- Email: [email protected]
Cue reactivity and opioid blockade in amphetamine dependence: A randomized, controlled fMRI study
Drug and Alcohol Dependence 191 (2018) 91–97
Contents lists available at ScienceDirect
Drug and Alcohol Dependence journal homepage: www.elsevier.com/locate/drugalcdep
Full length article
Cue reactivity and opioid blockade in amphetamine dependence: A randomized, controlled fMRI study
T
⁎
Joar Guterstama, , Nitya Jayaram-Lindströma, Jonathan Berrebib, Predrag Petrovicb, Martin Ingvarb, Peter Franssonb, Johan Francka a
Centre for Psychiatry Research, Department of Clinical Neuroscience, Karolinska Institutet, and Stockholm Health Care Services, Stockholm County Council, Norra Stationsgatan 69, SE-113 64, Stockholm, Sweden b Department of Clinical Neuroscience, Karolinska Institutet, SE-171 77, Stockholm, Sweden
A R T I C LE I N FO
A B S T R A C T
Keywords: Amphetamine dependence Stimulant use disorder Functional magnetic resonance imaging Naltrexone
Background: The opioid antagonist, naltrexone, has been shown to reduce the risk of relapse in amphetamine dependence, but the mechanisms behind this effect are not well understood. We aimed to investigate if naltrexone attenuates cue reactivity and craving in amphetamine dependence. Methods: Forty men with severe, intravenous amphetamine dependence were randomized to one dose of naltrexone (50 mg) or placebo. In a BOLD fMRI cue reactivity paradigm, they were exposed to drug-related and neutral films and gave subjective ratings of craving after each film. Twenty-nine patients left data of sufficient quality to be included in the final analysis. Results: The drug-related films elicited strong subjective craving and BOLD activations of the striatum, cingulate cortex, and occipito-temporal visual attention networks. Longer history of amphetamine use was associated with greater activations of the prefrontal cortex. Naltrexone as compared to placebo had no significant effects on brain activations or subjective ratings. Conclusion: Patients with severe stimulant use disorder exhibit strong neural cue reactivity, the patterns of which are modulated by duration of drug use. In this sample, we found no evidence for any effects of naltrexone on cue reactivity.
1. Introduction Amphetamine use disorder is a global health problem for which there is still no approved pharmacological treatment. In Northern Europe, amphetamine has dominated injection drug use for decades (Hakansson et al., 2009). One of the few promising pharmacological treatments for amphetamine dependence is the opioid antagonist naltrexone (Karila et al., 2010), which is currently used clinically for the treatment of alcohol and opioid dependence (Lobmaier et al., 2011; Rösner et al., 2010). In a number of human laboratory studies, it has consistently been shown that pre-treatment with naltrexone attenuates the subjective effects of amphetamine (Jayaram-Lindström et al., 2004, 2008b; Marks et al., 2014; Ray et al., 2015). Interestingly, in randomized clinical trials, a significant effect of naltrexone has also been shown to prolong the time to first amphetamine use, a fact which obviously cannot be explained by naltrexone’s modulation of amphetamine effects (Jayaram-Lindström et al., 2008a).
One hypothesis is that naltrexone might attenuate craving, which has long been recognized as an important cause of relapse in addictive disorders (Drummond, 2001). Craving serves as a diagnostic criterion for addiction in both DSM 5 and ICD 10 and is a central concept in most theoretical models of these disorders (Tiffany and Wray, 2012). For instance, the theory of incentive sensitization predicts that the drugdependent subject’s ‘wanting’ of drugs increases over time, while the ‘liking’ of the drug actually diminishes and eventually plays a minor role in relapse and maintenance of problematic drug use (Berridge et al., 2009). Therefore, attenuating craving might be an important treatment target in severe addiction. Since subjective ratings of craving are only weakly associated with time to relapse and other clinically important outcomes (Tiffany and Wray, 2012), an important research goal is to develop laboratory models that objectively assess the processes involved in craving and relapse. The biological mechanisms of craving are not fully understood, but striatal dopamine release seems to be of importance, at least for
⁎ Corresponding author at: Karolinska Institutet, Department of Clinical Neuroscience, Center for Psychiatry Research, Norra Stationsgatan 69, 7th floor, 113 64 Stockholm, Sweden. E-mail address: [email protected] (J. Guterstam).
https://doi.org/10.1016/j.drugalcdep.2018.06.023 Received 20 February 2018; Received in revised form 5 June 2018; Accepted 18 June 2018 Available online 26 July 2018 0376-8716/ © 2018 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/BY-NC-ND/4.0/).
Drug and Alcohol Dependence 191 (2018) 91–97
J. Guterstam et al.
2.2. Study procedures
stimulant drugs (Berger et al., 1996; Volkow et al., 2006). However, there is also evidence for the involvement of the brain opioid system, and in alcohol dependence naltrexone has been shown to attenuate cueinduced craving and neural cue reactivity as measured with BOLD fMRI (Lukas et al., 2013; Myrick et al., 2008). This led us to hypothesize that naltrexone might also attenuate craving and neural cue reactivity in amphetamine dependent patients. At the start of this project, there were still no published studies of cue reactivity in amphetamine dependent patients. Since then, three fMRI studies with methamphetamine users have appeared (Courtney et al., 2016; Malcolm et al., 2016; Yin et al., 2012). However, two of these studies did not require the participants to fulfill diagnostic criteria for amphetamine dependence (Courtney et al., 2016; Yin et al., 2012), and their cases generally represent mild or moderate addiction severity. Courtney et al. also studied the effects of naltrexone vs. placebo using cross-over randomization and found that naltrexone attenuated activity in primary sensory and motor areas when viewing drug-related pictures (Courtney et al., 2016). In the present study, we only included patients with long histories of compulsive, intravenous amphetamine use and used a paradigm of movie cues designed to induce strong craving reactions in the participants.
After a first screening visit, eligible patients were scheduled for a test day. Having arrived in the clinic on the test day, the patients were asked for a supervised urine test to exclude current use of drugs other than amphetamine. If the result was negative and a breathalyser test showed no trace of alcohol, the patient was included in the study and randomized to one capsule of 50 mg naltrexone (naltrexone hydrochloride, APL, Stockholm) or an identical capsule with placebo. This dose was chosen based on our earlier laboratory work and clinical trials with amphetamine dependent patients, which have all used 50 mg of naltrexone (Jayaram-Lindström et al., 2005, 2008b, 2008a). Block randomization was performed by the providers of the study medication, and block size was unknown to the investigators. After an interval of at least 60 min following ingestion of the study medication, the fMRI procedures were started. The experiment described here consisted of the patients seeing film clips depicting drug related scenes (i.e., people preparing and taking drugs, both nasally and by injecting) or neutral scenes (e.g., old people drinking coffee or chatting). The visual content of the film types was not matched in detail but was broadly similar in colour, luminance, and presence of humans and faces. The drug related film clips had been developed in our lab, and pilot trials had shown them to elicit strong but transient craving responses in amphetamine dependent individuals (unpublished data). Nine film clips of each type, lasting 16 s each, were shown in a pseudo-randomized order. The films were separated by a short break in which the participants were asked to rate with a trackball their level of amphetamine craving on a Visual Analogue Scale on the screen, with 0 in one end representing no craving at all and 100 in the other end representing the maximum level of craving imaginable (Kober et al., 2016; Milella et al., 2016). This was followed by a brief fixation cross. In total, the experiment lasted for 7 min. After the examination, the patients were debriefed and asked about lingering craving and possible adverse events. All patients were offered clinical follow-up.
2. Material and methods This study was designed as a randomized, double-blind, placebocontrolled clinical trial with two parallel groups of amphetamine dependent patients. Each patient received one oral dose of naltrexone (50 mg) or identical placebo and then went through an fMRI examination with a paradigm of visual drug cues. The study was approved by the Stockholm Ethics Review Board and the Swedish Medical Products Agency. It was preregistered in the EU Clinical Trials Register (EudraCT 2010-021384-33) and performed according to International Conference on Harmonisation guidelines for Good Clinical Practice, with external monitoring by the Karolinska Trial Alliance. All participants provided their written, informed consent before participating in any study procedures.
2.3. Magnetic resonance imaging MRI examinations were performed with a 3 T instrument (GE MR750 Discovery) with an 8-channel head coil at the Karolinska MR Research Center. Each subject went through a total of four fMRI paradigms and structural imaging (T1 and T2 flair) lasting for about 50 min in total. The experiment described here was the last to be performed before the structural imaging and was preceded by a resting state examination and two different paradigms with still pictures (none of which elicited any significant drug craving in the subjects). For this experiment, we used EPI with gradient echo, slice thickness = 2.9 mm, number of slices = 43, TR = 2500 ms, TE = 30 ms, flip angle = 75°, FOV = 230 mm.
2.1. Participants 40 male, non-treatment seeking amphetamine users aged 20–65 were recruited via advertisement and word of mouth at the needle exchange program and in shelters in the Stockholm region. The exclusion of female participants was based on earlier fMRI studies suggesting different neural patterns between male and female stimulant users (Kilts et al., 2004), and we therefore chose to include only the most prevalent sex in this patient group to avoid introducing sex as a confounding variable in this small, experimental study. All participants were screened by a study physician, including psychiatric assessment with the Structured Clinical Interview for DSM-IV, Axis 1 (SCID-1), and detailed assessment of substance use history and other background variables with the Addiction Severity Index (McLellan et al., 1992). Inclusion criteria included DSM-IV diagnosis of amphetamine dependence since at least two years prior, history of intravenous amphetamine use, amphetamine use for a minimum of 12 days in the last 12 weeks, and having been drug free 1–30 days (minimum 24 h). The study was started before the release of DSM-5, but all patients would qualify for a diagnosis of severe amphetamine use disorder according to DSM-5. Exclusion criteria were other ongoing substance dependence (except nicotine), schizophrenia or bipolar disorder I, left-handedness, clinical signs of amphetamine intoxication at the day of testing, traces of cannabis, opiates, cocaine, or benzodiazepines in the urine at the day of testing, traces of alcohol as measured by breathalyser at the day of testing, or presence of severe somatic disorder. Patients with contraindications to MRI or the study medication were also excluded.
