Introduction

Drug addiction is a chronic, relapsing mental disorder characterized by compulsive drug-seeking despite severe negative subsequences [1]. Clinically, drug addiction/substance-use disorder, which is diagnosed by standardized examinations such as the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-V), is distinct with the occasional use of drugs of abuse. To be diagnosed as drug addiction by DSM-V, patients should meet the diagnostic criteria for each drug of abuse, which are defined as mild, moderate, or severe, to indicate the level of severity. In general, drug addiction develops chronically with the following four stages: occasional use, recreational use, regular use, and addiction [2]. Current studies investigating the neural mechanisms of drug addiction has been focusing on compulsive intake when addicted to drugs and relapse after abstinence [1, 3]. Along with the neck-breakingly fast developments of biological research approaches, especially molecular neuroscience, it seems that we have moved into a new era that considerable approaches are promising to be translated into effective clinic tools for treating mental disorders. However, the lost in translation from basic research to clinical treatments is a common challenge in various pathological disorders, including addiction treatments [4].

Pharmacological interventions are the major therapies for treating human diseases. However, similar to other mental disorders, drug addiction is profoundly influenced by emotion, consciousness, and cognition [5]. Besides pharmacological interventions, behavioral strategies are also promising for treating drug addiction. To some extent, non-pharmacological treatments may have greater advantages because of their limited side effects [6]. Here we review potential strategies that we deem most important and promising for treating drug addiction and preventing relapse.

Pharmacological targets

Current pharmacological treatments of drug addiction primarily target the specific receptors that the drugs of abuse act on [7]. For example, the most successful and recommended treatment for opioid-use disorder is opioid agonist maintenance treatment by using the opioid partial agonist buprenorphine or opioid full agonist methadone [8, 9]. In addition, the opioid antagonist naltrexone that displaces opioid agonists has been used as a medicine for the treatment of opioid dependence and relapse for decades [10]. Likewise, nicotine addiction is usually treated by modulating the activity of nicotine receptors. Currently, three major pharmaceutics that were approved by United States Food and Drug Administration to treat nicotine addiction are the nicotine receptor antagonist bupropion, the nicotine receptor partial agonist varenicline [11], and nicotine replacement therapy [12]. However, the current medicines cannot effectively prevent the high rate of relapse after abstinence [12]. Furthermore, current treatments could produce serious side effects such as opioid overdose in opioid agonist maintenance treatment [8]. Effective and safer pharmacological treatments are in dire need to treat drug addiction.

Dopaminergic system

Growing evidence has identified the critical mesolimbic dopaminergic circuit as well as important molecules in this neural pathway that mediate addiction to specific drugs and excessive behaviors [13]. Dopamine, one of the major neurotransmitters in the brain, is believed to be the “culprit” that results in drugs of abuse-induced “high” [14]. As most drugs of abuse eventually activate the dopaminergic system and elevate the dopamine transmission in the brain, modulating the dopamine system could effectively treat most drug addiction, at least in theory [15]. For this reason, many efforts have been made to investigate potential pharmacological targets within the dopaminergic system.

The dopamine system has been discovered for over 60 years [16]. Since then, accumulating evidence has demonstrated a crucial role of the dopamine system in regulating several brain diseases, especially drug addiction, Parkinson’s disorder, schizophrenia, anxiety, and depression [16]. The signal of dopamine is mediated by dopamine receptors and several downstream messengers, such as protein kinase A, protein kinase C, extracellular signal-regulated kinase, Ca2+/calmodulin-dependent kinase II, and DARPP-32 [14, 17, 18]. Among them, DARPP-32, which is enriched in the dopamine-innervated brain areas, could be a promising candidate for treating drug addiction [15]. As the downstream molecules of dopamine receptors also have common roles in mediating other G protein-coupled receptor signals [19], activation of these molecules could result in unwanted side effects. Thus, most of the downstream molecules of dopamine receptors are not appropriate therapeutic targets. Pharmacologically targeting the dopamine receptors for treating drug addiction is of great clinical interest. Dopamine receptors in the brain include two main subfamilies: D1-like receptors (D1 and D5) and D2-like receptors (D2, D3, and D4) [20]. Compared with D1 receptors, D2 receptors are more attractive as a candidate drug target, as D2 receptors are preferentially involved in chronic drug exposure and relapse behavior [20]. Other dopamine receptor subtypes, such as D3 receptors, are also promising candidate drug targets for addiction treatment [7, 21]. Due to the high structural homology between D2 and D3 receptors, compounds that target D3 receptors are usually able to bind D2 receptors. Fortunately, recent studies have developed some D3-selective ligands and these compounds showed potentials as modifiers of drug addiction [22]. It has been shown that GSK598809, a selective D3 receptor antagonist, is efficacious to reduce addiction-related behaviors in preclinical models of relapse [22]. A recent study showed that GSK598809 normalized the reward deficits in drug-dependent subjects [22]. It was further demonstrated that GSK598809 might selectively modulate the neural network underlying reward anticipation. More importantly, GSK598809 is currently in clinical trials for treating drug addiction [7].

