Oxytocin as a Treatment for Alcohol Use Disorder and Heavy Drinking: A Narrative Review

Abstract

Oxytocin is increasingly being studied for treating symptoms of alcohol use disorders and heavy drinking behavior. The neuropeptide oxytocin facilitates social relationships and modulates the body’s stress response by strengthening coping mechanisms and reducing anxiety. Relatedly, oxytocin is also thought to play a role in processes associated with craving and withdrawal from alcohol. This review aims to primarily provide an overview of preclinical and clinical literature on the applications of oxytocin in alcohol use, and additionally discuss a framework for types of trials and the variety of parameters that affect different study designs. A review of the existing literature in this area suggests that while low dosages of oxytocin do not affect drinking behavior and tolerance, higher dosages taken prior to alcohol exposure have varying behavioral and physiological results. Depending on quantity and timing, oxytocin treatments resulted in declines in withdrawal symptoms and alcohol self-administration in preclinical studies and may decrease neural cue-reactivity and withdrawal symptoms in clinical studies. Current ongoing trials are expanding on this work to thoroughly explore clinical applications of oxytocin.

Introduction

Background

Oxytocin (OT) is a neuropeptide comprising nine amino acids widely distributed in the central nervous system (CNS) (Hammock, 2015). It is synthesized in the supraoptic and paraventricular nuclei of the hypothalamus and released from axon terminals into the blood stream via oxytocinergic pathways (Hammock, 2015; Keating et al., 2013). OT plays a key role in the regulation and release of adenohypophyseal hormones including prolactin, gonadotropins, adrenocorticotropin (ACTH), and corticotrophin-releasing factor (CRF). Oxytocin facilitates processes within the endocrine system, including uterine contraction during childbirth as well as the stimulation of lactation in females. In males, OT is involved in sperm transportation and testosterone production (Marazziti et al., 2022). In the CNS, OT additionally facilitates social relationship development with effects that assist with pair bonding and maternal attachment (Lee & Weerts, 2016). However, OT also has a variety of documented neural effects on stress modulation of associative learning and motivational processes and processing of reward, which are implicated in the development of addiction and substance use disorders.

Animal studies have suggested that neuropeptides such as OT could be promising therapeutics for CNS disorders, however obstacles to neuropeptide administration of this nature include a limited half-life and poor penetration across the blood brain barrier (BBB), which limits bioavailability in the brain (Oppong-Damoah et al., 2019). To date, preclinical studies have yielded variable results on brain pentrance. Intranasal oxytocin (doses) produced minimal increases in OT levels in cerebral spinal fluid in some studies conducted in macaques and rats (Mens et al., 1983; Chang et al. 2012; Dal Monte et al., 2014), and two studies found no increase in OT (Leng & Ludwig, 2016). In contrast, CNS penetrance of IV administered OT (80 IU) and comparable pharmacokinetics and concentrations in cerebral spinal fluid of IV and intranasal routes was demonstrated in non-human primates (Lee, Scheidweiler et al. 2018). Therefore, discrepancies between studies may be related to dose, route of administration, and/or species differences.

OT and Vasopressin Receptors

OT receptors are widely distributed in the brain in regions involved rewards, social behavior, emotions, and memory, including the amygdala, hypothalamus, nucleus accumbens, and hippocampus (Bale et al., 2001; Stoop 2012; Baribeau & Anagnostou, 2015), and share structural and functional similarity with other compounds including vasopressin. While OT primarily binds to the oxytocin receptor (OTR), vasopressin has three distinct types: AVPR1a, AVPR1b, and V2. While both systems are involved in processes such as modulation of social behavior and stress, vasopressin is believed to contribute to aggressive and affiliative behaviors (Kelly & Goodson, 2014), whereas oxytocin is believed to help facilitate social and maternal bonds (Stoop, 2012). These systems are believed to interact and cross-talk with one another, and may contribute to complimentary processes (Baribeau & Anagnostou, 2015).

The Endogenous OT System and its Role in the Modulation of Anxiety and Stress

The stress response is modulated by three key neuroendocrine circuits: the hypothalamic-pituitary-adrenal (HPA) axis, the extrahypothalamic CRF system, and the adrenomedullary system (Lee & Weerts, 2016). Of these, the HPA axis is responsible for the regulation of arousal and stress (Joseph & Whirledge, 2017). The HPA axis releases corticotrophin-releasing hormone and vasopressin, which stimulate the production of adrenocorticotropin and corticosteroids (Aguilera, 2011). Norepinephrine and corticotrophin-releasing hormone expression in extrahypothalamic brain regions additionally play a role in the expression of stress and anxiety (Lee & Weerts, 2016).

Stress is measured as a response to challenges that cause physical, psychological, or emotional distress and create a threatened homeostatic state (Takayanagi and Onaka, 2022). The relationship between OT and stress has been supported by animal and human studies, which demonstrate that OT can have a protective effect when an individual is confronted with fear-inducing stimuli (Riem et al., 2020; Ballas et al., 2021; Flanagan et al., 2019; Ayers et al., 2011; László et al., 2022). Further, endogenous OT appears to strengthen adaptive responses to stressors by decreasing anxiety and facilitating accurate threat recognition (Janecek & Dabrowska, 2019). Given that acute and chronic stress increase self-administration for a range of drugs with misuse potential, there is theoretical support that OT administration may have downstream effects on alcohol and other forms of substance use (Brady & Sonne, 1999).