2.4. Data analysis fMRI data was analyzed with SPM 12 (http://www.fil.ion.ucl.ac.uk/ spm/software/spm12/). The data was realigned, co-registered to the participant’s structural image, segmented, and smoothed with a 7 mm FWHM Gaussian kernel. The images were then normalized to an MNI template. Since movement artifacts often represent a major problem in fMRI, we quantified this using frame-wise displacement (FD) (Power et al., 2012). The FD is computed for each frame (image volume) as the root-mean-squared difference between adjacent images in terms of rotational and translational movements (in total 6 movement parameters). We decided to set a limit for the level of movement allowed such that patients were excluded if more than 25% of their volumes had a FD > 0.3. The functional data were analyzed as a block design using the 92
Drug and Alcohol Dependence 191 (2018) 91–97
J. Guterstam et al.
analysis due to excessive head movements during the examination, i.e., > 25% of the image volumes above 0.3 FD. This meant that 29 patients (14 naltrexone, 15 placebo) had complete data of sufficient quality to be included in the analysis. The participants’ craving ratings were significantly higher after the drug-related compared to the neutral film clips (mean scores of 68 ± 31 vs. 16 ± 24, p < 0.001). There were no differences between the naltrexone and placebo groups on the craving ratings in any of the categories (p > 0.1). Comparing the BOLD activity when viewing drug-related compared to neutral scenes, we found statistically significant activations in the occipital, parietal and temporal cortices, the striatum, and cingular cortex (Fig. 1 and Table 2). However, there were no statistically significant differences between the naltrexone and placebo groups in the cue-induced BOLD activity of our regions of interest. The same results were found in the whole-brain analysis. In secondary analyses, we looked at moderators of cue reactivity in the whole sample. In earlier reviews of the literature, both the length of addiction history and the amount of recent drug use have been linked to increased cue reactivity in addiction (Jasinska et al., 2014). When added as a co-variate in the model, the number of amphetamine-use days in the last month did not correlate with cue-induced BOLD reactivity. However, years of drug use had a positive correlation with cueinduced activation of a cluster in the left dorsolateral prefrontal cortex (DLPFC), and this finding remained statistically significant at the cluster level after FWE correction (Fig. 2). Next, we computed the extent of spatial overlap between the fMRI activation pattern yielded by the drug-related versus neutral scenes depicted in Fig. 1 with each of the 29 brain networks delineated in (Thompson and Fransson, 2017). The largest overlap was found for network 2.4 (for reference see Fig. 3 in Thompson and Fransson (2017)) with a spatial overlap of 30.5 percent (Fig. 3). The psychological and anatomical network constructs most strongly associated with network 2.4 in the previous fMRI literature are phrases such as “recognition”, “occipitotemporal”, “object recognition” and “motion”. This means that the fMRI activation pattern shown in Fig. 1 is spatially overlapping with previous fMRI studies that have described their brain activations in the context of “recognition network”, “occipitotemporal network”, “motion network”, etc.
general linear model with main contrast being Drug movies > Neutral movies. All results were corrected for multiple comparisons using Family-Wise Error (FWE) correction with level of significance set at p < 0.05 (two-tailed). In the analysis of naltrexone vs placebo, we used the WFU PickAtlas software (http://fmri.wfubmc.edu/software/pickatlas) to assemble a mask in order to look only at pre-specified regions of interest. Based on the prior literature on cue reactivity in stimulant dependence (Chase et al., 2011), the mask included the ventral tegmental area, striatum, anterior cingulate cortex (ACC), orbitofrontal cortex, and medial prefrontal cortex. We also did a whole-brain analysis to investigate possible effects of naltrexone on other brain regions. We decided to present our results in the context of large-scale brain networks in order to provide a better understanding of the relationship between the neural correlates of cue reactivity and functional brain anatomy. To this end, we employed a recently developed networkbased functional brain atlas (Thompson and Fransson, 2017). The network atlas is formed by a mass-meta analysis of previously reported fMRI activations in Neurosynth (Yarkoni et al., 2011) combined with a survey of the literature (PubMed database) of search terms associated with network constructs. A hierarchical clustering technique (Rosvall and Bergstrom, 2008) allowed for the construction of a hierarchical spatial similarity map of psychologically and anatomically rooted constructs of brain networks resulting in 29 different network maps. The subjective craving outcomes were analyzed as the average of the Visual Analogue Scale ratings after the drug-related and neutral films, respectively. t-tests were used to compare the values of the two film categories and the naltrexone and placebo groups. 3. Results The participants recruited were 43 ± 10 years old and exhibited long histories of amphetamine dependence. They had a mean of 17 years of amphetamine use and on average had used amphetamine 22 days in the latest month. Most were unemployed and had a criminal history; they had spent a mean of 5 years incarcerated. There were no significant differences in baseline variables between the naltrexone and placebo groups. For more details on demographics, see Table 1. Forty patients were randomized, but some of them were not able to perform the fMRI examination because of suspected adverse reactions to the medication (n = 3; consisting of nausea or anxiety, two in the naltrexone and one in the placebo group), claustrophobia (n = 1), or technical problems (n = 3). One patient was lost to follow-up for unclear reasons and never took part in the fMRI examination; after randomization and having ingested the study medication at the research clinic, he disappeared during the un-supervised 50 m walk to the MR center. In total, 32 patients contributed data for analysis. No serious adverse events occurred during the study. For the fMRI analysis, three participants were excluded from the
4. Discussion In this randomized controlled trial, we examined the effects of one oral dose of naltrexone 50 mg as compared to placebo on cue-reactivity in non-treatment seeking individuals with severe amphetamine use disorder. The participants reported strong craving reactions to the drugrelated film cues, which also elicited highly significant BOLD activations, especially of visual attention networks. However, we found no evidence of any naltrexone effect on subjective craving or BOLD reactivity. Recently, a study of cue-induced craving in methamphetamine users found no effects of naltrexone on subjective craving ratings (Courtney et al., 2016), which is in line with our results. These findings can be compared with earlier laboratory studies that have consistently shown naltrexone pre-treatment to attenuate craving after amphetamine intake in amphetamine dependent patients (Jayaram-Lindström et al., 2008b; Ray et al., 2015). This discrepancy suggests that there might be other mechanisms behind cue-induced, as opposed to drug priminginduced, craving reactions in stimulant dependence. In clinical treatment trials, naltrexone-treated amphetamine dependent patients have generally reported attenuated craving over time (Jayaram-Lindström et al., 2008a; Tiihonen et al., 2012). In line with the argument above, this might be because naltrexone affects other forms of craving, e.g., related to drug priming or stress rather than that induced by conditioned cues. Long-term treatment with naltrexone, such as in a clinical trial, might also have different effects than a single
Table 1 Demographics of the patients included in the final analysis. Numeric variables are presented as means ± standard deviations.
Age Years of amphetamine use Days of amphetamine use during last month Days since last amphetamine use Chronic viral hepatitis HIV Nicotine dependence Living alone Years in school Months incarcerated
Naltrexone (n = 14)
Placebo (n = 15)
40.1 ± 10,2 12.6 ± 7.9 19.4 ± 10.6
45.9 ± 9.6 20.8 ± 11.2 24.5 ± 9.4
5.2 ± 4.6 64% 7% 79% 100% 10.8 ± 1.6 48.6 ± 90.8
4.6 ± 4.8 67% 0% 93% 67% 10.0 ± 3.9 76.5 ± 113.9
93
Drug and Alcohol Dependence 191 (2018) 91–97
J. Guterstam et al.
Fig. 1. Drug movie > Neutral movie activations for the whole sample, regardless of treatment condition (n = 29). Whole-brain results, p < 0.05, with Family-Wise Error correction for multiple comparisons.
measurements are hard to perform in the fMRI environment during an experiment with rapidly shifting stimuli, especially with an impulsive and restless patient population such as in the present study. Given the problems with existing subjective measures of craving, one could also argue that studies of cue reactivity should not limit themselves to mirroring such self-reports. Rather, the objective should be to work out pathophysiological information of more diagnostic and prognostic relevance that may translate into biomarkers useful both in the clinic and in evaluating new treatments. Another psychometric issue is that several patients rated their craving as 100% after most of the film clips, which means there could be a “ceiling effect” in the ratings. In this study, we prioritized the efficiency of the fMRI paradigm, and further studies need to consider how to balance this against the comprehensiveness of the subjective craving assessments. Our sample is the largest of any cue reactivity study in amphetamine dependence published so far, and our fMRI paradigm proved to be highly efficient. The BOLD activations during drug-related as opposed to neutral film cues reached whole-brain statistical significance even with strict controls for multiple comparisons. Several regional brain activations found in our study (e.g., of the striatum and cingular cortex) are consistent with the results of previous meta-analyses of cue reactivity in addiction (Chase et al., 2011; Jasinska et al., 2014). However, the most significant activations in the present sample were found in attention and visual networks of the occipital, parietal, and temporal lobes, which have also been reported in some earlier studies, albeit not consistently. This might be due to differing analytic approaches; some studies have not reported whole-brain results but only regions of interest, primarily in the limbic system (Chase et al., 2011). From a neurophysiological perspective, it is hardly surprising that attention networks are activated when the subject is presented with highly salient stimuli, such as drug-related films for persons with severe substance dependence.