Glutamate system

As mentioned above, drug addiction developed after repeated drug exposure is different from occasional use of drugs, which is mainly dependent on the acute effects of drugs [3, 23]. The neuroadaptations induced by chronic drug use involve several important neural systems [23]. Increasing evidence has shown that the glutamatergic system in the ventral striatum/nucleus accumbens (NAc) and the prefrontal cortex (PFC) emerges as the dominant factor that mediates drug-seeking and relapse after abstinence from drugs [24, 25].

Glutamate system in the brain contributes to the adaptive alterations induced by chronic drug exposure [24, 26]. Actions of glutamate are mediated through its receptors, among which the α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptors (AMPARs), the N-methyl-d-aspartate receptors (NMDARs), and the metabotropic glutamate receptors (mGluRs) are the major glutamate receptors that are involved in drug addiction [24, 27, 28]. Activation of glutamate system in the corticostriatal areas partially contributes to the relapse to continued drug-taking and -seeking [23]. Pharmacological modulations of AMPARs and NMDARs could attenuate abuse-related behaviors of drugs. The interested readers are referred to other excellent reviews for more information on this topic [29,30,31,32]. It should be noted that although many compounds that antagonize ionotropic glutamate receptors show efficacy to prevent addictive-like behaviors, most of them are not suitable for clinical use because of their potential side effects. Recently, it was demonstrated that subanesthetic doses of ketamine, a NMDAR antagonist, attenuated cocaine addiction without inducing psychiatric disturbances [33, 34]. Although ketamine has abuse potential per se, these data strongly support that subanestheic ketamine could be used as a novel pharmacotherapy to treat drug addiction. It is especially intriguing that subanesthetic ketamine has been shown to possess rapid antidepressant effects in both preclinical and clinical studies [35,36,37]. Abundant epidemiological data have revealed high comorbidity of drug addiction and major depression [38]; therefore, the above results suggest that subanesthetic ketamine could be especially effective to treat the comorbidity of drug addiction and depression [39]. Besides directly modulating the ionotropic glutamate receptors, recent studies have demonstrated promising effects for the use of allosteric modulators of mGluRs in drug addiction [40]. Preliminary studies showed that positive and negative allosteric modulators were able to reduce various addictive-like behaviors in animal models [40].

Pharmacological modulation of glutamate lactate transporters (GLTs), which are important for extracellular glutamate clearance, is also a promising strategy for treating drug addiction [41]. GLT-1, also referred as solute carrier family 1 member 2 and excitatory amino acid transporter 2, is the predominant transporter attributing to glutamate uptake in the brain [42]. Preclinical and clinical studies investigating the role of GLT-1 in drug addiction has been systematically reviewed recently [41]. In general, most preclinical evidence revealed that reversal of the GLT-1 dysfunction that was induced by drug exposure could effectively suppress drug-seeking behavior [43]. There are several clinical trials that investigated the efficacy of N-acetylcysteine (NAC), a drug regulating GLT-1, for treating several drugs of abuse, such as cocaine, methamphetamine, cannabis, and nicotine [44]. However, the clinical results are mixed. For example, it was reported that NAC reduced self-reported cocaine use in a small trial [45]. However, in a large double-blind placebo-controlled study, it was shown that NAC was effective to prolong the time to relapse but had no effect on self-reported cocaine use [46]. A study by using the proton magnetic resonance spectroscopy showed that NAC reduced the elevated glutamate levels in the dorsal anterior cingulate cortex in cocaine-dependent patients, suggesting that the inhibitory effects of NAC were partially due to its ability to normalize the abnormal glutamate transmission [47]. Recently, it is demonstrated that the NAC derivative NAC-amide, which has better bioavailability than NAC, was also able to attenuate cocaine-seeking behaviors, suggesting a potential use or even greater efficacy of NAC-amide in clinical settings [48, 49]. It should be kept in mind that NAC and NAC-amide are antioxidant and anti-stress molecules [50]. The effects of these compounds on addiction are possibly due to their anti-stress properties, for the stress/anti-reward system also has a critical role in drug addiction [51].

