Medications for substance use disorders (SUD): Emerging approaches

Abstract

Introduction:

Substance use disorders are a group of chronic relapsing disorders of the brain, which have massive public health and societal impact. In some disorders (e.g., heroin / prescription opioid addictions) approved medications have a major long-term benefit. For other substances (e.g., cocaine, amphetamines and cannabis) there are no approved medications, and for alcohol there are approved treatments, which are not in wide usage. Approved treatments for tobacco use disorders are available, and novel medications are also under study.

Areas covered:

Medication-based approaches which are in advanced preclinical stages, or which have reached proof-of concept clinical laboratory studies, as well as clinical trials.

Expert opinion:

Current challenges involve optimizing translation between preclinical and clinical development, and between clinical laboratory studies to therapeutic clinical trials. Comorbidities including depression or anxiety are challenges for study design and analysis. Improved pharmacogenomics, biomarker and phenotyping approaches are areas of interest. Pharmacological mechanisms currently under investigation include modulation of glutamatergic, GABA, vasopressin and κ-receptor function, as well as inhibition of monoamine re-uptake. Other factors that affect potential market size for emerging medications include stigma, availability of treatment settings, adoption by clinicians, and the prevalence of persons with SUD who are not actively treatment-seeking.

1. Background:

Substance use disorders (SUD) are usually chronic diseases, due to excessive and uncontrolled intake of licit or illicit psychoactive substances. The main licit drugs that result in substantial SUD are alcohol and tobacco (in many countries). The other major drugs of abuse that will be discussed here are heroin and other μ-opioid receptor agonists, cocaine and other stimulants, and cannabis (which is in legislative flux across jurisdictions) [1]. This review will focus on advanced preclinical studies, on the medication pipeline, and on planned clinical trials, based on published grants and clinical trial registries. The reader may refer to several other articles for more extensive historical background on medications for SUD, and basic methodologies used during development [2–6]. Several major trends are currently observed in this area. Firstly, an increasing amount of prescriptions for certain medications for opioid use disorders (e.g., buprenorphine-naloxone), have been observed, versus stable use of other effective medications, such as methadone [7]. Secondly, medications for alcohol use disorders are used only to a limited extent [8,9]. Thirdly, expenditures for SUD pharmacotherapies remains low, compared to overall expenditures in the care of these disorders [7].

2. Medical need:

The ongoing societal and public health costs of SUD clearly show that the medical need (including behavioral and psychiatric health) for these diseases is large and is increasing [1,10,11]. Diverse societal issues (including stigma) [1] can complicate and limit availability of medications for substance use disorders, even when these are approved. Thus, whereas there are several approved medications for specific conditions (see below), their availability and usage and remains uneven.

As mentioned above, SUD tend to be chronic relapsing disorders, and different stages in the disease trajectory for each drug of abuse (e.g., escalation, withdrawal/abstinence and relapse prevention) may need to differentially managed with specific pharmacotherapeutic approaches. There are few available treatments that are demonstrated to be disease-modifying; this could likely be due to the robust, complex and long-lasting neurobiological changes that occur after chronic drug exposure. These long-lasting changes are also manifested as changes in neurocognitive function, and in mood and anxiety.

2.1. Epidemiological estimates of persons with SUD:

The WHO Global Burden of Disease project estimated that there were approximately 125 million persons worldwide with Alcohol Use Disorders (including approximately 24 million in the Americas, and approximately 27 million in Europe) [12]. The United Nations World Drug Report for 2016 estimates that 29 million people worldwide have an SUD (excluding alcohol and tobacco) [1]. Based on the U.S. National Survey on Drug Use and Health (NSDUH), it is estimated that in 2015, 15.7 million people in the U.S. had an alcohol use disorder, and 7.7 million people had an illicit drug use disorder (i.e., excluding alcohol and tobacco) [13]. Thus, NSDUH data show that among Americans 12 years old and older, approximately 2.6 million people had an opioid use disorder (i.e., heroin or illicitly used prescription opioids), 4 million had a cannabis use disorder, 0.9 million had a cocaine use disorder, and 0.9 million had a methamphetamine use disorder [13].

2.2. Estimates of persons in SUD treatment:

Of relevance to the size of the potential market for SUD pharmacotherapy, The United Nations World Drug Report for 2016 estimates that only approximately 17% of persons with an SUD (excluding cases with alcohol or tobacco disorders) were receiving treatment [1]. Also, the U.S. NSDUH data show that only approximately 10% of persons needing treatment for any SUD were receiving it in 2015 [13]. Furthermore, insurance expenditures for SUD in major industrialized countries (e.g., U.S.) have remained relatively static from 2004–2013 [7]. Overall, the limits on available SUD pharmacotherapy also have a major impact on costs related to comorbid infectious diseases (e.g., HIV and Hepatitis C) [1].

Taken together, the above data indicate that the potential market for specific SUD is large, especially in view of their chronic relapsing nature. However, a major expansion in use of novel pharmacotherapies would have to address considerable systemic factors, including availability of treatment providers, adoption of the pharmacotherapeutic modality by clinicians, and reimbursement for long-term medication use [7,14]. A further major factor that affects actual market size is that most persons with SUD do not seek treatment in a timely manner [15]. This may be due to several factors, including lack of recognition of the SUD by the person, as well as other obstacles, including professional or personal stigma, and perceived availability of treatment [7,16–19]. The enhancement of medical education on the management of SUD (both in medical schools and in continuing education efforts) is an important current goal [10,20,21].

