This research was supported by the National Institute on Drug Abuse grant #DA-03746. We gratefully acknowledge the efforts of Michaela Bamdad and Catalina Saldaña, who read an earlier version of the manuscript and made helpful suggestions. MG thanks the Instituto de Salud Carlos III for making his contribution possible through a Rio Hortega grant (CM18/00168)ISCIII/FSE
Cannabis, which comprises Δ 9 -tetrahydrocannabinol-containing products including marijuana and hashish, is the most widely used illicit drug in the world, with 183 million people reporting annual use in 2015. During 2015, in the United States alone, an estimated 22.2 million (8.3%) individuals report current marijuana use, defined as use within the past 30 days. Most users of cannabis consume the drug infrequently and without apparent negative consequences. There is, however, a small proportion of users who experience problems related to frequent cannabis use. It has been estimated that the cumulative probability of transitioning from use to dependence is 8.9% for cannabis users. Although this number is low compared with dependence rates for nicotine users (67.5% of tobacco users will become dependent), rates of cannabis dependence in several countries have increased substantially over the past decade as well as the number of individuals seeking treatment for cannabis-related problems . , The terms “dependence” and “dependent” encompass the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision (DSM-IV-TR) and the 10th revision of the International Statistical Classification of Diseases and Related Health Problems .
Although the total number of cannabis-dependent individuals who seek treatment is higher than the number of individuals who seek treatment for other illicit drugs, the relative proportion of those seeking treatment for cannabis dependence is low. For example, in the United States, the percentage of regular drug users who received treatment for a cannabis use disorder (includes cannabis abuse and dependence) in 2015 was around 5%, whereas this number was nearly 20% for cocaine users. Several possible explanations for the relatively low percentage of cannabis treatment seekers include the fact that many individuals perceive cannabis as a relatively innocuous drug. However, several investigators have reported that heavy, daily cannabis use is associated with an abstinence syndrome upon cessation of the drug (for review, see Budney ). Although cannabis withdrawal is not life-threatening, the accompanying symptoms such as irritability, anxiety, sleep disruptions, aches, and pains can be quite unpleasant. In addition, many individuals seeking treatment for cannabis dependence reported that these symptoms made it more difficult to maintain abstinence.
In addition, heavy cannabis use has been reported to be associated with poor cognitive performance. For example, Bolla and colleagues reported that heavy use of cannabis was associated with poorer cognitive performance on a wide range of tasks (e.g., memory and executive functioning) and that decreased performance persisted as long as 28 days of abstinence. Lifetime exposure to cannabis alone higher than 7 joint-years is also possibly associated with pulmonary dysfunction. The concept of joint-years indicates the cumulative dose of cannabis ingested in lifetime, being, for example, in the case of 7 joint-years either one joint every day during 7 years or seven joints every day during 1 year. Regarding the controverted relationship between cannabis and psychosis, Ksir and colleagues published a recent review. In their study, they concluded that both early use of cannabis and heavy use of cannabis were more likely in individuals with a vulnerability to a variety of problems, such as early or heavy use of cigarettes or alcohol, use of other illicit drugs, and poor school performance. In some individuals, the same vulnerability also resulted in increased risk for psychosis or some other mental disorder.
Some investigators have speculated that the low percentage of individuals seeking treatment for cannabis dependence may be related to the fact that there are relatively few specific treatments for cannabis dependence, although this issue does not appear to deter treatment-seeking cocaine abusers. Regular cannabis users may also be reluctant to participate in treatment programs dominated by alcohol-, cocaine-, and opioid-dependent individuals. Preference to quit without treatment and fear of stigma seem to be among the main barriers to seeking treatment. There are data indicating that some cannabis dependence–specific therapies are successful in decreasing drug use and many associated negative consequences. Other data, however, show that cannabis-dependent individuals exhibit high rates of relapse, similar to those found with other substances of abuse. To date, the majority of treatment studies have investigated behavioral/psychosocial therapies. The development of pharmacotherapy presents another option that would be available to cannabis-dependent individuals who have a high relapse rate. Pharmacotherapies may be used alone, in combination with behavioral/psychosocial therapies, or in a staged manner following inadequate response to behavioral/psychosocial therapies. In general, the problem in treating substance-dependent individuals has been less that of treating withdrawal and more of preventing relapse. However, treating withdrawal symptoms continues to be an important first step in eventual success and one that clinicians often need to begin this therapeutic endeavor. The treatment of cannabis dependence in this regard is similar to efforts underway for decades for opioids, cocaine, and alcohol dependence.