Trace amine system

Trace amines are a group of amines that are structurally similar to classic amino amines such as dopamine [52, 53]. Compared with the extensive accumulation of classic amines in the brain, the concentrations of endogenous trace amines are relatively low (nanomolar) [53]. A family of trace amine-associated receptors (TAARs) have been cloned in 2001 [53]. Among them, TAAR1 is the best-studied receptor of trace amines [54]. Recent studies have revealed that TAAR1 participates in several mental disorders including depression, schizophrenia, and drug addiction [52].

TAAR1 is broadly expressed in the brain with a relative high accumulation in the mesocoticolimbic system [52]. It is demonstrated that modulation of TAAR1 could regulate dopamine transmission, indicating a possible role of TAAR1 in dopamine-related behaviors [54]. Studies from our lab and others have shown that TAAR1 agonists were effective to attenuate the behavioral and biochemical effects of stimulants such as cocaine and amphetamines [52]. Our recent study also demonstrated that activation of TAAR1 was effective to reduce nicotine-induced dopamine release and neural activation in the NAc [55]. Furthermore, TAAR1 agonists attenuated several kinds of nicotine-related behaviors, e.g., nicotine-induced behavioral sensitization, nicotine discrimination, nicotine self-administration, and reinstatement of nicotine-seeking [55]. Furthermore, we demonstrated that the dosages of TAAR1 agonists that effectively inhibited addiction-related behaviors had no effect on general motor activity [55]. These results indicated that modulation of TAAR1 could selectively regulate abuse-related behaviors of several drugs of abuse. Currently, several compounds that modulate the activity of TAAR1 are in clinical trials designed to treat schizophrenia. Nevertheless, although TAAR1 is a promising target for addiction treatment, the function and the neural mechanism of TAAR1 are still poorly understood [52]. Future studies are needed to determine the exact neural mechanisms of TAAR1 in regulating drug addiction.

Stress/anti-reward system

Besides the positive reinforcing effects of drug use, e.g., euphoria or “high,” negative reinforcing effects associated with abstinence are also well-known factors that contribute to the development of drug addiction [2, 56]. The negative reinforcement of drugs recruits a brain system that is opposed to the rewarding system, termed the anti-reward/stress system in the brain [2, 57]. It is believed that the neuroanatomical substrate of the anti-reward system is the extended amygdala that includes the bed nucleus of the stria terminalis, the central amygdala, and the NAc shell [2, 51]. These brain regions were demonstrated to participate in both the physical withdrawal symptoms and negative emotional states that drive compulsive drug use, drug-seeking, and relapse [51, 58]. As mentioned above, the comorbidity of addiction and other mental disorders are common in patients with substance-use disorder. Drug use- and withdrawal-induced emotional alterations may interact with or facilitate the occurrence of other mental disorders, which in turn promote drug use and relapse [59, 60].

Several important molecular targets in the anti-reward system have been studied [3]. Among them, corticotropin-releasing factor (CRF) is one of the best-studied molecules [3]. A very large number of studies have verified the role of CRF in stress-related behaviors and drug-seeking behavior associated with major drugs of abuse. Preclinical studies demonstrated that CRF peptide and receptor antagonists prevented anxiety-like responses and aversive-like motivational effects induced by withdrawal from several drugs of abuse, as well as attenuated the increased self-administration of drugs in the rat extended-access model of drug intake [51]. Recently, a clinical study investigated the effects of CRF1 antagonist verucerfont on alcohol addiction [61]. Unfortunately, although verucerfont was able to prevent the hypothalamic-pituitary-adrenal axis response, it did not affect stress-induced alcohol craving [61]. Nevertheless, it is too early to conclude that targeting the CRF system to treat drug addiction is deemed to fail, given the very strong preclinical evidence to suggest otherwise [51]. More carefully designed clinical trials are required to test the translational possibility of other CRF antagonists from animals to humans.