Indeed, multiple reviews underscore the importance of OT on neurochemistry and behavioral responses to substances such as cannabis, opiates, and stimulants, and have suggested that OT is involved in the development maintenance of addictive behaviors across multiple substances (King et al., 2020; Lee & Weerts, 2016; Edinoff et al., 2022; Lee et al., 2016). The aims of the present review are to update and build upon this literature by reviewing the effects of OT on alcohol administration, tolerance, craving, and withdrawal, across both clinical and preclinical studies.

Alcohol Use Disorder (AUD)

According to the 2021 National Survey on Drug Use and Health conducted by the Substance Abuse and Mental Health Association (SAMHSA), 133.1 million people aged 12 or older reported past month alcohol use, which represents nearly half of the U.S. population (47.5%). Alcohol use encompasses a broad spectrum of drinking patterns, ranging from occasional social drinking to binge drinking and heavy drinking. Binge drinking is defined by SAMHSA as the consumption of five or more drinks on a single occasion for males, or four or more drinks in a single occasion for females, at least once in the past 30 days. Heavy drinking is defined as binge drinking for five or more days in the past 30 days. About 16.3 million people reported past-month heavy drinking behavior (Substance Abuse and Mental Health Association [SAMHSA], 2021). Binge drinking and/or heavy drinking behaviors are associated with greater risk for an Alcohol Use Disorder (AUD).

An AUD diagnosis describes a pattern of alcohol use that leads to clinically significant impairment. AUD is classified along a spectrum of mild, moderate or severe, contingent upon the number of symptoms experienced. Symptoms of AUD can include repeated use despite social, emotional, or physical consequences; difficulties controlling use; alcohol craving; tolerance; and, in severe cases, the development of physical dependence on alcohol and the emergence of withdrawal symptoms upon discontinuation (American Psychiatric Association [APA], 2022). AUD can cause or exacerbate a host of physical conditions, including liver disease, various cancers, and cardiovascular disease (Shield et al. 2014; Lee et al., 2020). Furthermore, AUD often co-occurs with mental health conditions, including anxiety disorders, major depressive disorder, post-traumatic stress disorder, bipolar disorder, attention-deficit hyperactivity disorder, borderline personality disorder, antisocial personality disorder, and other substance use disorders (Castillo-Carniglia et al., 2019).

In addition to sociocultural contextual variables, such as life adverse events and socioeconomic status, the development of AUD is linked to other social factors like isolation and lack of support (Sudhinaraset et al., 2016). Social ostracization and poor access to community support, for example, are associated with a higher prevalence of alcohol and substance use disorders (Rapier et al., 2019).²³ Perceived burdensomeness on others can also lead to social withdraw and rejection of social support, heightening the risk of regular alcohol use to cope with stress (Le et al., 2021). This pattern was witnessed during the height of the COVID-19 pandemic, where social isolation resulting from stay-at-home orders preceded a widespread increase in alcohol use and heavy drinking behavior (Killgore et al., 2021; Stockwell et al., 2021; Sharma et al., 2020). This was particularly pronounced among vulnerable groups, including persons experiencing financial and social stress (Murthy & Narasimha, 2021).

Design and Outcome Measures in Clinical Trials for Alcohol Use

There have been several clinical trials examining potential medications for AUD, including naltrexone, acamprosate, nalmefene, baclofen, gabapentin, and topiramate. Results from this literature suggest that participant AUD severity and drinking level impact both study power and may also influence the effectiveness of different medications. For example, across studies, Baclofen has been shown to produce particularly pronounced effects among individuals who have higher daily alcohol use at baseline, suggesting that this medication may be particularly beneficial for those with greater AUD severity (Pierce et al., 2018). However, a recent meta-regression analysis found that individuals with milder alcohol use severity are more likely to reduce their drinking early in a study and are more likely to demonstrate a placebo response, suggesting limited generalizability and potential risk for type II error in studies with a large proportion of participants with mild AUD. Unfortunately, AUD severity at baseline is infrequently controlled for in clinical trials (Scherrer et al., 2021). Further, some studies have required abstinence from participants to qualify for the study. This requirement may be difficult to achieve for individuals engaging in regular heavy drinking and those for whom abstinence is not a goal.

Further, studies on medications for AUD have varied regarding the clinical endpoints used. The U.S. Food and Drug Administration (FDA) has historically recommended abstinence or no heavy drinking as efficacy endpoints in trials for AUD (U.S. Food and Drug Administration [FDA], 2015). However, these endpoints may not accurately capture reductions in drinking among individuals looking to reduce their drinking but not stop completely. Thus, there has been increased discussion around alternative meaningful clinical endpoints in AUD trials to capture reductions in use among individuals for which abstinence is not a goal (Falk et al., 2019; U.S. FDA, 2018). ^32,31 Alternative and/or potential endpoints have included reduction in World Health Organization (WHO) risk levels, percentage of heavy drinking days, drinks per drinking day at the end of treatment, adverse events, dropout, and biomarkers of liver function or the alcohol metabolite phosphatidylethanol (PEth) (Minozzi et al., 2018; Falk et al., 2019; Sobell, 2003). Many of these metrics can be assessed using the Alcohol Timeline Followback (TLFB), a psychometrically validated assessment in which individuals provide estimates of their past drinking history on a day-by-day basis (Sobell, 2003). This tool is useful in clinical trials for providing accurate drinking estimates, and can be used to establish a baseline for a participant prior to study enrollment.