3. Commonalities among addictive diseases and downstream mechanisms:

Addictive diseases have some common behavioral signs (e.g., escalation of drug use, use despite health consequences, difficulty in reducing use, relapse). Some downstream effects of several major abused drugs are similar. For example, major drugs of abuse mentioned above, when acutely administered, tend to cause an increase in synaptic dopamine levels in terminal fields of the nigro-striatal and meso-limbic pathways [22,23] [24]. It is also known that other downstream counter-regulatory mechanisms are upregulated after chronic exposure to certain drugs. For example, activity in the κ-receptor / dynorphin system is upregulated after exposure to drugs including cocaine, μ-agonists, or alcohol [22,23,25]. It is known that activation of the κ-receptor / dynorphin system may increase vulnerability to relapse-like behaviors, and to depressant-like behaviors [26,27]. Also, while μ-receptors are the primary target of abused opioids (e.g., heroin), μ-receptors are also important in the downstream effects of other drugs of abuse, including alcohol and cocaine [23,28].

3.1. Poly-drug exposure:

Specific types of poly-drug exposure (e.g., smoking/cocaine/alcohol, or cocaine/heroin) are common in persons with substance use disorders, and particular medication strategies have been investigated [29,30]. There are no medications currently approved for persons who have poly-drug diagnoses, although this group may be at special risk [31,32].

3.2. Comorbid psychiatric diseases:

Psychiatric disorders (especially depression and anxiety) are often comorbid with persons with substance use disorders [33–35]. A number of studies have investigated whether approved psychiatric medications are effective in persons with dual mood and substance use disorders [34]. Treatment of these comorbid psychiatric disorders is important for quality of life and treatment engagement, but may not necessarily have a major impact on SUD per se [36]. Further pharmacogenomic or phenotypic refinements may determine if a subset of these dually diagnosed persons would have a more global benefit from anti-depressant or anxiolytic medication.

3.3. Pharmacogenomics:

Modern approaches to addiction pharmacotherapy also include pharmacogenomic analyses. Several genetic variants that may affect the effectiveness of proposed and approved pharmacotherapeutic agents. For example, genetic variants in pharmacokinetic gene targets affect the clinical profile of methadone [37,38]. Other studies have shown that polymorphisms in a pharmacodynamic target of naltrexone (OPRM1, the gene encoding μ-receptors) affect the clinical effectiveness of naltrexone in alcohol use disorders, as well as attributable risk of developing these disorders [39,40]. A polymorphism in the dopamine-beta-hydroxylase gene has also been associated with clinical response to disulfiram, in persons with cocaine use disorders [41]. Polymorphisms in the serotonin transporter gene have also been associated with differential clinical response to the 5HT3 antagonist ondansetron, in the treatment of alcohol use disorders [42,43]. Given the ongoing dissemination of genotyping technology outside research settings, pharmacogenetic approaches may eventually become established as clinical tools for personalized treatment of specific addictive diseases.

3.4. Societal and financial implications of addictions:

The costs associated with substance use disorders worldwide are massive, especially when considering co-morbid infectious diseases (HIV and HCV) and other societal costs [1]. These costs are estimated to reach approximately $700 billion annually in the U.S. [44].

4. Scientific rationale for the main approaches considered:

Specific effects of major types of drugs of abuse:

There are major facets of the neurobiology, etiology, course and prognosis that differ across the aforementioned types of drugs of abuse. The abuse potential and neurobiological effects of specific drugs in the same category (e.g., μ-agonists or dopamine re-uptake inhibitors) may also differ according to their particular pharmacodynamic and pharmacokinetic properties. Therefore, pharmacotherapeutic approaches to each of these drugs have to be targeted to the individual agent, and in some cases to particular stages in the trajectory of the addictive disease (e.g., during active use, or in relapse prevention). Addictive diseases can also have different trajectories across the lifespan. It is also known that repeated drug exposure per se results in long-lasting neurobiological changes that can increase vulnerability to relapse or to comorbid psychiatric diseases (e.g., depression or anxiety).

4.1. Alcohol:

The primary pharmacological target of alcohol remains unknown, but alcohol has downstream effects on GABA, glutamatergic, μ-receptor and dopamine pathways [45]. Severe alcohol use disorder is characterized by tolerance (requiring increased doses to achieve a given effect) and withdrawal (emergence of signs and symptoms upon drug discontinuation). Depending on disease trajectory, persons may drink in “binges”, or in a regular daily pattern. Alcohol withdrawal is highly aversive, and causes anxiety-like symptoms, tremors and irritability, as well as potentially life-threatening consequences including seizures, and needs to be medically managed. There are major chronic health consequences in patients suffering from alcohol use disorder [46].

4.2. Tobacco / Nicotine:

Nicotine from tobacco (now also delivered by E-cigarettes) is an agonist at nicotinic acetyl-choline receptors. Nicotinic receptors are ionotropic (primarily permeable to Na+), and are made of 5 subunits which can be identical or different (these are termed α and β subunits, of [47]. The particular pentameric structure of nicotinic receptors differs by site and also by function. Therefore novel medications have aimed to target subtypes of nicotinic receptors based on their particular subunit structure. Tobacco-related products are a major cause of preventable deaths [48].

4.3. Cannabis and CB1-r agonists:

The CB1-r partial agonist Δ9-tetra-hydrocannabinol (Δ9-THC) is the main active product from Cannabis sativa [49]. Preparations containing Δ9-THC are the focus of much non-medical usage, and cannabis use disorders are relatively frequent [1,50]. CB1-r, which are also GPCRs, are found in diverse brain areas that mediate motivation, reward and cognition [51,52]. Several synthetic CB1-r agonists have also become the focus of abuse in recent years, and these have higher pharmacodynamic efficacy at CB1-r, compared to Δ9-THC [49]. Another product from the C. sativa plant is cannabidiol, which is not considered to have abuse potential per se [52]. The site(s) of action of cannabidiol are not fully understood, and there has been considerable progress in its therapeutic use for specific indications (including epilepsy and multiple sclerosis). Unfavorable health consequences of substantial cannabis exposure (presumably caused primarily Δ9-THC, acting at CB1-r), include mood, anxiety and cognitive changes, as well as sleep disruptions [53,54]. Substantial exposure to cannabis in adolescence may be associated with particular health concerns over the lifespan [55].