Immune system

Recent evidence has revealed a critical role of the neuroimmune system in psychiatric disorders including drug addiction [62, 63]. The central immune system is composed of multiple types of cells, cytokines, and intracellular signaling pathways in the brain [64]. Some studies demonstrated that drugs of abuse, such as opioids, cocaine, methamphetamine, and ethanol, can activate central immune signaling, which in turn enhances the reinforcing effects of drugs [65]. However, the role of immune system in drug addiction remains unclear and other studies also reported no effect of immune function modulators on addition-related behaviors [62]. Furthermore, although a growing number of immune-related targets, such as tumor necrosis factor-α, interleukin-1β (IL-1β), and IL-6, have been demonstrated to have important roles in regulating drug addiction, only a few of them are pharmacologically significant [62]. Here we will focus on one immune-related target, the Toll-like receptor 4 (TLR4), which we think is a promising pharmacological candidate for treating drug addiction [66].

TLR4 is a member of TLR family, which has a key role in the innate immune system [66]. Compared with other drugs of abuse, opioids are the best-studied drugs that could be regulated by TLR4. Opioids could directly activate the TLR4 to produce a neuroinflammatory response [67]. A growing number of studies revealed that inhibition of TLR4 suppressed morphine tolerance and the rewarding effects of morphine, and enhanced the analgesic properties of morphine [68, 69]. A recent study showed that chronic but not acute delivery of the TLR4 antagonist ( + )-naltrexone reduced the incubation of cue-induced heroin-seeking [70]. It was demonstrated that TLR4 is also involved in cocaine addiction. A study showed that TLR4 in the NAc mediates the reinforcing properties of cocaine [71]. In addition, activation of TLR4 in the ventral tegmental area (VTA) was sufficient to reinstate distinguished cocaine-seeking behavior [72]. Another study also reported that TLR4 mediated methamphetamine-induced neuroinflammatory response in astrocytes [73]. However, discrepant results were also reported. In a study, the dosages of ( + )-naltrexone that significantly inhibited remifentanil and cocaine self-administration produced an inhibitory effect on the rate of food-maintained responding, indicating a nonspecific inhibition of general activity rather than a specific anti-reward effect [74]. Moreover, ( + )-naltrexone had no effect on heroin self-administration and the incubation of cue-induced methamphetamine-seeking [70]. Taken together, the role of TLR4 in drug addiction remains somewhat elusive, which may be dependent on the types of drugs, treatment regimens, and experimental conditions.

Currently, there is no clinical evidence that modulation of central immune system is effective for treating drug addiction. As immune response is probably the most important biological reactions of the human body in responding to environmental and innate stimuli, it is a challenge to selectively inhibit the hyperactivity of central immune system induced by repeated drug exposure, while leaving the peripheral immune system intact.

Brain stimulation

Considerable evidence has demonstrated that modulating the activities of addiction-related brain areas is a promising strategy to control drug use and relapse [75]. Although there are various ways to modulate the activity of a specific brain area, such as random noise stimulation, ultrasound stimulation, electoral stimulation, and magnetic stimulation, recent research has been focusing on the latter two [76, 77]. Transcranial magnetic stimulation (TMS) and transcranial direct current stimulation (tDCS) are two of the best-studied approaches for treating drug addiction and other mental disorders [76,77,78,79]. Studies have shown that stimulating dorsal lateral PFC by TMS and tDCS were effective to reduce craving, improve cognition, and ameliorate the mood states in subjects using major drugs of abuse [76, 80]. Importantly, TMS and tDCS showed greater advantages than other invasive strategies including deep brain stimulation (DBS) that requires invasive electrodes to be implanted into the target brain areas [80]. It should be noted that, although TMS and tDCS are relatively safer than invasive stimulations, several disadvantages limit their application in the clinic. For example, (1) these techniques could inaccurately stimulate other nearby brain areas; (2) they cannot reach deeper brain areas that could be easily achieved by DBS; (3) an inaccurate manipulation could induce unwanted effects such as seizures [81]. The detailed technological issues of these approaches are beyond the scope of this review and interested readers can find more information elsewhere [81, 82].