At present, there are three medications currently approved for the treatment of AUD in the United States. The first of these is Naltrexone, an opioid antagonist that can be injected or taken as a pill (Fairbanks et al., 2020). Naltrexone competitively binds to opioid receptors in the brain, which blocks the release of endogenous opioids caused by alcohol. As a result, the euphoric feelings associated with alcohol are blocked, which reduces incentive to engage in alcohol use (Sudakin, 2016). Naltrexone has shown to be effective in reducing craving and heavy drinking, but is contraindicated for patients with acute hepatitis and/or liver failure (Witkiewitz et al., 2019).

A second treatment option is disulfiram, intended for patients whose goal is abstinence. Disulfiram inhibits aldehyde dehydrogenase, a liver enzyme that plays a key role in the metabolism of alcohol (Edenberg & McClintick, 2018). If alcohol is ingested in patients taking disulfiram, the inhibition of aldehyde dehydrogenase blocks alcohol metabolism in the liver, which leads to accumulation of acetaldehyde, and subsequent adverse reactions including tachycardia, nausea and vomiting, facial flushing, and palpitations that can last for an hour or longer. Though these resulting uncomfortable physical symptoms can discourage further alcohol ingestion, Disulfiram is not an optimal medication for persons for whom abstinence is not a treatment goal, for persons ambivalent about reducing drinking, and for persons with cardiovascular conditions that may be exacerbated if Disulfram is combined with alcohol (Jørgensen et al., 2011). Additionally, previous literature on Disulfram adherence suggests that compliance may be in issue. For this reason, it is recommended that the medication be taken with close supervision or reserved for individuals who are extremely motivated by abstinence (Walker et al., 2019).

Lastly, acamprosate targets the physiological stress caused by alcohol withdrawal and responses to alcohol cues by acting as a modulator of NMDA receptors, as well as through actions at glutamate and gamma-aminobutyric acid receptors. Acamprosate is not metabolized by the liver and can be used for treatment of AUD among people with liver disease, making it a potential useful treatment among this population (Witkiewitz et al., 2012). However, some mixed findings with acamprosate have been observed, suggesting that it may not be effective for all individuals with AUD and that further research is needed (Kranzler & Hartwell, 2023).

Though these three medications have been shown to help treat symptoms of alcohol use disorder, utilization of these medications is low. Around 2.7% of individuals with alcohol dependence reported receipt of a medication for alcohol use disorder in 2019 (Han et al., 2021), suggesting that exploring novel targets may increase treatment options for people with AUD.

OT and Alcohol

OT represents a promising medication worthy of further investigation. Broadly, there is evidence that repeated exposure to substances results in neuroadaptive changes to the endogenous OT system in brain regions implicated in the addiction process, including the mesolimbic dopamine system and systems involved in stress response (Lee & Weerts, 2016). Additionally, OT has been shown to affect key brain mechanisms associated with craving and withdrawal by reducing neuroadaptations to alcohol (Bowen et al., 2015; Hansson et al., 2018). Further, OT receptors are expressed in many different brain areas involved in executive function, emotion processing, and cognitive control, such as the prefrontal cortex regions and amygdala (King et al., 2021). These areas are impacted by alterations due to chronic alcohol exposure, suggesting that OT receptor activity may play a role in the addiction cycle in these regions (Pleil et al., 2015). Finally, chronic alcohol use exacerbates stress and anxiety, leading to increased secretion of corticotropin-releasing factor, norepinephrine, and cortisol (Sinha, 2012). The role of OT in the stress-mediating process could be a target mechanism for decreasing effects of alcohol craving and motivation to drink. Here, we review the preclinical and clinical literature on the effects of OT on alcohol administration and response to alcohol.

Method

Search Methods

A search was performed in the databases PubMed, ScienceDirect, Scopus, and Google Scholar between the months of September 2022 to April 2023. Search terms included “oxytocin,” “oxytocin and alcohol,” “alcohol use disorder,” and “oxytocin system.” Additional studies were found by examining the references section of existing reviews and studies. Inclusion criteria for the search were published preclinical and clinical studies that looked at the physiological effects of oxytocin dosages on various aspects of alcohol dependence and alcohol use disorder.

Transparency and Openness

Data and code were not utilized in conducting this review; thus, materials and analysis code for this study are not available. The review was not pre-registered.

Results

Preclinical Studies

Twenty preclinical studies analyzing the role of OT in physiological effects of alcohol were assessed. These studies suggest that OT can influence behavioral and physiological effects of alcohol, but route of administration, sex, time of administration, and frequency of administration are key factors. A summary of preclinical studies is presented in Table 1.

Table 1.