Table 2 MNI coordinates and Z-values for Drug movie > Neutral movie activations. Anatomical region
Cluster size
X
Y
Z
Peak Zvalue
Middle Occipital Gyrus (R) Inferior Temporal Gyrus (R) Inferior Temporal Gyrus (R) Inferior Occipital Lobule (L) Inferior Occipital Lobule (L) Inferior Temporal Gyrus (L) Superior Parietal Lobule (R) Angular Gyrus (R) Superior Parietal Lobule (R) Fusiform Gyrus (R) Precentral Lobule (L) Precentral Lobule (L) Middle Orbitofrontal Lobule (L) Rolandic Operculum (L) Superior Frontal Lobule (L) Middle Cingulum (L) Middle Frontal Gyrus (L)
4834
12
32 46 44 −48 −46 −46 26 28 36 38 −46 −46 −22
−82 −50 −66 −72 −64 −48 −64 −64 −52 −34 2 6 38
12 −10 −8 −4 −14 −18 54 44 62 −28 32 20 −18
7.36 7.15 6.76 7.35 7.11 6.88 5.84 5.70 5.29 5.25 5.08 4.96 5.06
8 31 6 9
−40 −22 −6 −46
−2 0 −24 40
16 56 42 18
4.98 4.97 4.80 4.77
8712
695
101 85
acute dose, although the latter is enough for a nearly complete blockade of the endogenous opioid system (Lee et al., 1988; Preston and Bigelow, 1993; Schuh et al., 1999) and attenuation of the acute effects of amphetamine (Jayaram-Lindström et al., 2008b). In general, there is a need for further studies combining functional neuroimaging with clinical follow-up in treatment trials in order to assess the predictive validity of the paradigms used. We used a one-item visual-analog scale to assess craving as is common in neuroimaging studies. A challenge is that this might not capture the whole phenomenon of drug craving (Tiffany and Wray, 2012). However, more elaborate psychometric and physiological 94
Drug and Alcohol Dependence 191 (2018) 91–97
J. Guterstam et al.
Fig. 2. Activations correlating with number of years of amphetamine use; p < 0.05 with cluster-based Family-Wise Error correction.
et al., 2016). At present it is not obvious how to interpret those findings, since these brain regions are rarely implicated in drug cue reactivity, and it is unclear why they would be specifically affected by opioid blockade (Chase et al., 2011). In our sample, we found no evidence of any naltrexone effect in these areas. This discrepancy could possibly be
It is also noteworthy that another major, recent fMRI study of cue reactivity in methamphetamine users found similar activation patterns in their sample (Courtney et al., 2016). As noted above, they also found that naltrexone as compared to placebo attenuated activity in primary sensory and motor areas when viewing drug-related pictures (Courtney
Fig. 3. Brain activity related to the contrast (Drug film > Neutral film, marked in red in the figure) shown in relation to a best fit of the networks described in the brain network atlas (Thompson and Fransson, 2017, marked in blue in the figure), which is based on queries of PubMed records of the usage of network contexts in the literature and meta-information of fMRI brain activity in the Neurosynth database. The spatial overlap between brain activation (Drug > Neutral) and the best fit of the network atlas is marked in purple (for reference see network 2.4 in Fig. 3 in Thompson and Fransson, 2017) (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).
95
Drug and Alcohol Dependence 191 (2018) 91–97
J. Guterstam et al.
consistent (Li et al., 2013). Since the present findings indicate that the left DLPFC is more involved in craving reactions for patients with longterm amphetamine dependence, it strengthens the rationale for studies of this population with a similar protocol.
explained by differences between the populations included in the studies (our sample had more severe addiction histories with intravenous drug use) and differences in the fMRI paradigms employed (we had movie clips instead of still pictures). A number of patients had to be excluded from the fMRI analysis due to movement artifacts or other problems during image acquisition. This probably resulted in a somewhat under-powered final sample for the between-subject analysis of naltrexone vs. placebo which might have contributed to the lack of evidence for an effect of naltrexone on the fMRI measures. A cross-over randomized study design might have been more optimal; however, trying to repeat fMRI sessions on different days with the severely addicted, non-treatment seeking amphetamine users included here would hardly be feasible. Challenges related to statistical power in the field of neuroimaging have been intensely debated in recent years (Button et al., 2013). Future studies should aim for larger sample sizes and try to minimize movement-related data loss in order to evaluate treatment effects on BOLD cue reactivity with more precision. To facilitate further studies in this field, we aim to make anonymized, individual data from this experiment available for future meta-analyses. There is a strong need for more imaging studies of cue reactivity, particularly with prospective follow-up, in order to elucidate the clinical relevance of neural cue reactivity as a biomarker in addiction. Such studies are also needed to determine the regional anatomical specificity of such a biomarker: in much of the PET and early fMRI literature, the focus was strictly on limbic structures, but the findings were highly heterogeneous and the individual studies often underpowered with poor statistical control for false positives (Jasinska et al., 2014). The activations of cortical visual attention networks found in this and several other recent studies might actually be more robustly detected with fMRI. In this study, we did not find evidence of prefrontal cue reactivity in the whole sample. This differs somewhat from earlier studies that have reported extensive involvement of the prefrontal cortex (PFC) in response to drug cues in non-treatment seeking but not in treatment seeking populations (Chase et al., 2011). This has been hypothesized to be due to the role of the PFC in planning drug-seeking behavior in nontreatment seekers who perceive the opportunity to use drugs (Wilson et al., 2004) or that treatment seeking patients might have more extensive drug-related brain damage and therefore fail to recruit prefrontal regions in response to drug cues (Chase et al., 2011). Since we found that years of amphetamine use correlated with prefrontal activations (specifically in the left DLPFC), our results do not obviously fit with any of these hypotheses. The participants in our sample had long histories of severe addiction and possibly even brain damage but still were not seeking treatment; a possible reason could be the lack of effective treatment available for stimulant use disorders. Although they all certainly had the opportunity to seek out and take amphetamine as soon as the experiment was over, only the participants with particularly long drug use histories exhibited any significant PFC activations. This could be an effect of long-term changes in mesocortical neurocircuitry but needs to be further studied in other samples with long histories of substance use. This might also have treatment implications, since preclinical models have shown that dopamine-opioid interactions are dynamic such that naltrexone attenuates amphetamine-induced dopamine release after chronic, but not acute, amphetamine administration (Jayaram-Lindström et al., 2017). Future imaging studies of stimulant addiction should aim to develop and validate clinically useful diagnostic markers and inspire new therapeutic options for this hard-totreat population. DLPFC cue reactivity is also interesting, since several studies of transcranial magnetic stimulation have found that targeting this particular region might attenuate craving in addicted patients (Enokibara et al., 2016; Gorelick et al., 2014). A recent study of cocaine-dependent patients specifically targeting the left DLPFC found reductions of craving and cocaine use (Terraneo et al., 2016), but the literature is not
4.1. Limitations As mentioned above, we had limited statistical power in the final analysis of naltrexone vs. placebo, not least because of drop-outs and movement artifacts. Other possible limitations are the lack of a healthy control group and lack of a film condition with content that could be as attention-grabbing as the drug films but not drug-related (if such content could be envisioned for this patient group). 4.2. Conclusion In summary, in a sample with severe amphetamine addiction, we found that movies with drug-related content caused strong subjective craving and BOLD activations in several motivationally relevant brain regions and visual attention and recognition networks. Longer duration of drug use correlated with cue-reactivity in prefrontal regions. Naltrexone pre-treatment as compared to placebo was not found to affect cue-reactivity, which was possibly related to lack of statistical power in this comparison. These results highlight the strength of neural cue reactivity in severe substance use disorders. Future studies need to consider a combination of subjective, clinical, and neural assessments in order to validate this as a potential biomarker of relapse risk. Conflict of interest All authors declare that they have no conflicts of interest. Contributors JG, NJL, PP, MI and JF were responsible for the study concept and design. JG recruited and screened the patients and, together with JB, performed the imaging data acquisition. PF, JB and JG were responsible for the data analysis. All authors contributed to the interpretation of the data. JG drafted the manuscript and all authors critically reviewed its content and approved its final version. Funding This work wasfunded by grants from the Stockholm County Council (Forskar-ST; a “research residency” grant), Märta Lundqvist Foundation and Professor Bror Gadelius Memorial Fund to Joar Guterstam, the Swedish Research Council (2013-61X-21444-04-3), the Swedish Research Council for Health, Working life and Welfare (2013-1849), and the Swedish Brain Foundation to Johan Franck, the Swedish Research Council to Nitya Jayaram-Lindström (2009-34201-69804-23). The funding sources had no role in the design or procedures of the study, nor in the interpretation of the data, the writing of the report or the decision to submit it for publication. Acknowledgements We thank Camilla Hellspong and the staff at the Magnus Huss clinic for assistance with the clinical procedures of the study and the Stockholm needle exchange and Stockholm Drug Users Union for helping to recruit study participants. References Berger, S.P., Hall, S., Mickalian, J.D., Reid, M.S., Crawford, C.A., Delucchi, K., Carr, K., Hall, S., 1996. Haloperidol antagonism of cue-elicited cocaine craving. Lancet 347, 504–508.