Recent preclinical studies have developed and utilized virus-mediated strategies to modulate the activities of brain areas [83]. Extensive evidence has demonstrated that chemogenetic and optogenetic modulations of reward and anti-reward systems, including the NAc, the PFC, the VTA, and the amygdala, were able to reduce drug-taking and -seeking [84]. These techniques by and large reconfirmed previous conclusions that were achieved by pharmacological interventions, as well as provided novel insights that may shed light on new neural mechanisms [84]. Chemogenetic and optogenetic approaches could easily target a specific circuit rather than just a specific brain area. Stimulation of a specific area could generally alter the activities of its projecting areas, which could induce unwanted side effects [85]. For example, stimulation of the PFC could affect activities of the NAc, hippocampus, and amygdala [86]. Because of the important roles of the hippocampus and amygdala in regulating memories and emotions, stimulating the PFC would also alter subjects’ emotional states and cause amnesia. Current approaches are impossible to specifically modulate a neural circuit in humans. Nevertheless, results from animal studies could provide us with clues of novel neural mechanisms and possible side effects induced by brain stimulation in humans, which in turn help us to adjust and optimize stimulation parameters. For example, a recent study showed that mimicking optogenetic stimulation-induced normalization of synaptic transmission in the NAc by the combination of DBS and pharmacological intervention persistently reduced cocaine-induced sensitization [87]. The study proved the validity of combining brain stimulation with other treatments such as drugs and cognitive behavioral treatments to treat drug addiction.

Accordingly, more studies are required by using optogenetic and chemogenetic approaches to identify novel brain areas that participate in drug addiction. It should be noted that neural circuit-related studies in animals are probably not translable to humans due to the great heterogeneity of brain anatomy between rodents and humans [88]. In addition, current techniques only allow us to stimulate one brain region at one time, which would limit the efficacy of brain stimulation. As mentioned above, a broad range of brain areas synergistically contribute to drug addiction. If we could develop novel techniques that allow us to simultaneously stimulate addiction-related brain areas, stronger and long-lasting effects could be expected.

Behavioral treatment

Behavioral approaches, such as cognitive behavioral therapy (CBT) and cue exposure therapy (CET) are effective for treating drug addiction [89, 90]. Drug addiction is a unique mental disorder in humans. Executive functions, especially the higher-order executive functions such as inhibitory control and cognitive flexibility, have a fundamental role in coping with the mood disturbances and incubated craving associated with protracted withdrawal from drugs of abuse [91]. The potential of CBT for treating drug addiction is based on the hypothesis that reversal of the maladaptive information processing and abrupt beliefs associated with drug addiction is able to reshape the addiction-related behaviors. CBT has been shown to be effective in reducing drug craving in clinical settings when applied alone or with other pharmacotherapy [89]. Furthermore, CBT is widely applicable across many psychiatry fields and has been implemented in a broad range of psychiatric disorders [92]. However, implementation of CBT for treating drug addiction is currently uncommon. This could be due to several disadvantages of CBT, including relatively high cost and requirement of physician specialists to apply the therapy [93]. Furthermore, although CBT shows a time-limit advantage compared with traditional talk therapy, additional treatment cycles are required to produce persistent therapeutic efficacy, otherwise patients would generally relapse within months. The computer-assisted CBT may be an alternative option, which has revealed significant advantages compared with the traditional CBT [94]. For example, computer-assisted CBT is low cost and in no need of clinicians with special skills. The acceptance and accessibility of computer-assisted CBT are growing due to the increasing access to computer and internet. Studies have shown that a six-module computer-based training in CBT (CBT4CBT) was an effective adjunct to standard drug addiction treatments [94, 95].

CET is another behavioral strategy that was widely accepted within drug addiction field and throughout psychiatry [90]. The premise of CET is that Pavlovian conditioned learning contributes to cue-induced craving and drug-seeking, and repeated exposure to drug-associated cues (contextual and discrete cues), which is called extinction, will eventually lead to a decrease in conditioned responses, e.g., cue-induced drug-seeking and craving. CET is based on the hypothesis that drug addiction is an abrupt memory that usurps the normal learning and memory system in the brain [6]. However, in contrast to the positive reports in other psychiatric disorders (e.g., anxiety disorders), studies have not shown a positive efficacy of CET for treating drug addiction in clinical settings. Another psychological process reconsolidation, a state that is triggered by memory retrieval and reactivation, is usually mentioned and discussed with extinction in the literature [96]. Modulation of reconsolidation and extinction of drug reward memory with either pharmacological or behavioral strategies have been demonstrated to be effective for treating addiction. We have reviewed the advances in this topic in a recent review [6]. It should be noted that the lack of efficacy of CET in humans may imply that targeting the cue-associated memories probably only has limited effects on addiction-related behaviors [90]. This could be due to that interventions on drug reward memory would effectively eliminate the cue-induced responses, while leaving the drug-induced maladaptive plasticity alterations intact. In addition, craving induced by drug-associated cues may not be the fuse triggering relapse. With this in mind, future research is warranted to address whether the modulation of drug reward memory could reverse the maladaptive behavioral and neural elasticities, and then develop novel memory-related strategies for treating addiction.