Since the original review in 2005, there have been several reviews addressing pharmacotherapies for cannabis dependence. This chapter incorporates most recent reviews and extends them by including the most recent studies. The chapter reviews findings from recent research on cannabinoids (a group of compounds related to Δ 9 -tetrahydrocannabinol, the primary psychopharmacologically active constituent of marijuana smoke) that may be relevant for the development of pharmacotherapies for cannabis dependence. Data from studies that assessed the ability of medications to attenuate cannabinoid-related abstinence symptoms in laboratory animals and in humans will be reviewed. In addition, results from studies that have investigated the effects of pharmacological agents on response to cannabinoids are reviewed because these data may prove useful in informing the development of cannabis relapse prevention medications. The review begins with a brief overview of the different phases of the dependence cycle that cannabis pharmacotherapies might target as well as cannabinoid relevant neuropharmacology.
Detoxification and Relapse Prevention or Maintenance Phase
Medications are typically initiated at two different phases of the dependence cycle: during detoxification and prevention of relapse. Detoxification is usually an initial and immediate goal during which medications are administered to assuage unpleasant abstinence symptoms that may appear following abrupt cessation of drug use, for example, the administration of a benzodiazepine during alcohol withdrawal. Medications used in the detoxification phase are also sometimes used in the relapse prevention or maintenance phase, for example, nicotine replacement medications. Thus the distinction between a detoxification medication and a relapse prevention medication is sometimes less clear. It is also important for us to note that although we recognize that the goal of this chapter is to review pharmacotherapies that may have some utility in decreasing cannabis withdrawal symptoms, we want to be careful not to overstate the problem of cannabis withdrawal because such symptoms may play a limited role in the addictive process when compared with other important psychosocial factors.
Maintenance medications can be viewed as a longer-term strategy used to help the dependent individual avoid relapsing to the abused drug. There are at least three major maintenance strategies. First, agonist or substitution therapy is used to induce cross-tolerance to the abused drug. For example, methadone (a long-acting μ-opioid agonist) and nicotine replacement medications have been used for opioid dependence and tobacco dependence, respectively, as agonist maintenance treatments to prevent relapse and cravings in individuals attempting to maintain abstinence. Agonist maintenance agents typically have safer routes of administration and diminished psychoactive effects relative to the abused drug. Second, antagonist therapy is used to produce extinction by preventing the user from experiencing the reinforcing effects of the abused drug. For example, the naltrexone blocks opioid mu receptors and agonists’ associated effects and is therefore used as an antagonist therapy for opioid dependence. Finally, punishment therapy produces an aversive reaction following ingestion of the abused drug. For example, disulfiram (Antabuse) is used in the treatment of alcohol dependence. Disulfiram inhibits aldehyde dehydrogenase, a major enzyme involved in alcohol metabolism, thereby preventing the complete breakdown of alcohol, and the resultant accumulation of aldehyde produces unpleasant symptoms including headache, vomiting, and breathing difficulties.
Over the past two decades, data from basic research have contributed to an increased understanding of neuronal mechanisms involved in the effects of cannabinoids. Although a comprehensive review of cannabinoid neuropharmacology is beyond the scope of the current manuscript and such reviews have already been published, , , a brief overview might be informative for the rationale regarding some of the medications presented in this review. Cannabinoids bind to two types of receptors: cannabinoid receptor 1 and cannabinoid receptor 2 (CB1 and CB2). These receptors are much more abundant than opioid receptors suggesting that the potential actions of cannabinoids are widespread. CB2 receptors are found mainly outside of the brain in immune cells, suggesting that cannabinoids may play a role in the modulation of the immune response. CB1 receptors are found throughout the body, but primarily in the central nervous system. The regions in which central nervous system CB1 receptors reside may provide some clues about their functions. For example, the highest density of CB1 receptors has been found in cells of the basal ganglia, the primary components of which include the caudate nucleus, putamen, and globus pallidus (for review, see Pertwee and Ross and Pertwee ). Cells of the basal ganglia are involved in coordinating body movements. Other regions that also contain a larger number of CB1 receptors include: the cerebellum , which coordinates fine body movements; the hippocampus , which is involved in aspects of memory storage; the cerebral cortex , which regulates the integration of higher cognitive functions; and the nucleus accumbens , which is involved in drug reinforcement. This suggests that endogenous cannabinoid activity modulates a broad range of behaviors.