96
Drug and Alcohol Dependence 191 (2018) 91–97
J. Guterstam et al.
NeuroImage 78, 176–185. Malcolm, R., Myrick, H., Li, X., Henderson, S., Brady, K.T., George, M.S., See, R.E., 2016. Regional brain activity in abstinent methamphetamine dependent males following cue exposure. J. Drug Abuse 2 pii 16. Marks, K.R., Lile, J.A., Stoops, W.W., Rush, C.R., 2014. Separate and combined impact of acute naltrexone and alprazolam on subjective and physiological effects of oral damphetamine in stimulant users. Psychopharmacology 231, 2741–2750. McLellan, A.T., Kushner, H., Metzger, D., Peters, R., Smith, I., Grissom, G., Pettinati, H., Argeriou, M., 1992. The fifth edition of the addiction severity index. J. Subst. Abuse Treat. 9, 199–213. Milella, M.S., Fotros, A., Gravel, P., Casey, K.F., Larcher, K., Verhaeghe, J.A.J., Cox, S.M.L., Reader, A.J., Dagher, A., Benkelfat, C., Leyton, M., 2016. Cocaine cue-induced dopamine release in the human prefrontal cortex. J. Psychiatry Neurosci. 41, 322–330. Myrick, H., Anton, R.F., Li, X., Henderson, S., Randall, P.K., Voronin, K., 2008. Effect of naltrexone and ondansetron on alcohol cue-induced activation of the ventral striatum in alcohol-dependent people. Arch. Gen. Psychiatry 65, 466–475. Power, J.D., Barnes, K.A., Snyder, A.Z., Schlaggar, B.L., Petersen, S.E., 2012. Spurious but systematic correlations in functional connectivity MRI networks arise from subject motion. NeuroImage 59, 2142–2154. Preston, K.L., Bigelow, G.E., 1993. Differential naltrexone antagonism of hydromorphone and pentazocine effects in human volunteers. J. Pharmacol. Exp. Ther. 264, 813–823. Ray, L.A., Bujarski, S., Courtney, K.E., Moallem, N.R., Lunny, K., Roche, D., Leventhal, A.M., Shoptaw, S., Heinzerling, K., London, E.D., Miotto, K., 2015. The effects of naltrexone on subjective response to methamphetamine in a clinical sample: A double-blind, placebo-controlled laboratory study. Neuropsychopharmacology 40, 2347–2356. Rösner, S., Hackl-Herrwerth, A., Leucht, S., Vecchi, S., Srisurapanont, M., Soyka, M., 2010. Opioid antagonists for alcohol dependence. Cochrane Database Syst. Rev. 12, CD001867. Rosvall, M., Bergstrom, C.T., 2008. Maps of random walks on complex networks reveal community structure. Proc. Natl. Acad. Sci. U. S. A. 105, 1118–1123. Schuh, K.J., Walsh, S.L., Stitzer, M.L., 1999. Onset, magnitude and duration of opioid blockade produced by buprenorphine and naltrexone in humans. Psychopharmacology 145, 162–174. Terraneo, A., Leggio, L., Saladini, M., Ermani, M., Bonci, A., Gallimberti, L., 2016. Transcranial magnetic stimulation of dorsolateral prefrontal cortex reduces cocaine use: a pilot study. Eur. Neuropsychopharmacol. J. Eur. Coll. Neuropsychopharmacol. 26, 37–44. Thompson, W.H., Fransson, P., 2017. Spatial confluence of psychological and anatomical network constructs in the human brain revealed by a mass meta-analysis of fMRI activation. Sci. Rep. 7, 44259. Tiffany, S.T., Wray, J.M., 2012. The clinical significance of drug craving. Ann. N. Y. Acad. Sci. 1248, 1–17. Tiihonen, J., Krupitsky, E., Verbitskaya, E., Blokhina, E., Mamontova, O., Föhr, J., Tuomola, P., Kuoppasalmi, K., Kiviniemi, V., Zwartau, E., 2012. Naltrexone implant for the treatment of polydrug dependence: a randomized controlled trial. Am. J. Psychiatry 169, 531–536. Volkow, N.D., Wang, G.-J., Telang, F., Fowler, J.S., Logan, J., Childress, A.-R., Jayne, M., Ma, Y., Wong, C., 2006. Cocaine cues and dopamine in dorsal striatum: mechanism of craving in cocaine addiction. J. Neurosci. 26, 6583–6588. Wilson, S.J., Sayette, M.A., Fiez, J.A., 2004. Prefrontal responses to drug cues: a neurocognitive analysis. Nat. Neurosci. 7, 211–214. Yarkoni, T., Poldrack, R.A., Nichols, T.E., Van Essen, D.C., Wager, T.D., 2011. Large-scale automated synthesis of human functional neuroimaging data. Nat. Methods 8, 665–670. Yin, J.-J., Ma, S.-H., Xu, K., Wang, Z.-X., Le, H.-B., Huang, J.-Z., Fang, K.-M., Liao, L.-M., Cai, Z.-L., 2012. Functional magnetic resonance imaging of methamphetamine craving. Clin. Imaging 36, 695–701.
Berridge, K.C., Robinson, T.E., Aldridge, J.W., 2009. Dissecting components of reward: “Liking”, “wanting”, and learning. Curr. Opin. Pharmacol. 9, 65–73. Button, K.S., Ioannidis, J.P.A., Mokrysz, C., Nosek, B.A., Flint, J., Robinson, E.S.J., Munafò, M.R., 2013. Power failure: why small sample size undermines the reliability of neuroscience. Nat. Rev. Neurosci. 14, 365–376. Chase, H.W., Eickhoff, S.B., Laird, A.R., Hogarth, L., 2011. The neural basis of drug stimulus processing and craving: an activation likelihood estimation meta-analysis. Biol. Psychiatry 70, 785–793. Courtney, K.E., Ghahremani, D.G., Ray, L.A., 2016. The effects of pharmacological opioid blockade on neural measures of drug cue-reactivity in humans. Neuropsychopharmacology 41, 2872–2881. Drummond, D.C., 2001. Theories of drug craving, ancient and modern. Addiction 96, 33–46. Enokibara, M., Trevizol, A., Shiozawa, P., Cordeiro, Q., 2016. Establishing an effective TMS protocol for craving in substance addiction: is it possible? Am. J. Addict. 25, 28–30. Gorelick, D.A., Zangen, A., George, M.S., 2014. Transcranial magnetic stimulation in the treatment of substance addiction. Ann. N. Y. Acad. Sci. 1327, 79–93. Hakansson, A., Schlyter, F., Berglund, M., 2009. Characteristics of primary amphetamine users in Sweden: a criminal justice population examined with the Addiction Severity Index. Eur. Addict. Res. 15, 10–18. Jasinska, A.J., Stein, E.A., Kaiser, J., Naumer, M.J., Yalachkov, Y., 2014. Factors modulating neural reactivity to drug cues in addiction: a survey of human neuroimaging studies. Neurosci. Biobehav. Rev. 38, 1–16. Jayaram-Lindström, N., Wennberg, P., Hurd, Y.L., Franck, J., 2004. Effects of naltrexone on the subjective response to amphetamine in healthy volunteers. J. Clin. Psychopharmacol. 24, 665–669. Jayaram-Lindström, N., Wennberg, P., Beck, O., Franck, J., 2005. An open clinical trial of naltrexone for amphetamine dependence: compliance and tolerability. Nord. J. Psychiatry 59, 167–171. Jayaram-Lindström, N., Hammarberg, A., Beck, O., Franck, J., 2008a. Naltrexone for the treatment of amphetamine dependence: a randomized, placebo-controlled trial. Am. J. Psychiatry 165, 1442–1448. Jayaram-Lindström, N., Konstenius, M., Eksborg, S., Beck, O., Hammarberg, A., Franck, J., 2008b. Naltrexone attenuates the subjective effects of amphetamine in patients with amphetamine dependence. Neuropsychopharmacology 33, 1856–1863. Jayaram-Lindström, N., Guterstam, J., Häggkvist, J., Ericson, M., Malmlöf, T., Schilström, B., Halldin, C., Cervenka, S., Saijo, T., Nordström, A.-L., Franck, J., 2017. Naltrexone modulates dopamine release following chronic, but not acute amphetamine administration: a translational study. Transl. Psychiatry 7, e1104. Karila, L., Weinstein, A., Aubin, H.-J., Benyamina, A., Reynaud, M., Batki, S.L., 2010. Pharmacological approaches to methamphetamine dependence: a focused review. Br. J. Clin. Pharmacol. 69, 578–592. Kilts, C.D., Gross, R.E., Ely, T.D., Drexler, K.P.G., 2004. The neural correlates of cueinduced craving in cocaine-dependent women. Am. J. Psychiatry 161, 233–241. Kober, H., Lacadie, C.M., Wexler, B.E., Malison, R.T., Sinha, R., Potenza, M.N., 2016. Brain activity during cocaine craving and gambling urges: an fMRI study. Neuropsychopharmacology 41, 628–637. Lee, M.C., Wagner, H.N., Tanada, S., Frost, J.J., Bice, A.N., Dannals, R.F., 1988. Duration of occupancy of opiate receptors by naltrexone. J. Nucl. Med. 29, 1207–1211. Li, X., Malcolm, R.J., Huebner, K., Hanlon, C.A., Taylor, J.J., Brady, K.T., George, M.S., See, R.E., 2013. Low frequency repetitive transcranial magnetic stimulation of the left dorsolateral prefrontal cortex transiently increases cue-induced craving for methamphetamine: a preliminary study. Drug Alcohol Depend. 133, 641–646. Lobmaier, P.P., Kunøe, N., Gossop, M., Waal, H., 2011. Naltrexone depot formulations for opioid and alcohol dependence: a systematic review. CNS Neurosci. Ther. 17, 629–636. Lukas, S.E., Lowen, S.B., Lindsey, K.P., Conn, N., Tartarini, W., Rodolico, J., Mallya, G., Palmer, C., Penetar, D.M., 2013. Extended-release naltrexone (XR-NTX) attenuates brain responses to alcohol cues in alcohol-dependent volunteers: a bold FMRI study.