Preclinical Studies on the Effects of OT on Alcohol-Related Behaviors

Source	Species	Sex	OT dose amount(s)*	OT dose frequency	Route(s)	Main outcome measure(s)	Main finding(s)
Bahi (2015)	C57BL/6 mice	Male	• 6.4 mg/kg	• Pretreatment before each ethanol administration	IP injection	• Alcohol-induced conditioned place preference response	• Carbetocin-induced activation of OT receptor (OTR) resulted in decreased acquisition of alcohol place preference, resulted in less time spent in the ethanol-paired compartment, and blocked instatement of alcohol CPP.
Bowen et al. (2011)	Australian Albino Wistar (AAW) rats	Male	• 1 mg/kg	• Daily treatments for 10 days, followed by a “booster shot” of OT after 25 days	IP injection	• Amount of alcohol consumed	• In a 9-day period, there was no difference in beer consumption between OT- and vehicle-treated groups. • In a 25-day period, rats pretreated with OT consumed significantly less alcohol compared to vehicle rats. • OT administration resulted in the long-term reduction of alcohol consumption in an environment with continual access to this substance. • The administration of a “booster shot” of OT following the 25-day period further inhibited alcohol consumption for the immediate 2.5 hours following.
Bowen et al. (2015) 44	Adult Albino Wistar rats	Male	• 1μg/5μL	• Singular pretreatment	ICV injection	• Righting reflex • Wire hanging test • Open-field test	• OT administered intracerebroventricularly immediately prior to ethanol dosing resulted in the attenuation of acute alcohol effects in a motor impairment test. • OT did not show an effect with higher quantities of ethanol.
Caruso et al. (2021)	C57BL/6J mice	Male and female	• 3 mg/kg	• Daily treatments for 4 days	IP injection	• Amount of alcohol consumed	• Mice treated with OT consumed significantly less alcohol compared to vehicle at 1-, 2-, and 3-hrs post-treatment on the first day of observation. • No effect was observed on days 2.  • On day 3, the OT group consumed more alcohol 2h and 3 post-administration, and more 13h.  • On day 4, there was lower alcohol administration in the OT group at 2h (trend-level)
Hansson et al. (2018)	Wistar rats	Male	• 10 nM	• Singular treatment following 3 weeks of alcohol abstinence	ICV injection	• Changes in OTR expression • Cue reinstatement"	• Alcohol-dependent rats had elevated levels of OTR mRNA and OTR proteins in several frontal and striatal brain areas following 3 weeks of abstinence. • OT administration also reduced cue reinstatement in alcohol-dependent rats.
King et al. (2017)	Adult C57BL/6J mice	Male	• 1, 3, or 10 mg/kg	• Singular treatment	IP injection	• Preference for alcohol vs. water • Locomotor activity • Alcohol self-administration	• OT significantly reduced ethanol consumption and BAC in mice. • Dosages of 3 and 10 mg/kg significantly reduced ethanol consumption compared to vehicle. • OT treatment also reduced ethanol consumption in a two-bottle choice task.
King & Becker (2019)	Adult C57BL/6J mice	Male and female	• 0.1, 0.5, or 1 mg/kg	• Singular treatment	IP injection	• Amount of alcohol consumed • Alcohol self-administration	• OT treatment significantly reduced alcohol consumption in response to predator odor • OT treatment appeared to reduce alcohol seeking and responding in response to yohimbine in mice.
MacFadyen et al. (2016)	Sprague–Dawley rats	Male	• 0.05, 0.1, 0.3, or 0.5 mg/kg	• Singular treatment prior to bottle access	IP injection	• Amount of alcohol consumed	• Alcohol intake was reduced in about 40% of days in which OT was given in a three-bottle choice task. • Following 0.1, 0.3, and 0.5 mg/kg of OT, 100% of animals decreased their alcohol consumption. • OT decreased ethanol consumption in a 30-minute session of alcohol intake.
McGregor & Bowen (2012)	Rats	Male	• 1 mg/kg	• Singular pretreatment • Daily pretreatments over the course of 10 days	IP injection	• Preference for alcohol vs. sucrose	• One dose of OT resulted in a 6 week decline in ethanol preference compared to sucrose solution. • Rats pretreated with OT displayed a significantly lower preference for ethanol.
Peters et al. (2013)	C57BL/6N mice	Male	• 10mg/kg • 0.5μg/2μl	• Singular treatment following 3 days of steady ethanol consumption	ICV or IP injection	• Amount of alcohol consumed	• 10mg/kg IP OT, but not 0.5μg/2μl ICV OT, reduced alcohol intake
Peters et al. (2017)	Wistar rats	Male	• 1 μg/5 μl	• Singular treatment following intermittent ethanol consumption, daily alcohol consumption, and in alcohol-naïve animals	ICV injection	•Preference for alcohol vs. water •Amount of alcohol consumed •Dopamine release in nucleus accumbens	• Mice treated with OT consumed less alcohol than control animals and showed lower alcohol preference in a 24-hr drinking period after treatment compared to before treatment. • No increase in local dopamine production in OT-treated animals, but there was an increase in non-OT-treated animals •Infusions of OT reduced the consumption of ethanol in repeatedly exposed mice
Puciklowski et al. (1985)	Wistar rats	Male	• 160, 800, or 2400 nmol/kg	• Daily treatment for 5 days	SC injection	• Tolerance to hypnotic effects of ethanol  • Hypothermic tolerance (body temperature measurements)	• Repeated OT administration reduced tolerance to the hypnotic effects of ethanol. • Low dosages of OT did not have any statistically significant effect on tolerance compared to saline-injected rats.   • OT administration inhibited hypothermic effects of ethanol.
Rae et al. (2018)	Swiss mice	Male	•6.4mg/kg	•Daily during and for 4 days prior to Conditioned Place Preference (CPP)	IP injection	•Alcohol CPP	• Carbetocin-induced activation of OT receptor (OTR) resulted in increased acquisition of alcohol place preference
Rigter et al. (1980)	Random-bred Swiss mice	Male	• 0.1, 1, or 10 μg	• Singular treatment given 30 minutes before ethanol or saline treatment	SC injection	• Hypothermic tolerance (body temperature measurements)	• High doses of OT enhanced the acute hypothermic effects of ethanol, but not baseline temperature.
Stevenson et al. (2017)	Adult prairie voles	Male and female	• 1.0, 3.0, and 10.0 mg/kg, 2.0 g/kg (locomotor task)	• Singular treatment • Two treatments	IP injection	• Preference for alcohol vs. water • Amount of alcohol consumed • Locomotor activity	• In a 24-hr period, alcohol consumption was reduced cumulatively and in the first hour of observation in prairie voles, but was not significantly reduced in animals with continued access during this time. • All OT doses produced similar alcohol consumption reduction in the first hour.
Szabo et al. (1985)	Random-bred CLFP albino mice	Male	• 0.1 IUa or 1 IU  • Graded doses of 2, 1, 0.5, 0.25, 0.02, or 0.002 IU	Single pretreatment	IPb injection	• Hypothermic tolerance (body temperature measurements) • Sleep onset and sleep duration • Righting reflex	• Singular dose treatments of oxytocin did not have a statistically significant effect on ethanol tolerance compared to controls. • Prior repetitive treatment of oxytocin prior to administration of ethanol prevented tolerance development to the hypothermic effects of ethanol compared to controls. • IU of OT less than 0.5 did not show this same effect.
Szabo et al. (1987)	Random-bred CLFP albino mice	Male	• Graded doses ranging from 0.02 to 2.00 IU	Daily, singular pre-treatment over the course of 3 days	SCc injection	• Withdrawal symptoms  • Mortality	• Pre-treating mice with .02 and .2 IU OT resulted in an increased incidence of seizures and elevated mortality rate. • Mice treated with 2.0 IU OT had milder withdrawal convulsions and lower mortality rates.
Szabo et al. (1989)	Random-bred CLFP albino mice	Male	• Graded doses of 3 pg, 30 pg, 300 pg, or 3 ng	• Daily, singular pre-treatment over the course of 3 days	ICVd injection	• Hypothermic tolerance (body temperature measurements)	• OT before the administration of the first dose of ethanol did not affect overall body temperature or the hypothermic response to ethanol. • Different effects of OT were observed depending on what phase of alcohol tolerance was assessed.
Tunstall et al. (2019)	Wistar and Sprague-Dawley rats	Male	• 0.125, 0.25, 0.5, and 1 mg/kg • 0.5 or 1 ml/kg	• Singular injection • Singular intranasal administration	IP injection and intranasal	• Amount of alcohol consumed • Alcohol self-administration • Locomotor activity • Grooming • Motor coordination	• OT administration removed the increased motivation for alcohol consumption in dependent rats. • OT reduced drinking in dependent rats but did not have an effect in non-dependent rats for both routes of administration.
Walcott & Ryabinin (2021)	Prairie Voles	Male and female	• 3mg/kg	• Daily injection over the course of 5 days	IP injection	• Alcohol consumption •Alcohol preference	• OT administration decreased alcohol consumption, temporarily. • OT did not impact alcohol preference