Data from microdialysis studies have revealed that dopaminergic transmission is increased in the nucleus accumbens following acute administration of cannabinoid agonists , , and this effect is blocked by the CB1 antagonist rimonabant (SR 141716A). Although it is possible that cannabinoid-induced dopamine elevations are a result of direct stimulation of dopamine neurons, accumulating evidence suggests a more likely mechanism of action is via disinhibition of dopamine-containing neurons that are under tonic γ-aminobutyric acid (GABA)ergic inhibition. Consistent neurochemical correlates during withdrawal from cannabinoids include reduced dopaminergic activity along the ventral tegmental area-nucleus accumbens pathway and upregulated expression and release of corticotropin-releasing hormone in the central nucleus of the amygdala. This growing body of knowledge, coupled with increasing numbers of individuals seeking treatment for cannabis dependence, has prompted research on the effects of cannabinoid antagonism on cannabis-associated reinforcement and research on the effects of cannabinoid agonists, as well as medications that decrease the stress response, on cannabis withdrawal.
Abstinence Symptoms Treatment Medications
Studies on Laboratory Animals
Prior to the availability of a cannabinoid antagonist, findings from investigations of cannabinoid-related withdrawal symptoms in laboratory animals were inconsistent. Some researchers found evidence of withdrawal symptoms upon abrupt cessation of drug administration, whereas others failed to observe signs of withdrawal when drug administration was terminated. Administration of the cannabinoid antagonist rimonabant, however, produces a reliable withdrawal syndrome in laboratory animals undergoing chronic cannabinoid treatment. Behaviorally, this syndrome is most consistently characterized in rodents by wet-dog shakes, paw tremors, piloerection, and increased grooming.
The fact that cannabinoid-related withdrawal symptoms are reliably produced in laboratory animals not only provided evidence for physiological cannabinoid dependence, but it also provided an opportunity to examine systematically pharmacological agents for effectiveness in attenuating these symptoms. Table 56.1 summarizes selected studies that have employed laboratory animals to evaluate medication effects on precipitated cannabinoid withdrawal symptoms. Although the number of studies conducted evaluating potential cannabinoid treatment medications continues to grow, compared with medications development research for other drugs of abuse, this number is conspicuously low. In one earlier study, Verberne et al. administered intravenous Δ 9 -tetrahydrocannabinol in escalating doses for five consecutive days to rats; on day 6, an acute dose of Δ 9 -tetrahydrocannabinol or placebo was given 30 minutes prior to an intraperitoneal injection of clomipramine, a selective serotonin reuptake inhibitor (SSRI). The investigators reasoned that clomipramine would precipitate withdrawal in animals chronically exposed to Δ 9 -tetrahydrocannabinol because fluoxetine, another SSRI, precipitated withdrawal in animals treated with a similar Δ 9 -tetrahydrocannabinol dosing regimen. Although clomipramine precipitated withdrawal symptoms in rats that received acute placebo, there were significantly fewer withdrawal symptoms observed in rats that received the acute dose of Δ 9 -tetrahydrocannabinol. These findings, together with data from the report showing that fluoxetine also induces behavioral signs of withdrawal in rats chronically administered Δ 9 -tetrahydrocannabinol, suggest that increased serotonergic activity following abrupt discontinuation of repeated cannabinoid agonist treatment may be an important component in the behavioral expression of cannabinoid withdrawal. More recently, however, Touriño et al. reported that the serotonin agonist 3,4-methylenedioxymethamphetamine dose-dependently attenuated rimonabant-precipitated Δ 9 -tetrahydrocannabinol withdrawal symptoms in mice. The reasons for these apparent incongruent findings are unclear, but might be related to the fact that the tested medications have multiple sites of action. Some of these actions may overlap, whereas others may not. It is also important to note that there have been no published reports of clomipramine- or fluoxetine-precipitated cannabis withdrawal in humans, so it is not known whether precipitated withdrawal has not occurred. Thus the impact of increased serotonin activity on cannabinoid-related withdrawal is unclear.
