The 5‐HT2C receptor as a therapeutic target for alcohol and methamphetamine use disorders: A pilot study in treatment‐seeking individuals

Abstract Alcohol use disorder (AUD) and methamphetamine use disorder (MUD) are prevalent and have high adverse impacts on both the individual and society. Current treatment strategies for these disorders are ineffective at a population level. Lorcaserin, a 5‐HT2C receptor agonist, has shown potential at reducing the symptoms of substance use disorder. This pilot study (initiated prior to market withdrawal) examined feasibility and safety of lorcaserin treatment in people undergoing residential detoxification and treatment for AUD and MUD. This was an open label pilot study of lorcaserin where participants (n = 10 AUD; n = 8 MUD) received 10‐mg lorcaserin daily for 4 days then twice daily for 1 month. Primary outcome measures included recruitment and retention rate, incidence of treatment‐emergent events, incidence of methamphetamine or alcohol withdrawal‐related events, heart rate, and blood pressure. Secondary measures included pharmacokinetic data and self‐reported alcohol or methamphetamine use, craving, and psychological distress. AUD participants were recruited faster and had a greater retention rate compared with MUD participants. Lorcaserin did not alter vital signs, was well tolerated, and had a similar pharmacokinetic profile to individuals with obesity. Lorcaserin reduced self‐reported alcohol and amphetamine‐type substance use and craving in AUD and MUD participants, respectively. Self‐reported psychological health also improved over the treatment period for all participants. Despite the pilot nature of this study, our data support the notion of 5‐HT2C receptors as a therapeutic target for drug and alcohol abuse.


| INTRODUC TI ON
Alcohol use disorder (AUD) poses a major social and economic burden to society, accounting for ~5% of deaths worldwide in 2016 and costing upward of $36 billion/year in Australia. 1 With regard to illicit drug use, methamphetamine is the second most regularly used drug following cannabis in Australia, and in the United States, there was a threefold increase in the number of methamphetamine-associated deaths between 2010 and 2015. 2,3 As a result, the recreational use and abuse of drugs continues to be a major global public health issue.
Importantly, poor retention and frequent relapse remain serious obstacles for the treatment of substance use disorders with continuous and intense cravings persisting both during and following treatment. 4,5 There are three Food and Drug Administration (FDA)approved pharmacotherapies for AUD, disulfiram, naltrexone, and acamprosate. Current pharmacotherapeutic treatments for AUD remain ineffective at a population level due to a combination of limited effectiveness and under-prescribing. 6,7 There are no FDA-approved pharmacotherapies specifically for methamphetamine use disorder (MUD). In a recent review, most medications evaluated were found to not have a statistically significant benefit. 8 Clearly, there is need for improved pharmacotherapeutic treatments aimed at reducing drug-related craving for both AUD and MUD.
The neural circuitry surrounding craving and relapse to drug use involves several brain structures including the ventral tegmental area, striatal complex, amygdala, hippocampus, and the prefrontal cortex. 9,10 Serotonin-containing cells (5-hydroxytryptamine  are located predominantly in the raphe nuclei of the mid/hind-brain, including the dorsal and median raphe. 11 From here, 5-HT fibers widely innervate the central nervous system including multiple nodes of reward-related circuitry. 12 Indeed, the serotonergic system has been implicated in substance use disorder and relapse for several decades and may represent an avenue for future pharmacological interventions. [13][14][15][16][17][18] There are 14 known subtypes of 5-HT receptors, and serotonin signaling is well known to modulate dopamine activity. 19 Specifically, the 5-HT 2C receptor is expressed in the hippocampus, striatum, and amygdala in both rat and human. [20][21][22] 5-HT 2C receptor signaling has been implicated in the development and maintenance of AUD and MUD. For example, the 5-HT 2C receptor agonist, Ro60-0175, decreased alcohol consumption in rats whereas the 5-HT 2C receptor antagonist, SB-242 084, increased alcohol consumption. 23 However, Ro60-0175 reduced both alcohol (gel solution) and vehicle (plain gel containing polycose) operant self-administration in rats suggesting possible nonspecific effects on caloric intake. 24 5-HT 2C receptor signaling is also involved in methamphetamine use; for example, Ro60-0175 reversed methamphetamine self-administrationinduced decreases in nucleus accumbens shell excitability. 25 Additionally, methamphetamine-induced behavioral sensitization is associated with a functional upregulation of 5-HT 2C receptors in the ventral pallidum. 26 Together, these preclinical studies highlight a role for the 5-HT 2C receptor in both alcohol and methamphetamine use.
The development of therapeutic drugs that selectively target individual 5-HT 2 receptor subtypes is difficult. Indeed, most 5-HT 2C receptor agonists also bind to 5-HT 2A and/or 5-HT 2B receptors. 27 Lorcaserin is a selective serotonin 2C (5-HT 2C ) receptor agonist with a 3-benzazepine scaffold, 28 developed as an anti-obesity medication. Notably, lorcaserin reduces the consumption of alcohol in rats and methamphetamine use in rhesus monkeys, 29,30 implicating the 5-HT 2C receptor as a potential treatment target for alcohol and substance use disorders. 15 Until recently, lorcaserin (Belviq ® ) was FDAapproved as an anti-obesity medication 31 but was withdrawn from the market after a safety trial indicated an increased occurrence of cancer, where 7.7% of participants receiving drug developed cancer relative to 7.1% in the placebo arm. 31 Here, we carried out a pilot study, prior to market withdrawal, to evaluate the ability of lorcaserin to suppress alcohol and methamphetamine craving and consumption in treatment-seeking AUD or MUD participants.