4.4. Heroin and other abused μ -agonists:

Abused opioids are primarily short acting μ-agonists. The main illicit μ-agonist is heroin, and there is also considerable abuse of “prescription opioid analgesics” (such as morphine, oxycodone and congeners, and fentanyl and congeners). Recent data also suggest that fentanyl and other potent synthetic analogs are entering the illicit supply, causing a substantial risk of overdose [56]. μ-receptors are also GPCR, coupled primarily to G_i and other second messengers. Chronic exposure to μ-agonists results in a classic syndrome of dependence and withdrawal, upon discontinuation [57]. The most severe withdrawal signs resolve within 1 week of drug discontinuation, but prolonged neuroplasticity and behavioral changes have also been studied in recent preclinical models [58,59]. Avoidance of withdrawal becomes an important “negative reinforcer” in persons with severe opioid use disorders. Persons with an opioid dependence diagnosis (DSM IV), or severe opioid use disorder (DSM5) may self-administer drugs in a relatively regular pattern during the day, and across days.

Acute toxicity due to μ-agonists is primarily due to overdose, a result of respiratory depressant effects mediated by μ-receptors in the brainstem [60,61]. Chronic opioid use disorders are also associated with other health concerns, including comorbid infectious diseases (e.g., HIV and HCV), especially in i.v. drug users [1]. Heroin is often injected intravenously, but prescription opioids (e.g., oxycodone) are also abused by the oral route. However, it is known that illicit tampering of prescription opioid formulations does occur, with the goal use by other routes. This latter risk pattern has given rise to development of several “abuse-deterrent” formulations of prescription opioids for analgesia [62].

4.5. Cocaine and abused stimulants / amphetamines:

The main abused stimulants are cocaine and several amphetamine analogs, including d-amphetamine and methamphetamine. The main direct effect of these stimulants is an increase in synaptic levels of dopamine levels in CNS, due to either inhibition of the dopamine re-uptake transporter (DAT), or by enhancing dopamine release (cocaine is not highly selective for DAT, and can also target SERT and NET). Other structurally related agents (such as methylenedioxymethamphetamine, MDMA “Ecstasy”) are also the focus of illicit use. Major medications used in the treatment of ADHD are dopamine re-uptake inhibitors or psychostimulants (e.g., methyphenidate and mixed amphetamine salts), which also have abuse potential [32,63]. Appropriate medical use of these compounds may be effective in the management of specific stimulant use disorders [4,64].

5. Existing treatments:

We summarize medications that are approved for pharmacotherapy of specific substance use disorders.

5.1. Alcohol:

The opioid antagonists naltrexone and nalmefene are approved for persons diagnosed with alcohol use disorders. Naltrexone is approved as a daily oral formulation, and also as a monthly depot injection [65]. Oral nalmefene is approved on an “as needed” basis for this indication, in several European countries, but not in the United States at this time [66,67]. Both naltrexone and nalmefene have μ-antagonist effects, and at sufficient doses, also exert low efficacy actions at κ receptors (e.g., κ-antagonist or partial agonist effects)[68–70]. Some pharmacogenomic studies show that the A118G SNP in the μ-receptor gene (OPRM1) is associated with increased pharmacotherapeutic effectiveness of naltrexone [39,40,71]. Acamprosate (calcium acetyl homotaurinate) is approved for maintenance of abstinence from alcohol in several countries. The mechanism of action of acamprosate is not well understood, and may involve GABA or glutamate neurotransmitter systems [72]. Disulfiram (an inhibitor of aldehyde dehydrogenase and dopamine-β-hydroxylase (DBH) has been long-approved for this indication, and causes aversive effects (e.g., nausea) if a person taking disulfiram ingests alcohol. The overall usage of the aforementioned medications, has remained relatively modest, for a variety of medical, socio-cultural and reimbursement factors [72,73]. Novel pharmacotherapeutic approaches (see Table), or further patient stratification strategies (including pharmacogenomics and phenotyping strategies) may therefore be of value.

Table 1:

Competitive environment

Compound	Company	Structure	Indication	Stage of Devp.	Mechanism of Action
Unpublished	Opiant	Unpublished	cocaine	Unpublished	Opioid antagonist
CERC-501 (previously LY2456302)	Cerecor / Janssen	Aminobenzyl-oxyarylamide	smoking; Alcohol; other	Ph II	κ-antagonist
ALKS 6428	Alkermes	Unpublished	Opioid taper kit, prior to depot naltrexone	Ph III	Opioid ligands
Vigabatrin	Catalyst	GABA analog	Cocaine	Clinical; POC	GABA
AD04	Adial pharma	ondansetron	alcohol	Ph II	5HT3 antagonist
EMB-001	Embera	Metyrapone / Oxazepam	Smoking / other	Ph I	HPA and BZ
Cannabidiol;	Insys; GW Pharmaceuticals	Cannabinoid ligands	Amphetamine; Heroin	Phase II;	Several possible mechanisms
ABT-436	NIAAA, Abbvie	ABT-436; not reported	Alcohol	Ph. II	vasopressin V1B-antagonist
NS2359	Saniona	Azabicyclo octane	cocaine	Ph II	Triple MA reuptake inhibitor
Desvenlafaxine	Pfizer	Desvenlafaxine	Opioid	Ph II	SNRI
Ibudilast	Medicinova	Ibudilast	opioid use disorder	Ph II	glial cell inhibition, other

5.2. Tobacco:

Nicotine patches, lozenges and gum (and other formulations) are approved for tobacco use disorder (i.e., with an “agonist-based” approach). These are often used for extended periods (although the knowledge base for extended use is limited) [74], and can decrease the probability of relapse. The other main medications approved for this disorder are bupropion and varenicline, which is a partial agonist at the α4β2 nicotinic receptor subtype. There is little information on the pharmacotherapeutic management of abuse of “e-cigarettes” (which typically vaporize nicotine solutions). There is little evidence in support of the use of e-cigarettes as aids to smoking cessation, [75,76], whereas there is emerging data showing that adolescent use of e-cigarettes increases the probability of cigarette smoking behaviors [77].