|Investigators||Species||Medication (dose)||Precipitant (dose)||Outcome|
|Verberne et al. 1981||Rat||Δ 9 -THC (6 mg/kg, i.v.)||Clomipramine (15 mg/kg, i.p.)||Δ 9 -THC reduced withdrawal symptoms.|
|Lichtman et al. 2001||Mouse||Δ 9 -THC (30 mg/kg, i.v.)||Rimonabant (10 mg/kg, i.p.)||Δ 9 -THC reversed withdrawal-related paw tremors.|
|Lichtman et al. 2001||Mouse||Clonidine (0.125–1 mg/kg, i.p.)||Rimonabant (10 mg/kg, i.p.)||Clonidine reversed withdrawal-related paw tremors and head shakes.|
|Anggadiredja et al. 2003||Mouse||Prostaglandin E 2 (1, 3.2 μg, i.c.v.)||Rimonabant (10 mg/kg, i.p.)||Prostaglandin E 2 lessened withdrawal symptoms.|
|Cui et al. 2001||Rat||Lithium (4, 8, 16 mEq/kg)||AM281 (3 mg/kg, i.p.)||Lithium blocked withdrawal symptoms.|
Lichtman and colleagues demonstrated that Δ 9 -tetrahydrocannabinol as well as clonidine, an α 2 -receptor agonist, lessened rimonabant-precipitated withdrawal symptoms in mice. In that study, mice were administered two daily subcutaneous injections of either Δ 9 -tetrahydrocannabinol or vehicle for 2 days; on the third day, animals were given one injection of their respective treatment, followed 4 hours later with an intraperitoneal injection of rimonabant or vehicle. Five minutes after the rimonabant challenge dose, mice were administered either an intravenous injection of Δ 9 -tetrahydrocannabinol (or placebo) or an intraperitoneal injection of clonidine (or placebo). Both Δ 9 -tetrahydrocannabinol and cloni dine reversed rimonabant-precipitated paw tremors, and this effect was independent of any generalized effects on movement. Although the finding that Δ 9 -tetrahydrocannabinol reversed precipitated cannabinoid withdrawal is consistent with previous data, these were the first published data to demonstrate that an α 2 -receptor agonist was effective in alleviating symptoms of cannabinoid withdrawal. Clonidine has been shown to attenuate some withdrawal symptoms associated with alcohol and opioid dependence in humans and laboratory animals, suggesting that some features of withdrawal syndromes associated with drugs of abuse may share common underlying pathophysiological mechanisms. For instance, it is possible that withdrawal symptoms, at least in part, may be mediated by noradrenergic hyperactivity. This view is consistent with the observation that many humans experiencing withdrawal from commonly abused drugs, including alcohol, opioids, and cannabis, often report increased anxiety. One exception to this speculation, however, is the efficacy of bupropion in the treatment of nicotine dependence (see below).
Another interesting line of research aimed at understanding mechanisms underlying cannabinoid withdrawal is the examination of the role of the arachidonic acid cascade. Anggadiredja et al. rendered mice Δ 9 -tetrahydrocannabinol-dependent by administering two daily intraperitoneal injections of Δ 9 -tetrahydrocannabinol for 5 days. On the sixth day, mice received one injection of Δ 9 -tetrahydrocannabinol, followed 4 hours later with an intraperitoneal injection of rimonabant to precipitate withdrawal. Mice in an additional treatment group were given an intraventricular injection of prostaglandin E 2 , an end-product of the arachidonic acid cascade, immediately before the rimonabant challenge dose. Prostaglandin E 2 dose-dependently attenuated rimonabant-precipitated withdrawal symptoms including forepaw tremors, forepaw licking, and facial preening. Although the exact mechanism(s) through which prostaglandin E 2 lessened withdrawal symptoms remains to be elucidated, it has been proposed that prostaglandin E 2 reduced symptoms of withdrawal via noradrenergic mechanisms.
Convergent evidence supports this hypothesis. In an in vitro study of the effects of prostaglandin E 2 on electrically evoked tritiated norepinephrine overflow, Exner and Schlicker found that prostaglandin E 2 inhibited the electrically evoked norepinephrine tritium overflow from mouse and rat brain cortex slices. Data from an earlier, similarly designed study were consistent with those obtained by Exner and Schlicker. In addition, as mentioned earlier, clonidine, administered 5 minutes after rimonabant in Δ 9 -tetrahydrocannabinol-dependent mice, reversed the precipitated withdrawal, providing further evidence for the role of noradrenergic processes in cannabinoid withdrawal.