| Study design and participants
This was an open label pilot study. The protocol and amendments were approved by the Human Research Ethics Committee of St Vincent's Hospital Melbourne (HREC 031/17). Eligible participants were males and females over the age of 18 years, diagnosed with alcohol or methamphetamine substance use disorder (DSM5).
Exclusion criteria included pregnant (urine βHCG positive) or breastfeeding; highly dependent on medical care for co-existing conditions; other medical treatments for substance dependence including anti-craving (e.g., acamprosate and naltrexone), aversive (e.g., disulfiram), or substitution (e.g., atomoxetine, dexamphetamine, and methylphenidate) treatments; known allergy to lorcaserin; already receiving lorcaserin; severe liver impairment (Child Pugh C); severe renal impairment (creatinine clearance <30 ml/min); hypertension; unstable diabetes; history of serotonin syndrome; low body mass index (BMI <20); and unstable mental state (including active psychosis or schizophrenia). Written informed consent was obtained from all participants.
The study protocol specified the recruitment of 10 AUD participants and 10 MUD participants. Recruitment was ceased when the initial FDA alert was issued in January 2020, resulting in the final recruitment of 10 alcohol-and 8 methamphetamine-dependent participants. Note that the HREC was advised immediately by the Chief Investigators of the original FDA alert and the subsequent product withdrawal. All participants were advised in writing of the product withdrawal and the health risks associated with their participation in the study.

| Procedures
Participants were treated with the lowest effective dose of lorcaserin (immediate release) used in the treatment of obesity.
Participants received lorcaserin 10 mg once daily for 4 days then twice daily for 1 month. All participants received symptomatic treatment using the St Vincent's Hospital standard protocol.
Withdrawal was actively managed as either an outpatient or, if required, in a residential setting. Alcohol withdrawal was managed in accordance with Clinical Institute Withdrawal Assessment of Alcohol (CIWA-Ar). 32 Prescribed medications consisted of 5-20 mg of diazepam 2 hourly with a maximum of 80 mg per 24-h period, anti-emetics for nausea, and paracetamol and nonsteroidal anti-inflammatories for aches. All medications were used only as required and ceased by Day 7 following participants' last use of alcohol or methamphetamine and the commencement of lorcaserin.  The Obsessive Compulsive Drinking Scale was used to assess alcohol craving. 33 A total craving score was calculated by summing the 14 items of this questionnaire, which were ranked on a Likert scale ranging from 1 to 7. Summing Items 1-6 calculated the obsessive subscale, and summing Items 7-14 calculated the Compulsive Subscale. A brief, 10-item Methamphetamine Craving Questionnaire, based on the Cocaine Craving Questionnaire, was used to assess methamphetamine craving over time. Scores across each item were averaged. 34,35 The Kessler Psychological Distress Scale (K10) yielded a global measure of distress, which was the sum of all 10 items.

| Pharmacokinetic sampling
The Australian Treatment Outcomes Profile (ATOP), a validated Australian version of the UK Treatment Outcome Profile, was used to assess self-reported drug use and health and well-being. 36 Higher scores on the substance use questions, measured using the timeline follow-back method, reflected more days of use whereas higher scores on health and well-being questions indicated greater selfrated health outcomes.

| Outcomes
Primary endpoints were prespecified. Feasibility endpoints were recruitment rate and retention in treatment at Days 7, 14, 21, and 28. Safety endpoints were incidence of treatment-emergent adverse events (AEs); incidence of methamphetamine or alcohol withdrawal-related treatment-emergent events; heart rate; and blood pressure.
Secondary endpoints were methamphetamine use (self-report, saliva screen, and urine drug screen) on Study Days 0, 7, 14, 21, and 28; alcohol use (self-report, breath alcohol, blood testing, and urine

| Statistical analysis
Unless otherwise stated, data were analyzed separately for AUD and MUD participants.
Demographic data were summarized as numbers and proportions for categorical data and mean (±standard error of mean pants had data missing from the health and well-being questions and were excluded from this analysis. Only two to three participants with MUD completed the ATOP self-report questionnaire for the duration of the study (baseline to Day 28); thus, data were also analyzed from baseline to Day 14, with five participants completing the questionnaire at these time points. All analyses were performed using SPSS v27 (α = .05). Data are presented as mean ± SEM.