5.3. Cannabis:

There are no medications approved for the management of cannabis use disorders. Some CB1-r agonist medications have been approved for other indications [78], and have been examined for their potential role in modulating SUD [79,80]. Cannabidiol is another major product from the Cannabis plant, and it is not thought to have abuse potential per se [81]. Its mechanisms of action are unclear, and cannabidiol is under study for its potential as a pharmacotherapy for SUD [81,82].

5.4. Heroin and abused μ -agonists:

There are two main medications approved for the medical management of severe opioid use disorders. These are methadone or buprenorphine / naloxone, taken on a daily basis. Methadone is a selective and long-lasting μ-agonist (although it also functions as an NMDA antagonist), whereas buprenorphine is a μ-partial agonist that also has κ-antagonist or partial agonist effects [23,83,84]. The antagonist naloxone is added to this buprenorphine formulation, as an abuse deterrent (i.e., to avoid diversion and injection). There are extensive long-term data showing considerable therapeutic benefit of chronic methadone and buprenorphine / naloxone maintenance, when used in high quality programs, providing sufficient doses of these medications [84,85].

The short-acting antagonist naloxone has been approved for decades as a parenteral injection, for the acute reversal of μ-agonist overdose. An intranasal formulation of naloxone was recently approved for this purpose, and may be especially helpful to laypersons and first responders [86]. α₂ adrenergic agonists such as clonidine (or lofexidine, in some countries) are used primarily to manage short-term symptoms of sympathetic over-activation that occur during withdrawal from chronic μ-agonists [87,88]. More extended use of clonidine in lapse and relapse prevention has also been investigated [89].

More recently, a monthly depot-formulation of naltrexone has been approved in the U.S., to prevent relapse in persons diagnosed with opioid dependence who have undergone detoxification [90–93]. Undergoing opioid detoxification and induction onto the depot naltrexone formulation is a challenge that can limit the applicability of such antagonist-based approaches, as many patients have difficulty in this transition [88]. Optimized use of depot naltrexone, including adjunctive treatments, its long-term clinical impact, and reimbursement are areas of recent interest [73,79]. Overall, usage of this naltrexone depot formulation remains considerably smaller than that of methadone and buprenorphine/naloxone [94]. A six-monthly implant of buprenorphine has been approved in 2016 in the U.S., as an option for persons who had previously been on stable buprenorphine/naloxone maintenance [95].

5.5. Cocaine and abused stimulants / amphetamines:

There are no medications approved for the management of cocaine or other stimulant use disorders. Current therapies focus on psychosocial approaches, including cognitive-behavioral therapies [96]. Several studies have explored the use of sustained release formulations of methylphenidate or amphetamine for this indication, thus using an “agonist-based” approach [97,98], with mixed results. A further pharmacotherapeutic strategy that has been examined for stimulant use disorders is the use of the opioid antagonist naltrexone (which acts primarily as a μ-antagonist and κ-antagonist/partial agonist) [99–101].

6. Market Review:

Medications for substance use disorders are relatively unique among clinical conditions, in that stigma, professional and societal attitudes, insurance reimbursement and legal considerations each have a major impact in the success and dissemination of emerging treatments. Please see also sections 2.1 and 2.2 above, describing general factors related to potential market size for emerging medications for SUD.

6.1. Alcohol:

Epidemiological studies from the Substance Abuse and Mental Health Services Administration (SAMHSA) estimate that approximately 16 million adults and 0.7 million adolescents in the U.S. had an alcohol use disorder in 2014 [102]. Only approximately 10% of these persons received treatment for these disorders, and much of that treatment does not typically include medication-based approaches (e.g., naltrexone or acamprosate) [102,103]. Of note, the NESARC epidemiologic survey from the U.S. indicates that only approximately 15% of persons with an AUD seek treatment for this disorder [15]. Overall, the above data indicate that increasing the proportion of persons with AUD who seek treatment, and expanding the capacity to deliver novel AUD medications appropriately are both critical targets.

6.2. Tobacco-related products:

use of tobacco-related products are the source of massive public health costs [102]. Trends in incidence of new users may differ across countries, and this may affect the size of the perceived market of medications to facilitate smoking cessation and decrease relapse vulnerability.

6.3. Cannabis and related products:

Approximately 4.2 million persons in U.S. had a cannabis use disorder in 2014, of whom approximately 700,000 were adolescents (who may be at particular long-term risk with heavy cannabis exposure) [102,104].

6.4. Heroin and abused prescription opioids:

In 2014, approximately 586,000 persons had a heroin use disorder in the U.S., and approximately 1.9 million persons had a prescription opioid use disorder [102]. It is also known that a some of these may progress to heroin use (in part due to cost), later in trajectory [105]. Only approximately 14% of these persons are in methadone or buprenorphine / naloxone maintenance treatment, due to limited availability of appropriate treatment programs [85], and a considerably smaller proportion are using depot naltrexone [94]. Thus, there is much space to expand the availability of approved treatments, and also to develop possible novel and adjunctive medications.