^aInternational unit,

^bIntraperitoneal,

^cSubcutaneous,

^dIntracerebroventricular

^*Placebo doses are not listed.

Dose Effects

Overall, results from the literature suggest that low dosages of OT did not influence drinking behavior and tolerance development as consistently as higher doses. These dose-dependent effects have been observed across outcomes, including both alcohol consumption and hypothermic tolerance. For example, repeated administration of OT (six doses) of 0.5, 1.0, or 2.0 IU per animal prior to administration of ethanol prevented tolerance development to the hypothermic effects of ethanol (as in, the decrease in core body temperature caused by alcohol) compared to controls. However, doses of OT less than 0.5 IU did not show this same effect (Szabo et al., 1985). Similarly, one study also showed how repeated OT administration reduced tolerance to the hypnotic effects of alcohol. However, the lowest tested dosage of 160nmol/kg did not have any statistically significant effect in changing tolerance compared to controls (saline-injected mice) (Puciklowski et al., 1985). In later work, King et al. (2017) also found evidence that doses of 3 and 10mg/kg of oxytoxin reduced ethanol consumption relative to vehicle, whereas lower doses of 1mg/kg did not.

However, these findings are somewhat inconsistent, as other studies observed effects at lower or found similar effects across doses, or mixed findings with regard to dose (e.g., Stevenson et al., 2017; Tunstall et al., 2019; MacFayden et al., 2016). For example, in bottle choice tasks, alcohol intake was reduced in about 40% of days in which OT was given in a three-bottle choice task, following doses of 0.1 mg/kg of OT (MacFadyen et al., 2016). Another study found positive effects of 1mg/kg of OT on alcohol preference (McGregor, Bowen, 2012). Further, it should be noted that, although effects are more consistently pronounced at higher doses, some variations have been observed based on the outcome under consideration.