| Nomenclature of targets and ligands
Key protein targets and ligands in this article are hyperlinked to corresponding entries in http://www.guide topha rmaco logy. org, the common portal for data from the IUPHAR/BPS Guide to PHARMACOLOGY, 38 and are permanently archived in the Concise Guide to PHARMACOLOGY 2019/20. 39

| Demographic information
The participants recruited with AUD were predominantly male   was 87.5%, 62.5%, 37.5%, and 50%, respectively. The main reason for participants discontinuing the trial was due to failing to attend appointments. Note that one patient missed Day 21 appointment but represented for Day 28, and this accounts for the dip in the retention rates over time at Day 21. Each was deemed "lost to followup" after multiple unsuccessful attempts by a clinician to contact the participant.

Treatment-emergent AEs
Eight AUD participants (80%) reported at least one AE, and seven MUD participants (87.5%). Of these, seven of the AUD participants (87.5%), and five MUD participants (71.4%), reported at least one AE that was considered by a study doctor to be related to lorcaserin.
These included ratings of definitely related, probably related, or possibly related ( Table 2). Of all AEs (n = 28), 19 were considered to be treatment emergent (67.9%).
The most frequently occurring treatment-emergent AEs were

TA B L E 1 (Continued)
"severe." One was classified as "moderate" severity. This event of headache reported at Day 21 lasted 2 days and resolved spontaneously. The remaining 18 treatment-emergent AEs were classified as "mild" severity. None of the treatment-emergent AEs was a serious AE.

Withdrawal-related treatment-emergent AEs (n = 1)
One event was "probably related" to lorcaserin. An AUD participant experienced ongoing fatigue. They dropped out at Day 9, and no further follow-up by the study staff was achieved.

Vital signs
For participants with AUD, temperature, pulse, systolic and diastolic blood pressure, and weight did not significantly change over time (Table S1). Respiratory rate (breaths/min) significantly increased over the treatment period (Table S1). For participants with MUD, temperature, pulse, respiratory rate, systolic and diastolic blood pressure, and weight did not significantly change over time (Table S2).

Alcohol use
Self-report. For the AUD group, the number of days drinking alcohol in the past week did not change between baseline and Day 28 (Table 3); however, the typical quantity of alcohol consumed on any given day did decrease over time (Table 3).
Clinicalcharacteristics. For the AUD group, breath alcohol content, liver function, blood glucose levels, and urinary drug screen results did not change over the treatment period (Table S3).

Methamphetamine use
Self-report. For the MUD group, from baseline to Day 14, the number of days using an amphetamine-type substance in the past week did not change (Table S4). From baseline to Day 28, the number of days using an amphetamine-type substance in the past week decreased (Table 3).
Clinicalcharacteristics. For the MUD group, liver function, blood glucose levels, urinary drug screen results, and oral fluid test results did not change over time (Table S5).

| ATOP self-report health and well-being data
For the AUD participants, psychological health status and quality of life significantly increased across the experimental period between baseline and Day 28, and physical health status approached a significant improvement (Table 3; Figure 1E).
For the MUD participants, from baseline to Day 14, the perceived health and well-being in the past week did not change (Table S4).
F I G U R E 1 Self-reported craving, psychological distress, and health and well-being after lorcaserin treatment in participants with alcohol use disorder (AUD) and methamphetamine use disorder (MUD). Total obsessive compulsive drinking score over the 28-day treatment period. Total n = 8. **p < .01 for each craving measure (A). Average craving score over the 28-day treatment period. n = 3-5. *p < .05 baseline to Day 14 (B). Kessler Psychological Distress self-report data over the 28-day treatment period. Total n = 8 for AUD participants, n = 3-5 for MUD participants. ***p < .001 (C and D). Psychological health status and quality of life increased over the treatment period; however, there was no significant change in physical health status for AUD participants. Total n = 8. **p < .01 for psychological health status and quality of life (E). Psychological health status increased over the treatment period; however, there was no significant change in physical health status or quality of life. n = 2-5. ***p = 0.001 for psychological health status (F). All data are presented as mean ± SEM From baseline to Day 28, the psychological health status in the past week increased; however, physical health status and quality of life did not change (Table 3; Figure 1F).