6.5. Cocaine and other psychostimulants, including methamphetamine:

Approximately 913,00 persons had cocaine use disorder in the U.S., in 2014 [102]. Together with methamphetamine, cocaine poses a major set of public health and societal costs. Use of these psychostimulants is also co-morbid with other drug use (e.g., alcohol or heroin) and the prognosis in persons with poly-drug exposure is especially complex [106,107], and could benefit from novel medication strategies.

7. Current research goals:

We examine below major pharmacological mechanisms that are targeted by medications that may undergo clinical trials in the foreseeable future, based on publications, funded grants and clinical trial registries.

7.1. Scientific rationale:

We describe below the main scientific rationales for the main emerging medications in this therapeutic area.

7.1.1. Opioid receptor mechanisms:

As mentioned above, several of the main medications already approved for substance use disorders target μ-receptors and/or κ-receptors. These include the major medical maintenance medications methadone and buprenorphine, as well as the primarily antagonist medications naltrexone and nalmefene. Future studies that target opioid receptor systems include naltrexone (combined with the benzodiazepine agonist oxazepam) for methamphetamine dependence. Some studies have shown that oral naltrexone can decrease some effects of amphetamine, methamphetamine or cocaine [99,108,109]. More recent studies have also used an extended release depot naltrexone formulation, which however did not show robust therapeutic effects [101].

Of interest, the standard oral naltrexone dose (50 mg) is sufficient to cause robust κ–receptor blockade [110], in addition to its expected μ-receptor blockade in humans [111]. The κ-antagonist CERC-501 (formerly LY2456302) [112,113] is being considered in tobacco use disorders and also in AUD, based on the rationale that upregulated signaling at the dynorphin/κ-receptor system is involved in the chronic relapsing nature of several SUDs [26,114]. One factor that may affect the success of this strategy is whether such upregulation in the dynorphin/κ-receptor system is only important at particular stages in the addiction cycle, or only in a subset of patients that share a categorical diagnosis, based on genetic or other characteristics [114,115].

Several groups are currently focusing on novel opioid-related compounds that would have strong analgesic effects, with substantially decreased undesirable effects (e.g., decreased respiratory depression, abuse potential or constipation). Current strategies for such optimized analgesics involve dual targeting on μ-receptors, and also other receptors, such as the orphanin/nociceptin receptor, κ- and δ-receptor and NK-1 receptor [116–119]. The value in these approaches is that the PD actions at the non μ-receptor site can result in either a potentiation of analgesia, or a decrease in the unwanted side effects of classic μ-agonists. A second strategy is the exploration of downstream signaling “bias” at μ-receptors with novel ligands [120,121]. The rationale of this approach is that a specific downstream pathway (e.g., β-arrestin recruitment) may mediate primarily undesirable effects of μ-agonists. It will be important to determine whether these novel PD targets and signaling mechanisms are robust enough to have a substantial impact in clinical analgesia.

7.1.2. Dopaminergic mechanisms:

Several prior studies have examined the potential therapeutic impact of dopaminergic ligands in SUD, but these compounds have not progressed to clinical approval [122,123]. Disulfiram is approved for alcohol use disorders (though not widely used; see section 5.1). Disulfiram has also been shown to inhibit dopamine-β-hydroxylase (DBH; an enzyme which catalyzes the production of norepinephrine from dopamine). This has led to studies of disulfiram as a potential treatment or adjunct in cocaine use disorders [124–126].

Bupropion, an anti-depressant with norepinephrine and dopamine re-uptake inhibitor properties (as well as antagonist effects at some nicotinic receptor subtypes) is approved for smoking cessation in the U.S. and some other countries. Methamphetamine is approved for narcolepsy, and amphetamine and methylphenidate are approved for primarily for ADHD. These compounds increase synaptic levels of dopamine and other monoamine neurotransmitters. It should be noted that amphetamines themselves are major drugs of abuse in several countries. Several investigators have examined the potential pharmacotherapeutic use of these compounds as “agonist-based” treatments, typically in extended release oral formulations that are thought to pose lesser abuse potential, with some encouraging results [64,127].

7.1.3. Serotonergic mechanisms:

Several serotonergic receptors are being investigated for potential therapeutic approaches in substance use disorders. These include lorcaserin, a 5HT3-agonist that is approved for weight loss, which is being evaluated for cocaine use disorders [128]. Also, ondansetron, a 5HT3-antagonist which is approved as an antiemetic, may be studied for its effects on smoking. Ondansetron has also been studied in alcohol use disorders, and potential pharmacogenetic associations to its effectiveness have been reported [42,129].

7.1.4. Adrenergic mechanisms:

Some recent studies examined the potential effectiveness of the norepinephrine reuptake inhibitor atomoxetine [130,131], which is currently approved for treatment of ADHD, without encouraging results. The α₂-agonist guanfacine, which is also approved for treatment of ADHD, may be further studied as a medication against smoking and alcohol abuse. α₂-agonists including clonidine and lofexidine (the latter only in some countries), are used primarily to manage short-term withdrawal in opioid-dependent persons.

7.1.5. Mixed monoamine system approaches:

Several planned trials target inhibition of the re-uptake of dopamine, serotonin, and norepinephrine, with the hypothesis that such multiple inhibitors (e.g., NS2359 and desvenlafaxine) may have a superior profile, compared to compounds that target just one of these transporters. Other “re-purposing” studies may also focus on compounds that are approved as mood stabilizers or atypical anti-psychotics, which target diverse monoaminergic receptors.