Timing of Alcohol and OT Administration

Studies examining effects of OT on alcohol-related outcomes have been variable with regard to the timing of both OT and alcohol administration. For example, in a chronic intermittent-access model, rats treated with OT consumed less alcohol in a 24-hour drinking period after OT treatment compared to before treatment. Therefore, infusions of OT reduced the voluntary consumption of ethanol in rats (Peters et al., 2017; Peters et al. 2013). A separate study in prairie voles tested groups with continuous or intermittent access to alcohol. In a 24-hour period of a voluntary consumption model, alcohol intake was reduced cumulatively and in the first hour of observation in prairie voles but was not significantly reduced in animals with continued access. All OT doses produced similar alcohol consumption reduction in the first hour. Therefore, in this species, the largest effects were present in the first hour, but the effects were less present in the hours following (Stevenson et al., 2017).

More information is needed to better understand how the timing of OT administration influences outcomes. Some studies have suggested that single or daily pre-treatments of OT can inhibit alcohol consumption (Bowen et al., 2011; Szabo et al., 1985; McGregor & Bowen, 2012; Peters et al. 2013; MacFadyen et al., 2016; Bowen et al., 2015). However, in another study no differences in alcohol consumption were observed after daily pre-treatments of OT, but interestingly, the administration of a booster of OT following the 25-day period further inhibited alcohol consumption for the immediate 2.5 hours following (Bowen et al., 2011).

Route of Administration

Most studies included in this review used a single route of administration. However, two studies compared routes of administration. For example, one study compared intranasal administration to injection (Tunstall et al., 2019) and found that these routes of administration had similar effects. Both intranasal administration and intraperitoneal injection of OT reduced drinking in dependent rats, irrespective of if the animals were pre-treated with vehicle or peripherally restricted OT receptor antagonist. The study further demonstrated that OT concentrations in the brain were far higher from intranasal administration compared to intravenous or intraperitoneal administration. However, there were hypothesized side effects noted with intraperitoneal injection that resulted in an impaired locomotor ability that was not present with rats receiving intranasal doses of OT (Tunstall et al., 2019). In a separate study, mice demonstrated reduced stress-induced alcohol intake following exposure to OT administered through IP, but not through ICV administration. Collectively, these findings highlight the importance of comparing routes of administration (Peters et al. 2013).

Species Effects

Many preclinical studies in this review focused on mice and rats (9 studies each), with fewer studies examining effects in prairie voles (2 studies). Comparison across species is challenging due to differences in size, body composition and social characteristics. However, despite these differences, positive of OT on alcohol related outcomes have been observed across species, with some nuances. Interestingly, in prairie voles, oxytocin reduced alcohol consumption, but these effects were influenced by both sex and the model of alcohol access (i.e., continuous vs. intermittent access; Stevenson et al., 2017). Additionally, another study in prairie voles found effects for alcohol consumption but not preference (Walcott & Rybanin, 2021). Collectively, these findings highlight the importance of replicating findings of OT across species.

Sex Differences

The majority of preclinical studies have been conducted on male animals. However, a small number of studies have explored whether effects of oxytocin vary by sex. While most studies suggest that oxytocin generally reduces alcohol administration across male and female animals (Caruso et al., 2021; King & Becker, 2019; Stevenson et al., 2017; Walcott & Ribanin, 2021), some sex-dependent effects have been observed. For example, although oxytocin reduced reinstatement of alcohol following exposure to stressors (i.e., predator odor, yohimbine) in both males and females, these effects were only present at the highest doses in males and induced at a lower dose in females (King & Becker, 2019). In another study, in a chronic intermittent access model – but not a continuous access model – oxytocin yielded greater reductions in alcohol consumption in male relative to female prairie voles (Stevenson et al., 2017). These results highlight that more research examining sex differences in male and female animals is needed.

Other effects of OT

OT may also impact the acute effects of alcohol, including motor impairment. Also, OT administered intracerebroventricularly immediately prior to ethanol dosing resulted in the attenuation of acute alcohol effects in a motor impairment test. Some of the effects of interest were sedative impairments, reflex, and overall locomotive capacity. However, OT did not improve motor skills following the consumption of higher quantities of ethanol (3g/1kg) (Bowen et al. 2015). Similar patterns were observed by Tunstall (2019), who did not note significant differences in motor coordination between dependent vs. nondependent rats, along with no significant deviations in performance following intranasal or intraperitoneal oxytocin administrations Tunstall (2019).

Body temperature measurements were also assessed by measures of hypothermic tolerance following alcohol administration. Results varied depending on OT dosage, but regular, repeated dosages of OT blocked the development of tolerance to the hypothermic effects of alcohol (Szabo et al., 1985; Puciklowski et al., 1985). This was not observed in singular pretreatments, where OT did not affect overall body temperature of hypothermic tolerance (Szabo et al., 1989; Rigter et al., 1980).

Conclusions from Preclinical Studies

In sum, OT appears to reduce alcohol self-administration, reduce ethanol-induced place preference, and attenuate motor effects caused by alcohol. However, more work is needed to better understand how results are influenced by dose, sex, species, experimental paradigm, and route of administration. The literature presented here suggests that OT is a promising agent worthy of further study but suggests that further investigation regarding the influence of each of these factors is an important area for future study.

Clinical Studies

Eight clinical studies, presented in Table 2, were included in this review. These studies also show the OT may have promising results depending on factors such as dose and frequency of administration. As with preclinical studies, the manner of dosing in clinical studies may impact the effectiveness of OT. Lower doses of OT given several times per day may not be sufficient to influence alcohol craving or use in human subjects (Table 2). In one study, participants who received 8 IU of OT administered three times daily, during a 25-day extension period did not experience significant differences in anxiety, depression or alcohol cravings relative to the placebo group (Melby et al., 2021). It is difficult to make firm conclusions with only one study with these dosing parameters; however, future studies may benefit from examining the optimal dosing of OT.