97
Contents lists available at ScienceDirect
Drug and Alcohol Dependence journal homepage: www.elsevier.com/locate/drugalcdep
Full length article
Cue reactivity and opioid blockade in amphetamine dependence: A randomized, controlled fMRI study
T
⁎
Joar Guterstama, , Nitya Jayaram-Lindströma, Jonathan Berrebib, Predrag Petrovicb, Martin Ingvarb, Peter Franssonb, Johan Francka a
Centre for Psychiatry Research, Department of Clinical Neuroscience, Karolinska Institutet, and Stockholm Health Care Services, Stockholm County Council, Norra Stationsgatan 69, SE-113 64, Stockholm, Sweden b Department of Clinical Neuroscience, Karolinska Institutet, SE-171 77, Stockholm, Sweden
A R T I C LE I N FO
A B S T R A C T
Keywords: Amphetamine dependence Stimulant use disorder Functional magnetic resonance imaging Naltrexone
Background: The opioid antagonist, naltrexone, has been shown to reduce the risk of relapse in amphetamine dependence, but the mechanisms behind this effect are not well understood. We aimed to investigate if naltrexone attenuates cue reactivity and craving in amphetamine dependence. Methods: Forty men with severe, intravenous amphetamine dependence were randomized to one dose of naltrexone (50 mg) or placebo. In a BOLD fMRI cue reactivity paradigm, they were exposed to drug-related and neutral films and gave subjective ratings of craving after each film. Twenty-nine patients left data of sufficient quality to be included in the final analysis. Results: The drug-related films elicited strong subjective craving and BOLD activations of the striatum, cingulate cortex, and occipito-temporal visual attention networks. Longer history of amphetamine use was associated with greater activations of the prefrontal cortex. Naltrexone as compared to placebo had no significant effects on brain activations or subjective ratings. Conclusion: Patients with severe stimulant use disorder exhibit strong neural cue reactivity, the patterns of which are modulated by duration of drug use. In this sample, we found no evidence for any effects of naltrexone on cue reactivity.
1. Introduction Amphetamine use disorder is a global health problem for which there is still no approved pharmacological treatment. In Northern Europe, amphetamine has dominated injection drug use for decades (Hakansson et al., 2009). One of the few promising pharmacological treatments for amphetamine dependence is the opioid antagonist naltrexone (Karila et al., 2010), which is currently used clinically for the treatment of alcohol and opioid dependence (Lobmaier et al., 2011; Rösner et al., 2010). In a number of human laboratory studies, it has consistently been shown that pre-treatment with naltrexone attenuates the subjective effects of amphetamine (Jayaram-Lindström et al., 2004, 2008b; Marks et al., 2014; Ray et al., 2015). Interestingly, in randomized clinical trials, a significant effect of naltrexone has also been shown to prolong the time to first amphetamine use, a fact which obviously cannot be explained by naltrexone’s modulation of amphetamine effects (Jayaram-Lindström et al., 2008a).
One hypothesis is that naltrexone might attenuate craving, which has long been recognized as an important cause of relapse in addictive disorders (Drummond, 2001). Craving serves as a diagnostic criterion for addiction in both DSM 5 and ICD 10 and is a central concept in most theoretical models of these disorders (Tiffany and Wray, 2012). For instance, the theory of incentive sensitization predicts that the drugdependent subject’s ‘wanting’ of drugs increases over time, while the ‘liking’ of the drug actually diminishes and eventually plays a minor role in relapse and maintenance of problematic drug use (Berridge et al., 2009). Therefore, attenuating craving might be an important treatment target in severe addiction. Since subjective ratings of craving are only weakly associated with time to relapse and other clinically important outcomes (Tiffany and Wray, 2012), an important research goal is to develop laboratory models that objectively assess the processes involved in craving and relapse. The biological mechanisms of craving are not fully understood, but striatal dopamine release seems to be of importance, at least for
⁎ Corresponding author at: Karolinska Institutet, Department of Clinical Neuroscience, Center for Psychiatry Research, Norra Stationsgatan 69, 7th floor, 113 64 Stockholm, Sweden. E-mail address: [email protected] (J. Guterstam).
https://doi.org/10.1016/j.drugalcdep.2018.06.023 Received 20 February 2018; Received in revised form 5 June 2018; Accepted 18 June 2018 Available online 26 July 2018 0376-8716/ © 2018 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/BY-NC-ND/4.0/).
Drug and Alcohol Dependence 191 (2018) 91–97
J. Guterstam et al.
2.2. Study procedures
stimulant drugs (Berger et al., 1996; Volkow et al., 2006). However, there is also evidence for the involvement of the brain opioid system, and in alcohol dependence naltrexone has been shown to attenuate cueinduced craving and neural cue reactivity as measured with BOLD fMRI (Lukas et al., 2013; Myrick et al., 2008). This led us to hypothesize that naltrexone might also attenuate craving and neural cue reactivity in amphetamine dependent patients. At the start of this project, there were still no published studies of cue reactivity in amphetamine dependent patients. Since then, three fMRI studies with methamphetamine users have appeared (Courtney et al., 2016; Malcolm et al., 2016; Yin et al., 2012). However, two of these studies did not require the participants to fulfill diagnostic criteria for amphetamine dependence (Courtney et al., 2016; Yin et al., 2012), and their cases generally represent mild or moderate addiction severity. Courtney et al. also studied the effects of naltrexone vs. placebo using cross-over randomization and found that naltrexone attenuated activity in primary sensory and motor areas when viewing drug-related pictures (Courtney et al., 2016). In the present study, we only included patients with long histories of compulsive, intravenous amphetamine use and used a paradigm of movie cues designed to induce strong craving reactions in the participants.
After a first screening visit, eligible patients were scheduled for a test day. Having arrived in the clinic on the test day, the patients were asked for a supervised urine test to exclude current use of drugs other than amphetamine. If the result was negative and a breathalyser test showed no trace of alcohol, the patient was included in the study and randomized to one capsule of 50 mg naltrexone (naltrexone hydrochloride, APL, Stockholm) or an identical capsule with placebo. This dose was chosen based on our earlier laboratory work and clinical trials with amphetamine dependent patients, which have all used 50 mg of naltrexone (Jayaram-Lindström et al., 2005, 2008b, 2008a). Block randomization was performed by the providers of the study medication, and block size was unknown to the investigators. After an interval of at least 60 min following ingestion of the study medication, the fMRI procedures were started. The experiment described here consisted of the patients seeing film clips depicting drug related scenes (i.e., people preparing and taking drugs, both nasally and by injecting) or neutral scenes (e.g., old people drinking coffee or chatting). The visual content of the film types was not matched in detail but was broadly similar in colour, luminance, and presence of humans and faces. The drug related film clips had been developed in our lab, and pilot trials had shown them to elicit strong but transient craving responses in amphetamine dependent individuals (unpublished data). Nine film clips of each type, lasting 16 s each, were shown in a pseudo-randomized order. The films were separated by a short break in which the participants were asked to rate with a trackball their level of amphetamine craving on a Visual Analogue Scale on the screen, with 0 in one end representing no craving at all and 100 in the other end representing the maximum level of craving imaginable (Kober et al., 2016; Milella et al., 2016). This was followed by a brief fixation cross. In total, the experiment lasted for 7 min. After the examination, the patients were debriefed and asked about lingering craving and possible adverse events. All patients were offered clinical follow-up.
2. Material and methods This study was designed as a randomized, double-blind, placebocontrolled clinical trial with two parallel groups of amphetamine dependent patients. Each patient received one oral dose of naltrexone (50 mg) or identical placebo and then went through an fMRI examination with a paradigm of visual drug cues. The study was approved by the Stockholm Ethics Review Board and the Swedish Medical Products Agency. It was preregistered in the EU Clinical Trials Register (EudraCT 2010-021384-33) and performed according to International Conference on Harmonisation guidelines for Good Clinical Practice, with external monitoring by the Karolinska Trial Alliance. All participants provided their written, informed consent before participating in any study procedures.
2.3. Magnetic resonance imaging MRI examinations were performed with a 3 T instrument (GE MR750 Discovery) with an 8-channel head coil at the Karolinska MR Research Center. Each subject went through a total of four fMRI paradigms and structural imaging (T1 and T2 flair) lasting for about 50 min in total. The experiment described here was the last to be performed before the structural imaging and was preceded by a resting state examination and two different paradigms with still pictures (none of which elicited any significant drug craving in the subjects). For this experiment, we used EPI with gradient echo, slice thickness = 2.9 mm, number of slices = 43, TR = 2500 ms, TE = 30 ms, flip angle = 75°, FOV = 230 mm.
2.1. Participants 40 male, non-treatment seeking amphetamine users aged 20–65 were recruited via advertisement and word of mouth at the needle exchange program and in shelters in the Stockholm region. The exclusion of female participants was based on earlier fMRI studies suggesting different neural patterns between male and female stimulant users (Kilts et al., 2004), and we therefore chose to include only the most prevalent sex in this patient group to avoid introducing sex as a confounding variable in this small, experimental study. All participants were screened by a study physician, including psychiatric assessment with the Structured Clinical Interview for DSM-IV, Axis 1 (SCID-1), and detailed assessment of substance use history and other background variables with the Addiction Severity Index (McLellan et al., 1992). Inclusion criteria included DSM-IV diagnosis of amphetamine dependence since at least two years prior, history of intravenous amphetamine use, amphetamine use for a minimum of 12 days in the last 12 weeks, and having been drug free 1–30 days (minimum 24 h). The study was started before the release of DSM-5, but all patients would qualify for a diagnosis of severe amphetamine use disorder according to DSM-5. Exclusion criteria were other ongoing substance dependence (except nicotine), schizophrenia or bipolar disorder I, left-handedness, clinical signs of amphetamine intoxication at the day of testing, traces of cannabis, opiates, cocaine, or benzodiazepines in the urine at the day of testing, traces of alcohol as measured by breathalyser at the day of testing, or presence of severe somatic disorder. Patients with contraindications to MRI or the study medication were also excluded.