7.1.6. GABA mechanisms:

GABA is the main inhibitory neurotransmitter in the CNS. Compounds that potentiate GABA signaling have been investigated for potential therapeutic effects against substance use disorders. Some planned studies may further examine this potential, for example with the GABA transaminase inhibitor vigabatrin, which is approved as an anti-epileptic agent. The anti-epileptic topiramate has potentiation of GABA-A receptor mechanisms as one of its actions, and has been under recent study (alone and in combination) for its effects on specific SUD [72,97]. A further combination product under study is EMB-001, a combination of the benzodiazepine-GABA site agonist oxazepam and the cortisol synthesis inhibitor metyrapone [132]. Other GABA-related medications which have been studied for the treatment of different SUD include gabapentin and pregabalin [133–135], which are thought to act primarily by blocking specific sub-units of voltage-gated calcium channels. Intriguingly, the abuse potential of gabapentin and pregabalin themselves has also been the cause of recent concern [136,137], therefore their use in patients with SUD would have to carefully evaluated.

7.1.7. Glutamatergic mechanisms:

Glutamate is a major excitatory neurotransmitter throughout the CNS, and acts through several metabotropic and ionotropic receptor subtypes [138]. As mentioned above, acamprosate is approved for the treatment of alcohol use disorders, and has modulation of NMDA receptor function as one of its mechanisms of action [138]. Compounds with non-competitive antagonist effects at ionotropic NMDA-subtype receptors, including memantine and ketamine (which is a DEA-scheduled drug itself), are approved for other indications. Several studies have examined whether such compounds (especially ketamine) could have pharmacotherapeutic potential in specific SUD [139,140], with other studies currently planned [141]. Emerging areas of work with ketamine include potential optimization of the PD profile of this compound, and also determination of ways to achieve safe and effective translation to clinical use for SUD [142,143].

7.1.8. Cannabinoid mechanisms:

The main psychoactive component of the Cannabis sativa plant is the CB1-r partial agonist Δ9-THC, which has abuse potential, and is thought to be the main compound that underlies cannabis-use disorders. Preparations containing Δ9-THC are also used in specific medical conditions, including AIDS-induced anorexia, and chemotherapy-induced nausea, in different countries [52,144].

Cannabidiol (CBD) is another major product from Cannabis sativa. Cannabidiol is not a CB1-r or CB2-r agonist, is not thought to have abuse potential, and has specific indications (e.g., in multiple sclerosis). Cannabidiol’s mechanism(s) of action are not fully understood [52]. CB1-r and its endogenous agonists (e.g., the endocannabinoids anandamide and 2-arachidonyl-glycerol) modulate major neurobiological pathways involved in SUD [145]. The potential pharmacotherapeutic use of cannabinoid formulations has been studied in specific SUD (including aspects of opioid use disorder), but their effectiveness is not broadly supported at this time [79,146,147]. CB1-r receptor systems can also be targeted by synthetic small molecules, and synthetic CB1-r full agonists are also abused [49,148]. Synthetic molecules with appropriate PD/PK (for example, CB1-r antagonists or partial agonists with prolonged durations of action) may be considered as pharmacotherapeutic strategies in specific SUD [49]. For example, a CB1-r antagonist or inverse agonist, rimonabant, showed some promising effects in smoking cessation in humans, as described in a recent review [149]. However, rimonabant was removed from the European market due to psychiatric side effects, therefore the potential of CB1-r antagonists or inverse agonists for SUD treatment in humans remains unclear. Rimonabant itself (and other CB1-r antagonists or inverse agonists) have produced effects consistent with pharmacotherapeutic potential against nicotine, cocaine or opioid use disorders, in preclinical models [150–153].

7.1.9. Cholinergic mechanisms:

Based on the direct action of nicotine at different receptor subtypes, different studies are designed to examine whether selective ligands may have encouraging profiles for smoking cessation studies. As mentioned above, cholinergic ligands including varenicline (and various nicotine formulations) are approved for use against smoking use disorders.

7.1.10. Orexin receptor system approaches:

The orexin (or hypocretin) neuropeptide and its two major GPCR receptors, OX1 and OX2, were originally implicated in CNS mechanisms including the sleep-wake cycle and appetitive behaviors. More recent work has shown that orexin receptor antagonism decreases drug and alcohol intake, in preclinical models [154–156].

7.1.11. Vasopressin receptor system approaches:

The neuropeptide (and hormone) vasopressin acts at three main receptors. Studies show that vasopressin systems (especially V1b receptors) are involved in the modulation of stress HPA-axis activity [157], and also drug and alcohol intake, based on preclinical and clinical studies [158,159]. A recent study with the V1b antagonist ABT-436 had favorable effects in a Phase II study, in persons with alcohol dependence [160].

7.1.12. Vaccine, antibody and, and enzyme-based approaches:

This paper focuses primarily on medications for substance use disorders, rather than on active immunity (i.e., vaccine), passive- immunity (injected antibody), or enzyme-based approaches. The reader is referred to recent reviews and articles on these topics [161–164]. Overall, the logic of the vaccine approaches is to produce a sufficient host immune response to a major drug of interest (e.g., cocaine or a specific μ-agonist), such that the actual blood concentrations of free drug are decreased sufficiently that the CNS-mediated effects of the drug are decreased. Most active immunity studies to date have contended with a variable immune response across participants, limiting clinical effectiveness [165]. Some studies have reported that particular genetic polymorphisms may be used to select patients who could have a more reliable clinical response to such vaccines [162]. Among the factors that are being considered by emerging vaccine approaches are achievement of a greater population of high affinity antibodies, with decreased variability across participants. A second group of factors being discussed are the optimal strategies to achieve a clinically valuable response. For example, a potential indication is the use of vaccination strategies in individuals at risk of progression of an addictive disease, rather than on persons who have already reached a diagnosis of severe substance use disorder. A further possibility is the combined use of passive and active immune approaches. In this situation, the passive (antibody or enzyme) treatment can result in a relatively rapid onset of protection, while the longer term active immunity response is activated by vaccines and boosters.