Table 2.

Clinical Studies on the Effects of OT on Alcohol-Related Behaviors

Source	Study population	Sex	Number of subjects		Length of study	OT treatment and dose amount(s)*	OT dose frequency	Route(s)	Main outcome measure(s)	Main findings
Bach et al. (2019)	Social drinkers from Germany	Male	13	2 sessions over 2 weeks		Treatment (24 IU/dose) on the first or second study session (placebo given before other session) with no prior abstinence required.	• Single dose	Intranasal	• Functional connectivity analysis • Cue-induced alcohol craving	• OT reduces alcohol-cue decreased connectivity between both the NAc and cuneus as well as the thalamus and lOC • OT enhances connectivity between the precentral gyrus and paracingulate gyrus.
Hansson et al. (2018)	Non-treatment-seeking heavy social drinkers from Germany	Male	12	1 day		Treatment (24 IU/dose) initiated on the day of study admission with no prior abstinence required.	• Single dose	Intranasal	• Cue-reactivity	• In an fMRI scan assessment, OT administration was associated with reduced cue-reactivity in the IC, hippocampal formation, Cg, inferior and medial frontal gyrus, and in visual and motor regions compared to placebo. • There was no observed increase in cue-reactivity when treated with OT.
Flanagan et al. (2019)	Veterans who met criteria for a DSM-5 diagnosis for PTSD and AUD from the USA	Male	67	1 day		Treatment (40 IU/dose) initiated on the day of study admission with no prior abstinence required.	• Single dose	Intranasal	• Cortisol reactivity • Self-reported alcohol craving	• Veterans in the OT group experienced less general cortisol reactivity in the stress task compared to placebo. • There was no significant difference in craving noted between the OT group and placebo group.
Melby et al. (2021)	Patients aged 18–65 years fulfilling the ICD-10 criteria for a diagnosis of alcohol dependence from Norway	Male and female	38	25 days		Treatment (8 IU/dose) initiated after inpatient withdrawal treatment.	• A maximum of thrice daily for 25 days	Intranasal	• Amount of alcohol consumed • Number of days to alcohol use • Proportion of subjects returning to use •Alcohol craving	• There were no significant differences in alcohol craving or consumption between the OT group and the placebo group. • There were no significant differences in time to alcohol use or proportion returning to use between the OT group and the placebo group.
Mitchell et al. (2016)	Persons meeting DSM-IV criteria for alcohol abuse	Male and female	32	2 sessions, 1 week apart		Treatment (40 IU/dose) on the first or second study session (placebo given before other session)	•Single dose	Intranasal	• Cue-induced alcohol craving	•Oxytocin reduced alcohol craving in persons with anxious attachment, and increased craving in those without anxious attachment.
Pedersen et al. (2013)	Alcohol-dependent inpatient subjects currently undergoing detoxification from the USA	Male and female	11	3 days		Treatment (24 IU/dose) initiated on the first day of study admission with no prior abstinence required.	• Twice daily for 3 days	Intranasal	• Total amount of lorazepam required to complete alcohol detoxification • Alcohol withdrawal	• Participants randomized to the OT group required less lorazepam for withdrawal symptom treatment. • Participants in the OT group experienced less alcohol withdrawal symptoms compared to participants taking placebo.
Stauffer et al. (2019)	Patients with comorbid diagnoses of PTSD and AUD from the USA	Male	83 (47 with AUD and PTSD, 36 controls)	3 sessions over a minimum of 3 weeks		Treatment (20 or 40 IU/dose) initiated on the day of study admission with no prior abstinence required.	• 3 doses	Intranasal	• Self-reported cue-induced alcohol craving • Heart rate following cue exposure	• There was no significant effect of OT on alcohol craving or heart rate.
Vena et al. (2018)	Healthy social drinkers from the USA	Male and female	35	2 sessions over a minimum of 1 week		Treatment (20 and 40 IU/dose) initiated following 24 hrs of abstinence.	• Single dose of each IU/dose	Intranasal	• Subjective alcohol response • Behavioral tasks • Self-reported alcohol craving	• OT did not affect subjective alcohol responses or performance on any behavioral tasks. • A single dose of OT did not affect craving.

Note: Placebo dosages are not listed.

Additionally, singular intranasal doses of OT (or doses administered several days apart) did not affect alcohol craving (Table 2). For example, a study of male US veterans with diagnosed PTSD and AUD assessed the role of a single administration of intranasal OT on alcohol craving in response to stress. However, there was no significant difference in craving noted between the OT group and placebo group in this population (Flanagan et al., 2019; Stauffer et al., 2019). Similarly, a trial of social drinkers from the USA received single doses of 20 and 40 IU and did not note statistically significant differences in subjective feelings of craving compared to controls (Vena et al., 2018). These null findings were observed despite other work suggesting that a single dose of 24/IU of oxytocin reduced connectivity between the NAc and ceneus, and enhanced connectivity between the precentral cyrus and paracingulate gyrus (Bach et al., 2019) and effects on cortisol reactivity (Flanagan et al., 2019).