2.4. Data analysis fMRI data was analyzed with SPM 12 (http://www.fil.ion.ucl.ac.uk/ spm/software/spm12/). The data was realigned, co-registered to the participant’s structural image, segmented, and smoothed with a 7 mm FWHM Gaussian kernel. The images were then normalized to an MNI template. Since movement artifacts often represent a major problem in fMRI, we quantified this using frame-wise displacement (FD) (Power et al., 2012). The FD is computed for each frame (image volume) as the root-mean-squared difference between adjacent images in terms of rotational and translational movements (in total 6 movement parameters). We decided to set a limit for the level of movement allowed such that patients were excluded if more than 25% of their volumes had a FD > 0.3. The functional data were analyzed as a block design using the 92
Drug and Alcohol Dependence 191 (2018) 91–97
J. Guterstam et al.
analysis due to excessive head movements during the examination, i.e., > 25% of the image volumes above 0.3 FD. This meant that 29 patients (14 naltrexone, 15 placebo) had complete data of sufficient quality to be included in the analysis. The participants’ craving ratings were significantly higher after the drug-related compared to the neutral film clips (mean scores of 68 ± 31 vs. 16 ± 24, p < 0.001). There were no differences between the naltrexone and placebo groups on the craving ratings in any of the categories (p > 0.1). Comparing the BOLD activity when viewing drug-related compared to neutral scenes, we found statistically significant activations in the occipital, parietal and temporal cortices, the striatum, and cingular cortex (Fig. 1 and Table 2). However, there were no statistically significant differences between the naltrexone and placebo groups in the cue-induced BOLD activity of our regions of interest. The same results were found in the whole-brain analysis. In secondary analyses, we looked at moderators of cue reactivity in the whole sample. In earlier reviews of the literature, both the length of addiction history and the amount of recent drug use have been linked to increased cue reactivity in addiction (Jasinska et al., 2014). When added as a co-variate in the model, the number of amphetamine-use days in the last month did not correlate with cue-induced BOLD reactivity. However, years of drug use had a positive correlation with cueinduced activation of a cluster in the left dorsolateral prefrontal cortex (DLPFC), and this finding remained statistically significant at the cluster level after FWE correction (Fig. 2). Next, we computed the extent of spatial overlap between the fMRI activation pattern yielded by the drug-related versus neutral scenes depicted in Fig. 1 with each of the 29 brain networks delineated in (Thompson and Fransson, 2017). The largest overlap was found for network 2.4 (for reference see Fig. 3 in Thompson and Fransson (2017)) with a spatial overlap of 30.5 percent (Fig. 3). The psychological and anatomical network constructs most strongly associated with network 2.4 in the previous fMRI literature are phrases such as “recognition”, “occipitotemporal”, “object recognition” and “motion”. This means that the fMRI activation pattern shown in Fig. 1 is spatially overlapping with previous fMRI studies that have described their brain activations in the context of “recognition network”, “occipitotemporal network”, “motion network”, etc.
general linear model with main contrast being Drug movies > Neutral movies. All results were corrected for multiple comparisons using Family-Wise Error (FWE) correction with level of significance set at p < 0.05 (two-tailed). In the analysis of naltrexone vs placebo, we used the WFU PickAtlas software (http://fmri.wfubmc.edu/software/pickatlas) to assemble a mask in order to look only at pre-specified regions of interest. Based on the prior literature on cue reactivity in stimulant dependence (Chase et al., 2011), the mask included the ventral tegmental area, striatum, anterior cingulate cortex (ACC), orbitofrontal cortex, and medial prefrontal cortex. We also did a whole-brain analysis to investigate possible effects of naltrexone on other brain regions. We decided to present our results in the context of large-scale brain networks in order to provide a better understanding of the relationship between the neural correlates of cue reactivity and functional brain anatomy. To this end, we employed a recently developed networkbased functional brain atlas (Thompson and Fransson, 2017). The network atlas is formed by a mass-meta analysis of previously reported fMRI activations in Neurosynth (Yarkoni et al., 2011) combined with a survey of the literature (PubMed database) of search terms associated with network constructs. A hierarchical clustering technique (Rosvall and Bergstrom, 2008) allowed for the construction of a hierarchical spatial similarity map of psychologically and anatomically rooted constructs of brain networks resulting in 29 different network maps. The subjective craving outcomes were analyzed as the average of the Visual Analogue Scale ratings after the drug-related and neutral films, respectively. t-tests were used to compare the values of the two film categories and the naltrexone and placebo groups. 3. Results The participants recruited were 43 ± 10 years old and exhibited long histories of amphetamine dependence. They had a mean of 17 years of amphetamine use and on average had used amphetamine 22 days in the latest month. Most were unemployed and had a criminal history; they had spent a mean of 5 years incarcerated. There were no significant differences in baseline variables between the naltrexone and placebo groups. For more details on demographics, see Table 1. Forty patients were randomized, but some of them were not able to perform the fMRI examination because of suspected adverse reactions to the medication (n = 3; consisting of nausea or anxiety, two in the naltrexone and one in the placebo group), claustrophobia (n = 1), or technical problems (n = 3). One patient was lost to follow-up for unclear reasons and never took part in the fMRI examination; after randomization and having ingested the study medication at the research clinic, he disappeared during the un-supervised 50 m walk to the MR center. In total, 32 patients contributed data for analysis. No serious adverse events occurred during the study. For the fMRI analysis, three participants were excluded from the
4. Discussion In this randomized controlled trial, we examined the effects of one oral dose of naltrexone 50 mg as compared to placebo on cue-reactivity in non-treatment seeking individuals with severe amphetamine use disorder. The participants reported strong craving reactions to the drugrelated film cues, which also elicited highly significant BOLD activations, especially of visual attention networks. However, we found no evidence of any naltrexone effect on subjective craving or BOLD reactivity. Recently, a study of cue-induced craving in methamphetamine users found no effects of naltrexone on subjective craving ratings (Courtney et al., 2016), which is in line with our results. These findings can be compared with earlier laboratory studies that have consistently shown naltrexone pre-treatment to attenuate craving after amphetamine intake in amphetamine dependent patients (Jayaram-Lindström et al., 2008b; Ray et al., 2015). This discrepancy suggests that there might be other mechanisms behind cue-induced, as opposed to drug priminginduced, craving reactions in stimulant dependence. In clinical treatment trials, naltrexone-treated amphetamine dependent patients have generally reported attenuated craving over time (Jayaram-Lindström et al., 2008a; Tiihonen et al., 2012). In line with the argument above, this might be because naltrexone affects other forms of craving, e.g., related to drug priming or stress rather than that induced by conditioned cues. Long-term treatment with naltrexone, such as in a clinical trial, might also have different effects than a single
Table 1 Demographics of the patients included in the final analysis. Numeric variables are presented as means ± standard deviations.
Age Years of amphetamine use Days of amphetamine use during last month Days since last amphetamine use Chronic viral hepatitis HIV Nicotine dependence Living alone Years in school Months incarcerated
Naltrexone (n = 14)
Placebo (n = 15)
40.1 ± 10,2 12.6 ± 7.9 19.4 ± 10.6
45.9 ± 9.6 20.8 ± 11.2 24.5 ± 9.4
5.2 ± 4.6 64% 7% 79% 100% 10.8 ± 1.6 48.6 ± 90.8
4.6 ± 4.8 67% 0% 93% 67% 10.0 ± 3.9 76.5 ± 113.9
93
Drug and Alcohol Dependence 191 (2018) 91–97
J. Guterstam et al.
Fig. 1. Drug movie > Neutral movie activations for the whole sample, regardless of treatment condition (n = 29). Whole-brain results, p < 0.05, with Family-Wise Error correction for multiple comparisons.
measurements are hard to perform in the fMRI environment during an experiment with rapidly shifting stimuli, especially with an impulsive and restless patient population such as in the present study. Given the problems with existing subjective measures of craving, one could also argue that studies of cue reactivity should not limit themselves to mirroring such self-reports. Rather, the objective should be to work out pathophysiological information of more diagnostic and prognostic relevance that may translate into biomarkers useful both in the clinic and in evaluating new treatments. Another psychometric issue is that several patients rated their craving as 100% after most of the film clips, which means there could be a “ceiling effect” in the ratings. In this study, we prioritized the efficiency of the fMRI paradigm, and further studies need to consider how to balance this against the comprehensiveness of the subjective craving assessments. Our sample is the largest of any cue reactivity study in amphetamine dependence published so far, and our fMRI paradigm proved to be highly efficient. The BOLD activations during drug-related as opposed to neutral film cues reached whole-brain statistical significance even with strict controls for multiple comparisons. Several regional brain activations found in our study (e.g., of the striatum and cingular cortex) are consistent with the results of previous meta-analyses of cue reactivity in addiction (Chase et al., 2011; Jasinska et al., 2014). However, the most significant activations in the present sample were found in attention and visual networks of the occipital, parietal, and temporal lobes, which have also been reported in some earlier studies, albeit not consistently. This might be due to differing analytic approaches; some studies have not reported whole-brain results but only regions of interest, primarily in the limbic system (Chase et al., 2011). From a neurophysiological perspective, it is hardly surprising that attention networks are activated when the subject is presented with highly salient stimuli, such as drug-related films for persons with severe substance dependence.