The effects of lithium, a commonly used mood-stabilizing medication for the treatment of bipolar disorder, have also been assessed on cannabinoid withdrawal symptoms. Examination of lithium was based on the clinical observation that increased irritability, anxiety and depression often accompanies cannabis withdrawal; lithium effectively decreases these symptoms. Cui et al. administered two daily injections of HU210, a synthetic cannabinoid agonist, to rats on 5 days; on the sixth day, animals were given one injection of their respective treatment, followed 4 hours later with an injection of AM281, a cannabinoid antagonist. The effects of lithium were examined by administering varying doses 15 minutes before the AM281 challenge dose. Lithium dose-dependently prevented symptoms of cannabinoid withdrawal. The investigators speculated that their findings were mediated via lithium-enhancing effects on central nervous system oxytocin activity and were not related to lithium-associated mood-stabilizing effects. This hypothesis was based on the following observations: (1) oxytocin administration mimicked the effects of lithium on cannabinoid withdrawal; (2) pretreatment with an oxytocin receptor antagonist blocked lithium-related effects on cannabinoid withdrawal; (3) pretreatment with an oxytocin receptor antagonist alone enhanced AM281-precipitated cannabinoid withdrawal ; and (4) divalproex, another mood stabilizer used in the treatment of mania, failed to attenuate AM281-precipitated cannabinoid withdrawal (unpublished observations from the same group of researchers). The fact that animals exhibit a stress-like response (e.g., increased grooming behaviors and increased release of corticotropin-releasing hormone) during cannabinoid withdrawal, however, suggests that the mechanisms involved in the stress response may also play a role in the cannabinoid withdrawal syndrome. Given that increased oxytocinergic transmission markedly diminishes the stress response, it seems plausible that oxytocin plays an integral role in lithium-related effects on cannabinoid withdrawal symptoms. Nevertheless, the exact mechanism responsible for lithium-associated ameliorating effects on cannabinoid withdrawal is an issue that can only be resolved with further research.
One important factor that might limit the generality of the above results is that cannabinoid drugs were administered to animals noncontingently; that is, they were not self-administered but were administered by the experimenter. Data from studies comparing noncontingent and contingent drug administration indicate that substantial differences (e.g., mortality rate and neurochemical) exist that are related to context of drug presentation. Future studies should assess the utility of medications to alleviate cannabinoid withdrawal symptoms in animals undergoing abrupt discontinuation of self-administered cannabinoids.
Despite this potential limitation, the above results suggest that the administration of oral Δ 9 -tetrahydrocannabinol might be a useful strategy to treat cannabis withdrawal. In addition, the data showing that clonidine mitigates cannabinoid withdrawal are encouraging and suggest that pharmacological agents that decrease noradrenergic output are excellent candidate medications to test in humans undergoing cannabis withdrawal. Although side effects, such as hypotension and sedation associated with clonidine may limit its clinical use for cannabis dependence, other α 2 -receptor agonists such as lofexidine, which has a more favorable side effect profile, may hold promise in treating cannabis withdrawal. Indeed, this strategy was investigated in a recent study employing human marijuana abusers (see below). Data indicating that lithium, as well as oxytocin, prevented cannabinoid withdrawal provide potentially novel treatment strategies, although the clinical use of systemic oxytocin for anti-cannabinoid withdrawal effects might be limited because high doses may be required, which increase the likelihood of unpleasant peripheral side effects. Because oxytocin has been shown to produce effects similar to those of benzodiazepines, an alternative approach might be to evaluate the effects of a benzodiazepine on cannabinoid withdrawal symptoms. Clinicians may be wary, however, about the potential for abuse associated with the use of some benzodiazepines, for example, alprazolam, particularly in sedative abusing populations, thus others such as clonazepam or oxazepam may be more likely candidates. Finally, the data regarding the role of serotonergic activity in cannabinoid withdrawal are less clear: findings from two studies indicate that medications that augment serotonin activity may precipitate or worsen withdrawal, whereas results from another suggest that increased serotonin activity dampens withdrawal.
Studies on Human Research Participants
Although the majority of cannabis users may not experience symptoms of withdrawal, data from a variety of human laboratory and clinical studies demonstrate that an abstinence syndrome can be observed following abrupt cessation of heavy, near-daily use of smoked cannabis or oral Δ 9 -tetrahydrocannabinol. Cannabinoid withdrawal syndrome in humans may include a variety of symptoms including increased negative mood states (e.g., increased anxiety, restlessness, depression, and irritability), disrupted sleep, decreased food intake, and in some cases, aggressive behavior. These symptoms have been reported to begin 1 day after cannabinoid cessation, peak effects are observed on days 2–6, and most effects persist from 4 to 14 days, depending on an individual’s level of cannabis dependence. Because cannabis withdrawal may be one factor maintaining continued cannabis use (i.e., frequent marijuana smokers may continue their use not only for marijuana-related intoxicating effects, but also to avoid undergoing withdrawal symptoms), medications that would alleviate cannabis withdrawal symptoms could be useful.