| Pharmacokinetics of lorcaserin
Plasma concentration time profiles (0, 2, 4, 8, and 12 h) for six participants are shown in Figure 2. The pharmacokinetic profile followed an oral absorption profile ( Figure S1) with a T max at 2 h and a C max of 102 ± 25 ng/ml before following a one phase exponential decay, resulting in a t 1/2 of 7.9 ± 2 h. Pharmacokinetic parameters are summarized in Table 4.

| DISCUSS ION
The aim of our pilot study was to examine the safety and efficacy of the 5-HT 2C receptor agonist lorcaserin in people undergoing residential withdrawal and seeking treatment for AUD and MUD.
There are limited data on the pharmacokinetics of lorcaserin, and in general, this has been observed in an obese population with a T max of 1.5 h, C max of 80 ng/ml with a variability of 26%, and a t ½ of 11.9 ± 1.5 h. 40 In AUD and MUD participants, we observed a slightly longer T max of 2.4 h, a higher C max of 102 ng/ml with similar variability of 26%, and a shorter t ½ of 7.9 ± 2 h. With the small sample sizes and similar variability of approximately 30% in both studies, it is unlikely that there is a different pharmacokinetic profile between an obese and AUD/MUD populations.
In the present study, AUD participants were recruited faster and had a greater retention rate compared with MUD participants.
There were no significant changes in vital signs over the treatment period except respiratory rate did increase for AUD participants.
Importantly, lorcaserin was well tolerated, having few, if any, side effects. Lorcaserin was not anorectic and had no cardiovascular side effects. We also tested the effect of lorcaserin on alcohol and Our findings are also somewhat consistent with preclinical literature showing reductions in alcohol consumption in rats following chronic lorcaserin treatment. 30 Lorcaserin can also decrease selfadministration of methamphetamine and cocaine in rhesus monkeys. 29 Although lorcaserin was recently withdrawn from the market following a safety concern, we contend the involvement of 5-HT 2C receptors in substance use disorder are relevant and targeting these receptors is still a viable avenue for treatment, particularly given our current findings and those discussed above. The 5-HT 2C receptor has been implicated in both compulsive-and impulsive-like behaviors in rodents, cardinal features of substance use disorder. 53 Additionally, 5-HT 2C receptor activity can modulate dopamine function. 54,55 Of note however is the complexity of the 5-HT system, particularly regarding the interaction between 5-HT 2C and 5-HT 2A receptors and their relationship to impulsive-like behavior. 56

| Methodological considerations and future research
A methodological limitation of the current study is that all participants received lorcaserin treatment; that is, there was no placebo control. Additionally, this study was open label, limiting the interpretation of our findings. Although overall sample size was small, particularly for MUD participants, those that completed the study seemingly benefitted from lorcaserin. The differential outcomes between AUD and MUD participants go some way to mitigate against these shortfalls; however, further studies are clearly warranted.
MUD participants were more difficult to retain and had lower retention in treatment. The observed difficulty in recruiting and maintaining MUD participants in the trial could possibly be because methamphetamine users seek treatment reluctantly due to a lack of perceived effective treatments. It could also simply reflect that there are more people with AUD than MUD. There was no difference in the rates of AEs between the two groups, so this is unlikely to be a factor. It is difficult to draw firm conclusions due to the small number of participants in the study. Any future study should include design elements to improve recruitment rates and mitigate against attrition. For example, contingency management approaches have been shown to be effective in the treatment of many substance use disorders and hold promise for MUD. 57,58

| CON CLUS ION
In summary, our data suggest that the 5-HT 2C system is still a viable treatment option for substance use disorders, particularly AUD. However, given the current FDA withdrawal of lorcaserin, other 5-HT 2C agonists with high selectivity may be beneficial.
Alternatively, due to the conserved nature of the orthosteric binding site for serotonin at 5-HT 2 family receptors, selective positive allosteric modulators, analogous to those for acetylcholine muscarinic receptors, 59 may provide a fruitful line of enquiry and enable more selective targeting. Another promising therapeutic target may be the use of psychedelics, targeting the serotonin system, for substance use disorder treatment. 60,61 However, animal studies have recently questioned the efficacy of psychedelics in models of AUD. 62 Nevertheless, our findings add to the literature supporting the 5-HT 2C receptor as a therapeutic target for AUD and MUD.

ACK N OWLED G M ENT
We acknowledge the Victorian State Government Operational Infrastructure Scheme.

D I SCLOS U R E
The authors report no conflict of interest.

AUTH O R CO NTR I B UTI O N S
AJL and YB conceived and designed research; YB, AP, LC, and AN collected the data; EJC analyzed data; PG and JJ conducted and analyzed pharmacokinetic data; EJC and AJL interpreted data and wrote the first draft of the manuscript; all authors contributed to, edited, and approved the final manuscript.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request. Some data may not be made available because of privacy or ethical restrictions.