Table 2 MNI coordinates and Z-values for Drug movie > Neutral movie activations. Anatomical region
Cluster size
X
Y
Z
Peak Zvalue
Middle Occipital Gyrus (R) Inferior Temporal Gyrus (R) Inferior Temporal Gyrus (R) Inferior Occipital Lobule (L) Inferior Occipital Lobule (L) Inferior Temporal Gyrus (L) Superior Parietal Lobule (R) Angular Gyrus (R) Superior Parietal Lobule (R) Fusiform Gyrus (R) Precentral Lobule (L) Precentral Lobule (L) Middle Orbitofrontal Lobule (L) Rolandic Operculum (L) Superior Frontal Lobule (L) Middle Cingulum (L) Middle Frontal Gyrus (L)
4834
12
32 46 44 −48 −46 −46 26 28 36 38 −46 −46 −22
−82 −50 −66 −72 −64 −48 −64 −64 −52 −34 2 6 38
12 −10 −8 −4 −14 −18 54 44 62 −28 32 20 −18
7.36 7.15 6.76 7.35 7.11 6.88 5.84 5.70 5.29 5.25 5.08 4.96 5.06
8 31 6 9
−40 −22 −6 −46
−2 0 −24 40
16 56 42 18
4.98 4.97 4.80 4.77
8712
695
101 85
acute dose, although the latter is enough for a nearly complete blockade of the endogenous opioid system (Lee et al., 1988; Preston and Bigelow, 1993; Schuh et al., 1999) and attenuation of the acute effects of amphetamine (Jayaram-Lindström et al., 2008b). In general, there is a need for further studies combining functional neuroimaging with clinical follow-up in treatment trials in order to assess the predictive validity of the paradigms used. We used a one-item visual-analog scale to assess craving as is common in neuroimaging studies. A challenge is that this might not capture the whole phenomenon of drug craving (Tiffany and Wray, 2012). However, more elaborate psychometric and physiological 94
Drug and Alcohol Dependence 191 (2018) 91–97
J. Guterstam et al.
Fig. 2. Activations correlating with number of years of amphetamine use; p < 0.05 with cluster-based Family-Wise Error correction.
et al., 2016). At present it is not obvious how to interpret those findings, since these brain regions are rarely implicated in drug cue reactivity, and it is unclear why they would be specifically affected by opioid blockade (Chase et al., 2011). In our sample, we found no evidence of any naltrexone effect in these areas. This discrepancy could possibly be
It is also noteworthy that another major, recent fMRI study of cue reactivity in methamphetamine users found similar activation patterns in their sample (Courtney et al., 2016). As noted above, they also found that naltrexone as compared to placebo attenuated activity in primary sensory and motor areas when viewing drug-related pictures (Courtney
Fig. 3. Brain activity related to the contrast (Drug film > Neutral film, marked in red in the figure) shown in relation to a best fit of the networks described in the brain network atlas (Thompson and Fransson, 2017, marked in blue in the figure), which is based on queries of PubMed records of the usage of network contexts in the literature and meta-information of fMRI brain activity in the Neurosynth database. The spatial overlap between brain activation (Drug > Neutral) and the best fit of the network atlas is marked in purple (for reference see network 2.4 in Fig. 3 in Thompson and Fransson, 2017) (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article).
95
Drug and Alcohol Dependence 191 (2018) 91–97
J. Guterstam et al.
consistent (Li et al., 2013). Since the present findings indicate that the left DLPFC is more involved in craving reactions for patients with longterm amphetamine dependence, it strengthens the rationale for studies of this population with a similar protocol.
explained by differences between the populations included in the studies (our sample had more severe addiction histories with intravenous drug use) and differences in the fMRI paradigms employed (we had movie clips instead of still pictures). A number of patients had to be excluded from the fMRI analysis due to movement artifacts or other problems during image acquisition. This probably resulted in a somewhat under-powered final sample for the between-subject analysis of naltrexone vs. placebo which might have contributed to the lack of evidence for an effect of naltrexone on the fMRI measures. A cross-over randomized study design might have been more optimal; however, trying to repeat fMRI sessions on different days with the severely addicted, non-treatment seeking amphetamine users included here would hardly be feasible. Challenges related to statistical power in the field of neuroimaging have been intensely debated in recent years (Button et al., 2013). Future studies should aim for larger sample sizes and try to minimize movement-related data loss in order to evaluate treatment effects on BOLD cue reactivity with more precision. To facilitate further studies in this field, we aim to make anonymized, individual data from this experiment available for future meta-analyses. There is a strong need for more imaging studies of cue reactivity, particularly with prospective follow-up, in order to elucidate the clinical relevance of neural cue reactivity as a biomarker in addiction. Such studies are also needed to determine the regional anatomical specificity of such a biomarker: in much of the PET and early fMRI literature, the focus was strictly on limbic structures, but the findings were highly heterogeneous and the individual studies often underpowered with poor statistical control for false positives (Jasinska et al., 2014). The activations of cortical visual attention networks found in this and several other recent studies might actually be more robustly detected with fMRI. In this study, we did not find evidence of prefrontal cue reactivity in the whole sample. This differs somewhat from earlier studies that have reported extensive involvement of the prefrontal cortex (PFC) in response to drug cues in non-treatment seeking but not in treatment seeking populations (Chase et al., 2011). This has been hypothesized to be due to the role of the PFC in planning drug-seeking behavior in nontreatment seekers who perceive the opportunity to use drugs (Wilson et al., 2004) or that treatment seeking patients might have more extensive drug-related brain damage and therefore fail to recruit prefrontal regions in response to drug cues (Chase et al., 2011). Since we found that years of amphetamine use correlated with prefrontal activations (specifically in the left DLPFC), our results do not obviously fit with any of these hypotheses. The participants in our sample had long histories of severe addiction and possibly even brain damage but still were not seeking treatment; a possible reason could be the lack of effective treatment available for stimulant use disorders. Although they all certainly had the opportunity to seek out and take amphetamine as soon as the experiment was over, only the participants with particularly long drug use histories exhibited any significant PFC activations. This could be an effect of long-term changes in mesocortical neurocircuitry but needs to be further studied in other samples with long histories of substance use. This might also have treatment implications, since preclinical models have shown that dopamine-opioid interactions are dynamic such that naltrexone attenuates amphetamine-induced dopamine release after chronic, but not acute, amphetamine administration (Jayaram-Lindström et al., 2017). Future imaging studies of stimulant addiction should aim to develop and validate clinically useful diagnostic markers and inspire new therapeutic options for this hard-totreat population. DLPFC cue reactivity is also interesting, since several studies of transcranial magnetic stimulation have found that targeting this particular region might attenuate craving in addicted patients (Enokibara et al., 2016; Gorelick et al., 2014). A recent study of cocaine-dependent patients specifically targeting the left DLPFC found reductions of craving and cocaine use (Terraneo et al., 2016), but the literature is not
4.1. Limitations As mentioned above, we had limited statistical power in the final analysis of naltrexone vs. placebo, not least because of drop-outs and movement artifacts. Other possible limitations are the lack of a healthy control group and lack of a film condition with content that could be as attention-grabbing as the drug films but not drug-related (if such content could be envisioned for this patient group). 4.2. Conclusion In summary, in a sample with severe amphetamine addiction, we found that movies with drug-related content caused strong subjective craving and BOLD activations in several motivationally relevant brain regions and visual attention and recognition networks. Longer duration of drug use correlated with cue-reactivity in prefrontal regions. Naltrexone pre-treatment as compared to placebo was not found to affect cue-reactivity, which was possibly related to lack of statistical power in this comparison. These results highlight the strength of neural cue reactivity in severe substance use disorders. Future studies need to consider a combination of subjective, clinical, and neural assessments in order to validate this as a potential biomarker of relapse risk. Conflict of interest All authors declare that they have no conflicts of interest. Contributors JG, NJL, PP, MI and JF were responsible for the study concept and design. JG recruited and screened the patients and, together with JB, performed the imaging data acquisition. PF, JB and JG were responsible for the data analysis. All authors contributed to the interpretation of the data. JG drafted the manuscript and all authors critically reviewed its content and approved its final version. Funding This work wasfunded by grants from the Stockholm County Council (Forskar-ST; a “research residency” grant), Märta Lundqvist Foundation and Professor Bror Gadelius Memorial Fund to Joar Guterstam, the Swedish Research Council (2013-61X-21444-04-3), the Swedish Research Council for Health, Working life and Welfare (2013-1849), and the Swedish Brain Foundation to Johan Franck, the Swedish Research Council to Nitya Jayaram-Lindström (2009-34201-69804-23). The funding sources had no role in the design or procedures of the study, nor in the interpretation of the data, the writing of the report or the decision to submit it for publication. Acknowledgements We thank Camilla Hellspong and the staff at the Magnus Huss clinic for assistance with the clinical procedures of the study and the Stockholm needle exchange and Stockholm Drug Users Union for helping to recruit study participants. References Berger, S.P., Hall, S., Mickalian, J.D., Reid, M.S., Crawford, C.A., Delucchi, K., Carr, K., Hall, S., 1996. Haloperidol antagonism of cue-elicited cocaine craving. Lancet 347, 504–508.