Table 56.2 summarizes the studies that have employed human research participants to evaluate the potential of medications to alleviate marijuana withdrawal symptoms. As can be seen, the majority of the published research in this area has been conducted in our laboratory. Our group at Columbia University/New York State Psychiatric Institute has conducted a series of carefully controlled, within-participant design, residential laboratory studies. During these studies, nontreatment seekers, frequent marijuana smokers smoked active marijuana cigarettes on several consecutive days, five times per day, followed by several days of marijuana abstinence. During abstinence, placebo marijuana cigarettes were smoked and the effectiveness of potential treatment medications to attenuate marijuana withdrawal symptoms was examined. The first medication tested in these studies was bupropion 0, 300 mg/day, a US Food and Drug Administration (FDA)–approved tobacco smoking cessation aid and antidepressant. Three hundred mg of bupropion were tested against a 0-mg control group. The rationale for the use of this medication was related to the observation that bupropion had been shown to maintain tobacco smoking abstinence, in part, because of its ability to decrease negative mood symptoms (e.g., increased anxiety, depression, and irritability) associated with nicotine withdrawal. Given that similar negative mood symptoms are also associated with marijuana withdrawal, bupropion was expected to improve symptoms of marijuana withdrawal. The data, however, indicated otherwise: bupropion worsened several ratings of mood, including irritability, restlessness and depression, and self-reported sleep quality. The mechanism(s) mediating bupropion-worsening effects on marijuana withdrawal is unclear, but the mechanism of action most commonly attributed to bupropion is inhibition of dopamine reuptake and, to a lesser extent, norepinephrine reuptake inhibition. Thus bupropion-associated effects on marijuana withdrawal symptoms could be related to enhanced norepinephrine activity. This suggestion is consistent with the above-cited data showing that clonidine, a medication that decreases noradrenergic activity, lessened precipitated Δ 9 -tetrahydrocannabinol withdrawal symptoms as well as the withdrawal symptoms associated with alcohol and opioid dependence.
|Investigators||Medication (dose, p.o.)||Outcome|
|Haney et al. 2001||Bupropion (0, 300 mg/d)||Bupropion worsened symptoms during withdrawal.|
|Haney et al. 2003||Nefazodone (0, 50 mg/d)||Nefazodone decreased some withdrawal symptoms, but it had no effect on most symptoms.|
|Haney et al. 2004||Divalproex (0, 1500 mg/d)||Divalproex worsened mood and psychomotor performance during marijuana consumption and during marijuana withdrawal.|
|Haney et al. 2004||Δ 9 -THC (0, 50 mg/d)||Δ 9 -THC reduced marijuana withdrawal symptoms and reversed the withdrawal-associated psychomotor performance decrements and weight loss associated with marijuana withdrawal.|
|Budney et al. 2007||Δ 9 -THC (0, 30, 90 mg/d)||Δ 9 -THC dose-dependently attenuated marijuana withdrawal symptoms.|
Another study conducted by our team investigated the effects of nefazodone 0, 450 mg/day on symptoms of marijuana withdrawal . Participants were either administered 450 mg of nefazodone daily in the treatment group or 0 mg in the placebo control group. Nefazodone, an atypical antidepressant, is thought to exert its major therapeutic effects via antagonistic actions at the serotonin-2a receptor, although it has also been shown to produce relatively weak inhibition of norepinephrine and serotonin uptake sites in vitro. A major reason for investigating nefazodone-related effects on marijuana withdrawal symptoms was that it had been demonstrated to effectively treat depression, agitation, and anxiety (symptoms also associated with marijuana withdrawal) in clinical populations. Data from the study by Haney et al. revealed that nefazodone decreased a few symptoms associated with marijuana withdrawal (i.e., ratings of “Anxious” and “Muscle Pain”), but it had no effect on most symptoms (e.g., ratings of “Irritable” and “Trouble Sleeping”). Because nefazodone did relieve some discomfort associated with marijuana withdrawal without worsening other symptoms and because only one active dose was tested, further study of this agent, using a broader dosing range, in the treatment of marijuana withdrawal could be warranted but may not occur because of the black box warning (i.e., the highlighted portion of the package insert) about hepatoxicity.