More promising findings were observed in the context of repeated, higher doses in the context of alcohol withdrawal. For example, a randomized double-blind trial assessed the effect of intranasal OT compared to placebo in eleven alcohol-dependent participants. Participants randomized to the OT group required less lorazepam for withdrawal symptom treatment and experienced less alcohol withdrawal symptoms compared to participants taking placebo (Pedersen et al., 2013). Future work should extend these findings in larger samples.

A pilot clinical trial was conducted to assess the ability of intranasal OT to potentially reduce neural reactivity to alcohol-related cues in a sample of 12 male heavy social drinkers. In an assessment of fMRI scans, those in the OT group had reduced cue-reactivity in the IC, the hippocampal/parahippocampal formation, the Cg, the inferior and the medial frontal gyrus, and invisual and motor regions compared to placebo. There was no increase in cue-reactivity observed for those in the OT group (Hansson et al., 2018).

Ongoing Clinical Trials

Current preclinical literature suggests that OT has potential for modulating alcohol-related behaviors such as behavioral response, craving, and withdrawal. The clinical literature has been less consistent but has found some promising findings in a sample that included three days of twice daily dosing (rather than a single dose), as well as hypothesized effects on cortisol reactivity and neural connectivity. OT is a promising candidate worthy of further clinical investigation because it has an excellent safety profile and low misuse potential, with the most common adverse effects being light dizziness and headache or nasal irritation (Lee & Weerts, 2016; MacDonald et al., 2011). Further, most individuals in intranasal OT trials have indistinguishable adverse effects for the receipt of OT versus placebo, suggesting that there are minimal risk factors associated with intranasal OT administration (MacDonald et al., 2011; Hansson et al., 2018; Melby et al., 2021; Vena et al., 2018).

There are multiple clinical trials currently investigating the role of OT as a potential therapeutic for AUD and heavy drinking behavior. Of the 20 clinical trials registered at the time of writing this review, seven have been completed with early results publicly available (NCT02251912, NCT02275611, NCT03610633, NCT02058251, NCT02742532, NCT01212185, NCT01829516). A Phase II multi-site RCT assessing the role of intranasal OT for AUD treatment led by the Clinical Investigations Group (NCIG) within the National Institute on Alcohol Abuse and Alcoholism (NIAAA) (NCT03878316) will also test effects of oxytocin. Since its development in 2007, NCIG has successfully completed six Phase II multi-site RCTs on potential medications to treat alcohol dependence. These trials have been conducted on drugs such as Levetiracetam, Varenicline, Quetiapine, and Gabapentin enacarbil extended release, and a seventh trial on the use of intranasal OT as a drug for AUD treatment is currently recruiting (National Institute on Alcohol Abuse and Alcoholism [NIAAA], n.d.).NCIG trials address the infrequent control for AUD severity in clinical trials by requiring a moderate level of AUD or higher at baseline, which entails the presentation of at least five symptoms (APA, 2022). Outcome measures of interest include percentage of heavy drinking days as opposed to days of complete abstinence, which may be more accurate in capturing drinking reduction in individuals who do not wish to be completely abstinent.

Conclusions

Preclinical studies have generally found promising – albeit nuanced - effects of oxytocin on reducing alcohol self-administration and physiological alcohol effects, with variations across doses and other study parameters. Results from clinical studies have been more mixed, though some promising findings on alcohol withdrawal and neural cue reactivity. At present, however, differing factors between these two types of trials can make comparisons difficult. For example, apart from Tunstall et al. 2019, all preclinical studies used ICF, IP, or SC injection as the route of administration, whereas all clinical OT administration involved intranasal administration. Doses and dose units varied within preclinical studies. We recommend that future studies report doses in a standardized format, including both the total dose administered and the dose per kilogram of body weight to increase the comparability of dosing regimens across species and studies. Additionally, although it is possible to make interspecies dose conversions (see Nair & Jacob, 2016), variances in body mass and composition across species further create difficulties with comparing across different animal models, and when comparing animals and humans. Ongoing clinical studies will add to the burgeoning literature in this area and may help better contextualize factors that influence potential beneficial effects of oxytocin.

Future Directions

Future and ongoing clinical trials in this area should address several important areas to learn more about the OT system and its applications for alcohol use treatment. Much is still unknown regarding the mechanisms by which OT may modulate alcohol craving and withdrawal. For this reason, more clinical and preclinical studies that test additional OT dose amounts in different time periods are necessary. Because most of the clinical studies discussed in this review took place within a short time frame of one day or single days spread out weeks apart, there is also a need for more trials where OT is administered over a continuous period. This is essential for better capturing long-term effects and any adverse reactions that may accompany prolonged use.

Furthermore, most of the studies discussed in this review focused on male populations, both clinically and preclinically. Others included both male and female participants but did not include analyses examining sex differences in response to oxytocin response. It is essential that future studies include both male and females and examine sex differences, as sex-specific neural effects have been observed (Dumais et al., 2017). Future studies should also incorporate more diverse study populations into trials, both in terms of biological sex as well as other sociodemographic variables. Increasing the number of female study participants across trials is critical because of sex differences in the OT system, making it important to measure any sex-specific effects for the application of OT as a potential alcohol use therapy. Despite these limitations, existing trials suggest that OT remains a promising target focus for investigation as a medication to treat AUD and heavy drinking.

Public Significance Statement.

This review suggests that additional investigation is needed to better understand the effects of oxytocin in clinic populations engaged in heavy drinking. However, main findings from current trials reveal that oxytocin represents a promising medication for continued investigation of alcohol use disorder treatment.