Therapeutic gene modulation

The long-lasting neuroplasticity changes induced by exposure to drugs of abuse are mediated by specific gene transcriptions [97]. Genetic analysis and functional studies have identified a wide range of candidate genes and their variants that regulate drug addiction, e.g., dopamine-related genes, glutamate-related genes, and transcription factors (e.g., the immediate early gene fosB) [97,98,99]. It has been shown that the development of addiction is positively associated with genetic factors, suggesting a high heritability feature of addiction [100]. Drug addiction is not heritable across generations, but personality traits that are associated with initiation of drug use are heritable [100]. Furthermore, epigenetic alterations and epigenetic regulators, e.g., chromatin-remodeling enzymes, histone acetyltransferases, and methyltransferases, also have critical roles in mediating the long-lasting effects of drug use [101]. Recent studies also found that microRNAs and other non-coding RNAs are also important factors that mediate rewarding properties of drugs of abuse, suggesting that modulation of posttranscriptional RNAs may be a possible pharmacotherapy to reverse drug-induced neuroplasticity in the brain [102, 103]. Taken together, current evidence indicates that therapeutic gene modulation altering the addiction-related genes and/or their expression patterns would be a promising therapy for treating drug addiction.

Gene therapy is usually achieved by delivering a variety of types of plasmids and vectors carrying therapeutic genes into human tissues. A recently developed technology known as CRISPR-Cas9 has been shown to be able to effectively alter genes in vivo [104]. Importantly, studies have revealed that CRISPR-Cas9 can highly selectively alter the target gene and do not cause off-target effects. Recently, a study published last year showing “unexpected mutations after CRISPR-Cas9 editing in vivo” has been retracted because of serious flaws of the experimental design [105]. Intriguingly, clinical trials are about to begin to test the validity of CRISPR-Cas9 in treating human diseases. Although current gene therapy is only used to treat cancer, genetic diseases, and infectious diseases [106], preclinical studies by using virus-mediated expression or knockdown of addiction-related genes have been revealed to be effective in altering drug-induced neuroplasticity and maladaptive behaviors [107]. For example, studies showed that helper-dependent adenoviral vector or adeno-associated virus vector-mediated expression of a butyrylcholinesterase-based cocaine hydrolase, which has great efficacy in metabolizing cocaine, was effective to block cocaine’s behavioral effects and relapse in animal models [108, 109]. We believe that it is about time to pay more serious attention to gene therapy in treating drug addiction. More importantly, therapeutic gene modulation might be possible to achieve true cure for drug addiction, because the results of gene therapy are usually long-lasting if not permanent.

Conclusion

Drug addiction is a complicated mental disorder, because multiple levels of biological and psychological systems are recruited into its development. Studies have demonstrated that the reward system (e.g., the mesocorticolimbic system) and the anti-reward system (e.g., the extended amygdala) synergistically contribute to relapse after protracted withdrawal. Besides the current treatments that target specific receptors, pharmacological approaches that target the molecular candidates in the reward and anti-reward systems is promising to treat drug addiction. The neuroimmune system is a relatively novel target that is of significant clinical interest. However, the role of neuroimmune system in drug addiction remains poorly understudied. Other strategies, such as brain stimulation and behavioral treatments, may have greater advantages than pharmacotherapies due to their safer profiles. Overall, the clinical efficacy of current treatments of drug addiction is limited and relapse remains a major clinical challenge. In regards to achieving long-term treatment success and possibly permanent cure, gene therapy could be a powerful strategy, despite that this approach remains in its infancy in the treatment of drug addiction.