Divalproex 0, 1500 mg/day, approved for the treatment of epilepsy, mood disorders, and migraine headaches, was evaluated for effectiveness in decreasing marijuana withdrawal symptoms. Study participants received either 1500 mg daily in the treatment group or 0 mg in the placebo control group. Divalproex’s precise neurochemical mechanism of action remains unknown, although some divalproex-related therapeutic effects have been attributed to its ability to dampen sustained repetitive neuronal firing via modulation of Na + channel activity. Other therapeutic effects might be related to its ability to increase central nervous system GABA activity. The rationale for testing the effects of divalproex on marijuana withdrawal symptoms was not based on a proposed neurochemical mechanism of action, but instead was based on clinical evidence indicating that the medication had been used successfully to treat some symptoms commonly associated with marijuana withdrawal (e.g., irritability and mood lability ). Unfortunately, divalproex did not reduce marijuana withdrawal symptoms. In fact, many withdrawal symptoms, including anxiety and irritability, were significantly increased when participants were maintained on divalproex compared to when they were maintained on placebo. Divalproex not only worsened marijuana withdrawal-associated mood, but it also produced psychomotor performance disruptions during marijuana consumption and during marijuana abstinence. The results are in agreement with data from the aforementioned unpublished study using rodents, which showed that divalproex had no effect on AM281-precipitated withdrawal. In short, these data do not support the use of divalproex as a marijuana treatment medication.
Two other groups of researchers have examined lithium carbonate, another mood stabilizer, for effectiveness in decreasing cannabis withdrawal symptoms. The rationale for testing lithium carbonate was based on encouraging data collected using laboratory animals in which the medication decreased cannabinoid-associated withdrawal symptoms. Bowen et al. and Winstock et al. conducted open-label trials of the effects of lithium carbonate (500–900 mg/day) on cannabis withdrawal. In general, the researchers reported that the medication reduced withdrawal severity for most study participants, but both studies contained important limitations that decrease the generality of the findings. For example, the noncontrolled nature of these experiments may have increased expectancy effects; that is, the researchers’ and the participants’ knowledge that participants were receiving an active treatment medication influenced participants reported cannabis withdrawal intensity.
Another strategy tested for efficacy in attenuating human marijuana withdrawal is the administration of oral Δ 9 -tetrahydrocannabinol . In a recently reported study, Haney and coworkers investigated the effects of oral Δ 9 -tetrahydrocannabinol (THC) 0, 10 mg administered five times per day, on marijuana withdrawal symptoms. Participants received either five capsules containing 0 mg of THC in the placebo control group or five capsules of 10 mg of THC each in the treatment group. The primary reason for evaluating the effects of oral Δ 9 -tetrahydrocannabinol on marijuana withdrawal was based on the idea of substituting a longer-acting pharmacologically equivalent drug for the abused substance, stabilizing the individual on that drug, and then gradually withdrawing the substituted drug. In this way, the likelihood of precipitating abstinence symptoms is decreased. Nicotine replacement therapies have been used extensively in this capacity for the treatment of tobacco-related withdrawal, as has methadone for opioid withdrawal. Haney et al. found that oral Δ 9 -tetrahydrocannabinol markedly reduced symptoms associated with marijuana abstinence including self-reported ratings of marijuana craving, anxiety, misery, and sleep disturbance. The medication also reversed the withdrawal-associated psychomotor performance decrements as well as the anorexia and weight loss associated with marijuana withdrawal. It is important to note, too, that these effects occurred at an oral Δ 9 -tetrahydrocannabinol dose indistinguishable from placebo (i.e., like placebo, active Δ 9 -tetrahydrocannabinol produced no apparent subjective effects), highlighting the pharmacological specificity of marijuana withdrawal. Budney and colleagues replicated and extended these findings by demonstrating that oral Δ 9 -tetrahydrocannabinol (30 and 90 mg/day) dose-dependently suppressed cannabis withdrawal in an outpatient environment. Together, these results are consistent with findings that showed that acute Δ 9 -tetrahydrocannabinol administration substantially assuaged precipitated cannabinoid withdrawal in laboratory animals ; more importantly, they indicate that oral Δ 9 -tetrahydrocannabinol might be beneficial in the treatment of marijuana dependence.