7.1.13. Glial-related and inflammatory mediators as target mechanisms:

Several studies have shown that μ−agonists and cocaine can result in glial activation, and also affect diverse mediators usually associated with neuroinflammation [166,167]. Also, Recent studies show that the glial activation inhibitor ibudilast can decrease signs of withdrawal in opioid-dependent persons, and can also reduce the reinforcing effects of oxycodone [168,169].

8. Competitive environment:

This section will focus on specific compounds that have reached proof-of concept clinical studies or planned therapeutic clinical trials. The readers may also refer to the sections above for more background on the pharmacology and neurobiology of some of these compounds. (see Table 1)

8.1. Potential development issues:

Substance use disorder medications have been developed with several rationales and translational paths. Most current approaches favor a mechanism-based targeting of particular systems, receptors or transporters. These may be strengthened by medicinal chemistry and PK/PD studies to achieve compounds or formulations with suitable duration of action, target engagement and bioavailability. Another trend in potential substance abuse medications is the “re-purposing” medications that are already approved for other indications, or that have successfully reached the clinical trial stage.

Some factors continue to affect the development path for SUD pharmacotherapies. These include abuse, tampering and diversion liability of the potential pharmacotherapies themselves. Other factors related to treatment compliance are also relevant, including the potential pharmacotherapy’s route of administration and duration of action. In general, oral or sub-lingual formulations with once-daily administration have been favored (e.g., daily methadone or buprenorphine/naloxone dosing). Shorter acting medications (especially with fast onset or offset; e.g., other prescription opioids) typically have greater abuse potential. In selected cases, depot or implant formulations (e.g., naltrexone and buprenorphine) have been approved in some countries, but their use remains limited to date.

Several preclinical models in rodents and non-human primates can be used to examine the potential effectiveness of novel medications in modulating reinforcing and neurobiological effects of drugs of abuse [5]. These include self-administration, conditioned place preference, and drug discrimination studies [3,170], as well as studies of dopamine microdialysis or intra-cranial self-stimulation [113,171]. Optimized candidates from these efforts can then undergo Phase I and clinical laboratory studies, to examine the impact of the candidate medication on rewarding, subjective or other effects of specific drugs of abuse [109,172]. When favorable conditions are met, this may be followed by Phase II and III clinical trials of the therapeutic effect of these medications [2,4].

However, the field as a whole has suffered from relatively high attrition and lack of success once the clinical stages are reached. It has been argued recently that a number of candidate medications for SUD in the past have not undergone a robust preclinical evaluation (for example in rodent and primate translational models), and that greater integration among preclinical and clinical methodologies should be prioritized [2,4]. In some cases, predictive validity of preclinical and Phase I studies may be compromised due to sub-optimal pharmacological and behavioral designs, which do not model important aspects of the clinical course of the SUD. For example, preclinical studies with animals with short-term or low-dose exposure to the drug of abuse may not model neuro-behavioral changes that occur clinically. In both preclinical and Phase I clinical studies, the use of single (rather than repeated or chronic) doses of the potential SUD medication may not be a robust predictor of therapeutic potential.

9. Conclusion:

Substance use disorders are the cause of massive public health and societal costs, and pose a diverse set of challenges for medication development, in addition to the overall challenges encountered for other CNS diseases. There is continuing need for broadly effective and disseminated medications for alcohol use disorder, cocaine and other psychostimulant use disorders, and cannabis use disorders. Psychiatric comorbidity and poly-drug exposure, as well as phenotyping approaches are current challenges to the examination of emerging medications.

10. Expert opinion:

10.1. Key findings:

Some underlying neurobiological systems are involved in several major substance use disorders; these include: dopaminergic, κ- and μ- opioid neuropeptide and receptors, glutamatergic, GABA, and also stress-axis systems.

10.2. Potential of current research:

Addition of pharmacogenomics (see above, section 3.3), neuroimaging and more mechanism-based and dimensional phenotyping may add to improved success for study design and analysis. PET imaging, in addition to its value for the study of PK/PD relationships and target engagement, can also be valuable for the mechanism-based analysis of SUD. For example, by studying particular PD targets in patients with different psychiatric diagnoses or exposure to drugs of abuse, it may be possible to determine whether the status of these systems may underlie specific aspects of SUD [114,173,174].

10. 3. Needs and challenges of current research:

Recent papers have emphasized the importance of optimizing translation between preclinical and clinical development for SUD, and also translation from clinical laboratory studies to Phase II and III trials [2,4,175,176]. These challenges in the development of novel medications for SUD have been recognized in the CNS field as a whole [177,178]. Different preclinical models of SUD can recapitulate some of the major aspects of the clinical condition (e.g., escalation of use, relapse-like behavior, withdrawal, underlying neurobiological changes due to drug exposure) [5,179,180]. However, our understanding of neurobiological changes that underlie clinical recovery from SUD (and resilience to relapse) is relatively limited [181–186]. Therefore, other than for medications which act at the main PD site of action of the drug of abuse (e.g., with “agonist-based” or “antagonist-based” approaches), target selection for therapeutic mechanisms remains a high-risk process. Recent studies have also examined how specific behavioral processes interact with the neurobiological effects of drugs of abuse themselves, and of SUD medications [124,187]. Behavioral research techniques, including ecological momentary assessment, can also allow a detailed examination of a participant’s behavior in her environment [188].

Comorbidities, including trauma- and stress - related disorders, as well as mood and anxiety disorders, are continuing challenges for clinical trial design in SUD. These comorbidities pose particular challenges both with respect to the predictive value of preclinical studies, and also for participant selection and stratification strategies in clinical trials.

