Turkheimer, Turkheimer, F. E., et aletal McHorney,C. A.,Ware, J. E. Jr., Lu. J. F., et aletal (19 94) (2002)(2002) Positron emission tomography compartmental The MOS 36-item short-form health survey (SF-36): III. models: a basis pursuit strategy for kinetic modelling. Tests of data quality, scaling assumptions, and reliability REFERENCES Journal of Cerebral Blood Flow Metabolism,, 22,1425^ across diverse patient groups. Medical Care,, 3232, 40^66.,40^66. 1439.1439. American Psychiatric Association (1994) Diagnostic Melichar, J. K., Myles, J. S., Eap, C.B., et aletal (2003)(2003) and Statistical Manual of Mental Disorders (4th edn) Haertzen, C. A. (1970) Subjective effects of narcotic Using saccadic eye movements as objective measures of (DSM^IV). APA. antagonists cyclazocine and nalorphine on the Addiction tolerance in methadone dependent individuals during Research Centre Inventory (ARCI). Psychopharmacologia,, the hydromorphone challenge test. Addiction Biology,, 88,, Anton, R. F., Moak, D. H. & Latham, P.K. (1996) TheThe 1818, 366^377.,366^377. 59^66. obsessive compulsive drinking scale: a new method of Hammers, A., Allom, R., Koepp, M. J., et aletal (2003)(2003) Melichar, J. K., Hume, S. P.,P.,Williams, Williams, T. M., et al assessing outcome in alcoholism treatment studies. 1111 Three-dimensional maximum probability atlas of the (2005) Using [[Using C]diprenorphine to image opioid Archives of General Psychiatry,, 53, 225^231.,225^231. human brain, with particular reference to the temporal receptor occupancy by methadone in opioid addiction: lobe. Human Brain Mapping,, 19, 224^247.,224^247. clinical and preclinical studies. Journal of Pharmacology Daglish,M.R.,Weinstein,A.,Malizia,A.L.,Daglish, M. R.,Weinstein, A., Malizia, A. L., et aletal and Experimental Therapeutics,, 312312, 309^315. (2001)(2001) Changes in regional cerebral blood flow elicited Heinz,A.,Riemold,M.,Wrase,J.,Heinz, A., Riemold, M., Wrase, J., et aletal (2005)(2005) by craving memories in abstinent opiate-dependent Correlation of stable elevations in striatal mu-opioid Nutt, D. J., Robbins,T.W., Stimson, G.V., et aletal (2006) subjects.subjects. American Journal of Psychiatry,, 158,1680^1686. receptor availability in detoxified alcoholic patients with Drugs and the Future: Brain Science, Addiction and Society.. alcohol craving: a positron emission tomography study Academic Press. Evans, A.C.,Collins,A. C., Collins, D. L., Mills, S. R., et aletal (19 93) 3D using carbon 11-labeled carfentanil. Archives of General statistical neuroanatomical models from 305 MRI Pfeiffer, A., Pasi, A., Mehraein, P., et aletal (19 82) Opiate PsychiatryPsychiatry,, 62, 57^64.,57^64. volumes. Nuclear Science Symposium and Medical Imaging receptor binding sites in human brain. Brain Research,, 248248, 87^96. Conference 1993: IEEE Conference Record,1813^1817., 1813^1817. Herz, A. (1997) Endogenous opioid systems and alcohol addiction. Psychopharmacology (Berlin),, 129,, Ranicar, A. S.,Williams, C.W., Schnorr, L., et aletal Eysenck, H. J. & Eysenck, S. B. G. (1975) Manual of the 999^111. 9 ^ 111. (19 91) The on-line monitoring of continuously Eysenck Personality Questionnaire..Hodder& Hodder & Stoughton. withdrawn arterial blood during PET studies using a Jasinski, D. R. (1997) Assessment of the abuse potential single BGO/photomultiplier assembly and non-stick Eysenck, S. B. G., Pearson, R. R., Easting, G., et aletal of morphine-like drugs (methods used in man). In Drug tubing. Medical ProgressThroughTechnologyProgress ThroughTechnology,, 1717,259^264., 259^264. (19 8 5)5)(19 Age norms for impulsiveness, venturesomeness, Addiction I: Morphine, Sedative/Hypnotic and Alcohol and empathy in adults. Personality and Individual Dependence. (Handbook of Experimental Pharmacology,, Ribeiro, S. C., Kennedy, S. E., Smith,Y. R., et aletal DifferencesDifferences,, 66,,613^619. 613^619. vol. 45), pp.197^258. Springer. (2005) InterfaceInterfaceofphysicalandemotionalstress of physical and emotional stress

6868

Downloaded from https://www.cambridge.org/core. 26 Sep 2021 at 22:39:25, subject to the Cambridge Core terms of use. BRBRAIN AIN OPIOID RECEPTOR BINDING IN EARLY ABSTINENCE

AUTHOR’ S P ROOF

regulation through the endogenous opioid system and Weinstein, A.,Wilson, S., Bailey, J., et aletal (19 9 7) in cocaine-dependent men is associated with cocaine mm-opioid receptors. Progress in Neuropsychopharmacology Imagery of craving in opiate addicts undergoing craving. Nature Medicine,, 22,1225^1229. and Biological Psychiatry,, 29,1264^1280. detoxification. Drug and Alcohol Dependence,, 48,,25^31. 25^31. Williams, J. T., Christie, M. J. & Manzoni, O. (2001) Zubieta, J., Dannals, R. F., Frost, J. J. (1999) Gender Schuster,C. R., Greenwald, M. K., Johanson, C. E., Cellular and synaptic adaptations mediating opioid and age influences on human brain mu-opioid receptor et aletal (19 995) 5) Measurement of drug craving during dependence. Physiology Review,, 81, 299^343.,299^343. binding measured by PET. American Journal of Psychiatry,, naloxone-precipitated withdrawal in methadone- 156,842^848., 842^848. maintained volunteers. Experimental and Clinical Zadina, J. E., Kastin, A. J., Harrison, L. M., et aletal (19 95) Opiate receptor changes after chronic exposure Psychopharmacology,, 33,424^431., 424^431. Zubieta, J., Greenwald, M. K., Lombardi,U., et aletal to agonists and antagonists. Annals of the New York (2000) Buprenorphine-induced changes in mu-opioid Academy of Sciences,, 757757, 353^361.,353^361. Spielberger,C.Spielberger, C. D. (1983) Manual for the State-TraitState -Trait receptor availability in male heroin-dependent Anxiety Inventory (FormY). Consulting Psychologists Zubieta, J., Gorelick, D. A., Stauffer, R., et aletal (19 9 6) volunteers: a preliminary study. Press. Increased mu opioid receptor binding detected by PET Neuropsychopharmacology,, 2323, 326^334.

Downloaded from https://www.cambridge.org/core. 26 Sep 2021 at 22:39:25, subject to the Cambridge Core terms of use.