Several limitations of the above studies should be noted. First, most of the studies employed only one active dose of the treatment medication. Perhaps more cannabis-related withdrawal symptoms would have been alleviated if a wider range of medication doses were examined. This point is particularly relevant for the study that examined nefazodone because the tested active dose (450 mg/day), which was lower than doses regularly used clinically to treat anxiety and depression, showed a trend toward improved withdrawal symptomatology. Second, most study participants were seeking treatment to abstain from cannabis use. Because the study of cannabis-related effects in humans requires the administration of carefully controlled doses of smoked marijuana, ethical considerations dictate that research volunteers not only have current cannabis use histories, but that they are also not seeking treatment for their cannabis use. , Thus it is possible that the above results may not generalize to persons who are requesting treatment for cannabis dependence. A related limitation is that although adolescents are more likely than adults to exhibit clinical features of cannabis dependence and experience difficulties abstaining from cannabis use, none of the above studies included participants younger than 21 years of age. This was done because the studies involved the administration of smoked marijuana (a drug of abuse); thus it was believed inappropriate to expose children to smoked marijuana in the laboratory, even if the potential participant had reported previous use. Nonetheless, in light of the fact that a large proportion of cannabis-dependent persons under the age of 21 report using cannabis to alleviate withdrawal symptoms, it may be important study the effects of potential treatment medications in older adolescents.
There are at least two issues of potential concern related to treating cannabis-dependent adolescents with medications such as oral Δ 9 -tetrahydrocannabinol: (1) administration of a psychoactive drug to individuals whose brains are still developing can potentially hamper development, especially in areas like the prefrontal cortex, which is slower to develop than other cortical regions ; and (2) replacement of one psychoactive drug with abuse potential with another drug that has abuse potential. Although these concerns deserve serious consideration, it is important to note that the route of administration is a critical determinant of neurochemical consequences associated with drug administration, in part because neurochemical effects depend on the rate of rise of drug concentrations and the maximum drug concentrations achieved. Thus administration of Δ 9 -tetrahydrocannabinol via the oral route would be expected to produce less deleterious neuronal consequences than smoked marijuana. Regarding concerns about the abuse potential of oral Δ 9 -tetrahydrocannabinol, data from a recent study completed in our laboratory showed that the drug produced low rates of self-administration in a sample of marijuana smokers, suggesting that the abuse potential of oral Δ 9 -tetrahydrocannabinol is limited. Note also that oral Δ 9 -tetrahydrocannabinol, unlike smoked marijuana, is not associated with an increased risk of lung toxicity. Hence, from a risk-benefit perspective, oral Δ 9 -tetrahydrocannabinol appears to be a safer therapeutic option. It should be noted that Gray et al. recently assessed oral Δ 9 -tetrahydrocannabinol (0, 2.5, 5, 10 mg/day) for tolerability in older adolescents (ages 16–21 years). They found that the drug produced dose-related increases in euphoria without producing significant effects on cardiovascular measures, psychomotor performance, or negative subjective-effect ratings. Another limitation worth noting is that the same group of researchers has collected most of the published data in this research area, which highlights the need for replication of previous results and additional data.
The above limitations notwithstanding, the data obtained in human research participants demonstrate that while a growing number of medications have been tested, few show promise as potential treatment strategies for the amelioration of cannabinoid withdrawal symptoms. Findings from studies of bupropion and divalproex were discouraging, as these medications failed to assuage many marijuana withdrawal symptoms. In some cases, symptoms were worsened by the medication. Of the agents tested, clearly, oral Δ 9 -tetrahydrocannabinol produced the most promising results. In addition, the limited results obtained in adolescents indicate that oral Δ 9 -tetrahydrocannabinol is well tolerated and suggest further study of this medication in adolescent marijuana abusers. Although no study has investigated the effects of benzodiazepines on human cannabis withdrawal symptoms , data obtained in laboratory animals suggest that future studies should examine the ability of agents such as clonazepam or oxazepam to lessen severity of the withdrawal syndrome.
Randomized Controlled Trials
Table 56.3 shows results from clinical trials that have assessed pharmacotherapies for treating cannabis dependence. As is stated in a recent Cochrane review, the evidence remains inconclusive, but there is moderate evidence indicating that oral THC increases the likelihood that participants will complete the trial. In addition, treatment with preparations containing THC reduces cannabis withdrawal symptoms and craving.