10.4. Other challenges in the development and clinical dissemination of medications for specific SUD:

Some of these challenges in the SUD pharmacotherapy space are similar to those for which occur in other CNS and mental health disorders. For example, even approved medications for certain mental health disorders are not effective in a substantial proportion of patients [189]. Also, the rate of placebo responders is high both in clinical trials for other mental health disorders, in addition to SUD [190,191]. The possibility that specific medications may be effective in a subset of patients based on endophenotypes or specific genetic polymorphisms is also the focus of current work. Of note, the U.S. National Institute of Mental Health (NIMH) has developed the Research Domain Criteria (RDoC) framework, with the goal to understand mechanisms of mental health diseases in a dimensional manner, in addition to standard categorical diagnoses [192]. Such criteria may also “map” more directly onto underlying neurobiological mechanisms, and may cut across categorical diagnoses.

Other challenges may be more unique to the SUD field, and even to specific SUD. Among these remains the stigma of SUD (especially for illicit substances) [1,19]. This is reflected in societal and legal attitudes which may affect the type of treatment, its reimbursement and availability. Certain stakeholders involved in the therapeutic process (including social support groups and different types of treatment centers and clinicians) may be philosophically opposed to pharmacotherapy of SUD, or may have the perception that these pharmacotherapies are not efficacious [9,193,194]. Clearly, both these phenomena can affect dissemination of novel pharmacotherapies.

The chronic relapsing nature of SUD may pose further challenges in diagnosis and pharmacotherapeutic management. It has also been suggested that specific SUD be further subdivided into “stages”, for the purposes of preclinical and clinical research [22,114,195]. Therefore, clinical effectiveness of a given pharmacotherapy may depend based on the “stage” of a specific SUD, due to differences in underlying neurobiological adaptations that occur over time. Thus, various preclinical and clinical studies have shown that behavioral, neurobiological and neuroendocrine responses remain altered for prolonged periods after the end of drug exposure [158,173,196,197]. Other studies also show that repeated cycles of substantial drug exposure and drug-free periods (modeling “abstinence” or withdrawal) can result in progressive behavioral and neurobiological plasticity [180,198]. The likelihood that such plasticity may differ across patients (depending on the stage of their disease, as well due to genetic and epigenetic factors) may increase the heterogeneity of response to a candidate pharmacotherapy.

Development of pharmacotherapies for SUDs may also be relatively unique in that some clinical studies are carried out in diagnosed participants who are “non-treatment seeking”. This may occur for a variety of practical, ethical and regulatory reasons [199]. It is unclear if the study of such “non-treatment seeking” participants results in challenges to appropriate “go/no-go” decisions for progression to further clinical development [18,199]. It is possible that the practice of studying non-treatment seekers may affect development of different types of pharmacotherapies in a different manner. For example, adequate examination of medications that act at the same PD site as the drug of abuse (i.e., using either “agonist-based” or “antagonist-based” models) could be achieved in non-treatment seeking participants. However, examination of candidate medications that target other mechanisms (including stress-responsivity, motivation, neurocognitive and emotional regulation) may be less effectively studied in this manner.

10.5. Potential evolution of the field in the foreseeable future:

As mentioned above, pharmacogenomic, neuroimaging and more mechanism-based and dimensional phenotyping may be undertaken, to “de-risk” studies in this field. Development of robust neuroimaging and peripheral biomarkers of SUD (e.g., based on protein, mRNA, genetic or epigenetic data) is also an area of emerging interest [200–203]. The goal of such biomarker development is to generate a broader array of diagnostic and prognostic factors in SUD, in addition to standard approaches based on patient reports and clinical history. Transdiagnostic approaches are also under study, based on shared neurobiological mechanisms across several substances of abuse, and with specific psychiatric comorbidities [174,192]. These efforts may eventually lead to medication approaches that ameliorate downstream targets affected across drugs of abuse, and that are impacted in SUD as well as comorbid psychiatric diseases.

10.5. Medications that hold promise:

Medications that have a strong mechanistic rationale (e.g., novel κ- opioid, orexin and vasopressin receptor ligands) are important for proof-of-concept studies. These may be involved in shared underlying mechanisms for SUD, including modulation of dopamine, mood (dysphoria/euphoria) and reward systems, as well as stress responsivity. Maintenance-based (“agonist based”) approaches with long-lasting amphetamine-like analogs (or extended release formulations), and multiple monoamine re-uptake inhibitors may also hold promise for cocaine and other stimulant use disorders.

10.6. Areas of current interest:

As mentioned above, there is active interest in novel studies with cannabidiol, and also with re-purposed compounds acting at glutamatergic or GABA-ergic systems, and also different neuropeptide systems (κ-receptor/dynorphin, vasopressin and orexin). Novel inhibitors of multiple monoamine re-uptake are also under study.

Abbreviations:

Δ9-THC

Δ9-tetra-hydrocannabinol

ADHD

Attention deficit and hyperactivity disorder

AUD

Alcohol use disorder

BZ

benzodiazepine – GABA receptor site

CB1-r

cannabinoid-1 receptors

DA

dopamine

δ-receptor

delta-opioid receptor

DBH

dopamine-β-hydroxylase

DEA

U.S. Drug Enforcement Administration

GPCR

7-transmembrane domain G-protein coupled receptors

HCV

Hepatitis C virus

HIV

Human immunodeficiency virus

HPA Axis

Hypothalamic-Pituitary Adrenal axis

κ-receptor

kappa-opioid receptor

MA

monoamine neurotransmitters (i.e., dopamine, norepinephrine, serotonin)

μ-receptor

mu-opioid receptor

NK-1

Neurokinin-1

NLX

naloxone

NSDUH

U.S. National survey on drug use and health

PD

Pharmacodynamic

PK

Pharmacokinetic

POC

proof-of concept

RDoC

Research Domain Criteria framework of the U.S. National Institute for Mental Health

SUD

substance use disorder