96
Drug and Alcohol Dependence 191 (2018) 91–97
J. Guterstam et al.
NeuroImage 78, 176–185. Malcolm, R., Myrick, H., Li, X., Henderson, S., Brady, K.T., George, M.S., See, R.E., 2016. Regional brain activity in abstinent methamphetamine dependent males following cue exposure. J. Drug Abuse 2 pii 16. Marks, K.R., Lile, J.A., Stoops, W.W., Rush, C.R., 2014. Separate and combined impact of acute naltrexone and alprazolam on subjective and physiological effects of oral damphetamine in stimulant users. Psychopharmacology 231, 2741–2750. McLellan, A.T., Kushner, H., Metzger, D., Peters, R., Smith, I., Grissom, G., Pettinati, H., Argeriou, M., 1992. The fifth edition of the addiction severity index. J. Subst. Abuse Treat. 9, 199–213. Milella, M.S., Fotros, A., Gravel, P., Casey, K.F., Larcher, K., Verhaeghe, J.A.J., Cox, S.M.L., Reader, A.J., Dagher, A., Benkelfat, C., Leyton, M., 2016. Cocaine cue-induced dopamine release in the human prefrontal cortex. J. Psychiatry Neurosci. 41, 322–330. Myrick, H., Anton, R.F., Li, X., Henderson, S., Randall, P.K., Voronin, K., 2008. Effect of naltrexone and ondansetron on alcohol cue-induced activation of the ventral striatum in alcohol-dependent people. Arch. Gen. Psychiatry 65, 466–475. Power, J.D., Barnes, K.A., Snyder, A.Z., Schlaggar, B.L., Petersen, S.E., 2012. Spurious but systematic correlations in functional connectivity MRI networks arise from subject motion. NeuroImage 59, 2142–2154. Preston, K.L., Bigelow, G.E., 1993. Differential naltrexone antagonism of hydromorphone and pentazocine effects in human volunteers. J. Pharmacol. Exp. Ther. 264, 813–823. Ray, L.A., Bujarski, S., Courtney, K.E., Moallem, N.R., Lunny, K., Roche, D., Leventhal, A.M., Shoptaw, S., Heinzerling, K., London, E.D., Miotto, K., 2015. The effects of naltrexone on subjective response to methamphetamine in a clinical sample: A double-blind, placebo-controlled laboratory study. Neuropsychopharmacology 40, 2347–2356. Rösner, S., Hackl-Herrwerth, A., Leucht, S., Vecchi, S., Srisurapanont, M., Soyka, M., 2010. Opioid antagonists for alcohol dependence. Cochrane Database Syst. Rev. 12, CD001867. Rosvall, M., Bergstrom, C.T., 2008. Maps of random walks on complex networks reveal community structure. Proc. Natl. Acad. Sci. U. S. A. 105, 1118–1123. Schuh, K.J., Walsh, S.L., Stitzer, M.L., 1999. Onset, magnitude and duration of opioid blockade produced by buprenorphine and naltrexone in humans. Psychopharmacology 145, 162–174. Terraneo, A., Leggio, L., Saladini, M., Ermani, M., Bonci, A., Gallimberti, L., 2016. Transcranial magnetic stimulation of dorsolateral prefrontal cortex reduces cocaine use: a pilot study. Eur. Neuropsychopharmacol. J. Eur. Coll. Neuropsychopharmacol. 26, 37–44. Thompson, W.H., Fransson, P., 2017. Spatial confluence of psychological and anatomical network constructs in the human brain revealed by a mass meta-analysis of fMRI activation. Sci. Rep. 7, 44259. Tiffany, S.T., Wray, J.M., 2012. The clinical significance of drug craving. Ann. N. Y. Acad. Sci. 1248, 1–17. Tiihonen, J., Krupitsky, E., Verbitskaya, E., Blokhina, E., Mamontova, O., Föhr, J., Tuomola, P., Kuoppasalmi, K., Kiviniemi, V., Zwartau, E., 2012. Naltrexone implant for the treatment of polydrug dependence: a randomized controlled trial. Am. J. Psychiatry 169, 531–536. Volkow, N.D., Wang, G.-J., Telang, F., Fowler, J.S., Logan, J., Childress, A.-R., Jayne, M., Ma, Y., Wong, C., 2006. Cocaine cues and dopamine in dorsal striatum: mechanism of craving in cocaine addiction. J. Neurosci. 26, 6583–6588. Wilson, S.J., Sayette, M.A., Fiez, J.A., 2004. Prefrontal responses to drug cues: a neurocognitive analysis. Nat. Neurosci. 7, 211–214. Yarkoni, T., Poldrack, R.A., Nichols, T.E., Van Essen, D.C., Wager, T.D., 2011. Large-scale automated synthesis of human functional neuroimaging data. Nat. Methods 8, 665–670. Yin, J.-J., Ma, S.-H., Xu, K., Wang, Z.-X., Le, H.-B., Huang, J.-Z., Fang, K.-M., Liao, L.-M., Cai, Z.-L., 2012. Functional magnetic resonance imaging of methamphetamine craving. Clin. Imaging 36, 695–701.
Berridge, K.C., Robinson, T.E., Aldridge, J.W., 2009. Dissecting components of reward: “Liking”, “wanting”, and learning. Curr. Opin. Pharmacol. 9, 65–73. Button, K.S., Ioannidis, J.P.A., Mokrysz, C., Nosek, B.A., Flint, J., Robinson, E.S.J., Munafò, M.R., 2013. Power failure: why small sample size undermines the reliability of neuroscience. Nat. Rev. Neurosci. 14, 365–376. Chase, H.W., Eickhoff, S.B., Laird, A.R., Hogarth, L., 2011. The neural basis of drug stimulus processing and craving: an activation likelihood estimation meta-analysis. Biol. Psychiatry 70, 785–793. Courtney, K.E., Ghahremani, D.G., Ray, L.A., 2016. The effects of pharmacological opioid blockade on neural measures of drug cue-reactivity in humans. Neuropsychopharmacology 41, 2872–2881. Drummond, D.C., 2001. Theories of drug craving, ancient and modern. Addiction 96, 33–46. Enokibara, M., Trevizol, A., Shiozawa, P., Cordeiro, Q., 2016. Establishing an effective TMS protocol for craving in substance addiction: is it possible? Am. J. Addict. 25, 28–30. Gorelick, D.A., Zangen, A., George, M.S., 2014. Transcranial magnetic stimulation in the treatment of substance addiction. Ann. N. Y. Acad. Sci. 1327, 79–93. Hakansson, A., Schlyter, F., Berglund, M., 2009. Characteristics of primary amphetamine users in Sweden: a criminal justice population examined with the Addiction Severity Index. Eur. Addict. Res. 15, 10–18. Jasinska, A.J., Stein, E.A., Kaiser, J., Naumer, M.J., Yalachkov, Y., 2014. Factors modulating neural reactivity to drug cues in addiction: a survey of human neuroimaging studies. Neurosci. Biobehav. Rev. 38, 1–16. Jayaram-Lindström, N., Wennberg, P., Hurd, Y.L., Franck, J., 2004. Effects of naltrexone on the subjective response to amphetamine in healthy volunteers. J. Clin. Psychopharmacol. 24, 665–669. Jayaram-Lindström, N., Wennberg, P., Beck, O., Franck, J., 2005. An open clinical trial of naltrexone for amphetamine dependence: compliance and tolerability. Nord. J. Psychiatry 59, 167–171. Jayaram-Lindström, N., Hammarberg, A., Beck, O., Franck, J., 2008a. Naltrexone for the treatment of amphetamine dependence: a randomized, placebo-controlled trial. Am. J. Psychiatry 165, 1442–1448. Jayaram-Lindström, N., Konstenius, M., Eksborg, S., Beck, O., Hammarberg, A., Franck, J., 2008b. Naltrexone attenuates the subjective effects of amphetamine in patients with amphetamine dependence. Neuropsychopharmacology 33, 1856–1863. Jayaram-Lindström, N., Guterstam, J., Häggkvist, J., Ericson, M., Malmlöf, T., Schilström, B., Halldin, C., Cervenka, S., Saijo, T., Nordström, A.-L., Franck, J., 2017. Naltrexone modulates dopamine release following chronic, but not acute amphetamine administration: a translational study. Transl. Psychiatry 7, e1104. Karila, L., Weinstein, A., Aubin, H.-J., Benyamina, A., Reynaud, M., Batki, S.L., 2010. Pharmacological approaches to methamphetamine dependence: a focused review. Br. J. Clin. Pharmacol. 69, 578–592. Kilts, C.D., Gross, R.E., Ely, T.D., Drexler, K.P.G., 2004. The neural correlates of cueinduced craving in cocaine-dependent women. Am. J. Psychiatry 161, 233–241. Kober, H., Lacadie, C.M., Wexler, B.E., Malison, R.T., Sinha, R., Potenza, M.N., 2016. Brain activity during cocaine craving and gambling urges: an fMRI study. Neuropsychopharmacology 41, 628–637. Lee, M.C., Wagner, H.N., Tanada, S., Frost, J.J., Bice, A.N., Dannals, R.F., 1988. Duration of occupancy of opiate receptors by naltrexone. J. Nucl. Med. 29, 1207–1211. Li, X., Malcolm, R.J., Huebner, K., Hanlon, C.A., Taylor, J.J., Brady, K.T., George, M.S., See, R.E., 2013. Low frequency repetitive transcranial magnetic stimulation of the left dorsolateral prefrontal cortex transiently increases cue-induced craving for methamphetamine: a preliminary study. Drug Alcohol Depend. 133, 641–646. Lobmaier, P.P., Kunøe, N., Gossop, M., Waal, H., 2011. Naltrexone depot formulations for opioid and alcohol dependence: a systematic review. CNS Neurosci. Ther. 17, 629–636. Lukas, S.E., Lowen, S.B., Lindsey, K.P., Conn, N., Tartarini, W., Rodolico, J., Mallya, G., Palmer, C., Penetar, D.M., 2013. Extended-release naltrexone (XR-NTX) attenuates brain responses to alcohol cues in alcohol-dependent volunteers: a bold FMRI study.
97