Evaluating the Abuse Potential of Lenabasum, a Selective Cannabinoid Receptor 2 Agonist

Abstract

Endocannabinoids, which are present throughout the central nervous system (CNS), can activate cannabinoid receptors 1 and 2 (CB1 and CB2). CB1 and CB2 agonists exhibit broad anti-inflammatory properties, suggesting their potential to treat inflammatory diseases. However, careful evaluation of abuse potential is necessary. This study evaluated the abuse potential of lenabasum, a selective CB2 receptor agonist in participants (n = 56) endorsing recreational cannabis use. Three doses of lenabasum (20, 60, and 120 mg) were compared with placebo and nabilone (3 and 6 mg). The primary endpoint was the peak effect (E_max) on a bipolar Drug Liking visual analog scale (VAS). Secondary VAS and pharmacokinetic (PK) endpoints and adverse events were assessed. Lenabasum was safe and well tolerated. Compared with placebo, a 20-mg dose of lenabasum did not increase ratings of Drug Liking and had no distinguishable effect on other VAS endpoints. Dose-dependent increases in ratings of Drug Liking were observed with 60 and 120 mg lenabasum. Drug Liking and all other VAS outcomes were greatest for nabilone 3 mg and 6 mg, a medication currently approved by the US Food and Drug Administration (FDA). At a target therapeutic dose (20 mg), lenabasum did not elicit subjective ratings of Drug Liking. However, supratherapeutic doses of lenabasum (60 and 120 mg) did elicit subjective ratings of Drug Liking compared with placebo. Although both doses of lenabasum were associated with lower ratings of Drug Liking compared with 3 mg and 6 mg nabilone, lenabasum does have abuse potential and should be used cautiously in clinical settings.

SIGNIFICANCE STATEMENT

This work provides evidence that in people with a history of recreational cannabis use, lenabasum was safe and well tolerated, although it did demonstrate abuse potential. This work supports further development of lenabasum for potential therapeutic indications.

Introduction

The endocannabinoid system is widely distributed throughout the central nervous system (CNS) and is involved in CNS development and activity, synaptic pruning, and immune response. Endogenous and exogenous cannabinoids activate receptors within the endocannabinoid system, primarily cannabinoid receptors 1 and 2 (CB1 and CB2). CB1 is mainly found throughout the central nervous system, whereas CB2 is located throughout the peripheral nervous system and immune cells. Delta-9-tetrahydrocannabinol (THC) and cannabidiol (CBD) are the primary exogenous cannabinoids found in cannabis, both acting on the endocannabinoid system in distinct ways. Existing work suggests that THC has a high affinity for CB1, mediating its psychoactive, reinforcing, and abuse-related effects and partial agonist activity at CB2 receptors. Although still evolving, additional work suggests that CBD is a partial CB2 agonist with allosteric modulation of CB1 and complex downstream activation of G-protein–coupled receptors (Laprairie et al., 2015; Lu and Mackie, 2016; Krishna Kumar et al., 2019; Shahbazi et al., 2020; Haney, 2022; Martinez Naya et al., 2023). CB2 activation is associated with reductions in inflammation in animal models (Rom and Persidsky, 2013; del Río et al., 2016).

Given the ubiquity of endocannabinoids in the CNS, scientific inquiry has been focused on identifying potential therapeutic uses for drugs acting at cannabinoid receptors (Baratta et al., 2022) (http://www.ncbi.nlm.nih.gov/books/NBK423845/). Despite documented use for medicinal purposes for thousands of years, regulatory obstacles and concerns about potential psychoactive effects have made it difficult for researchers to study the potential therapeutic effects of this class of drugs in experimental trials (Russo, 2007; Pisanti et al., 2017; Haney, 2020). Still, existing research supports the role of cannabinoids in treating chemotherapy-induced nausea and vomiting (nabilone and dronabinol), loss of appetite and weight loss in individuals with AIDS (dronabinol), and seizures associated with Lennox Gastaut and Dravet syndromes (CBD).

Prior work broadly suggests anti-inflammatory properties of CB2 and, to a lesser extent, CB1 receptor agonists, providing support for its use in disorders that involve chronic inflammation (Burstein, 2015; Pertwee, 2015; Turcotte et al., 2016) and inflammatory pain (Clayton et al., 2002). Cystic fibrosis, systemic sclerosis, and dermatomyositis are disorders that produce chronic inflammation, each with characteristic clusters of symptoms triggered and sustained by autoimmune responses (Marvi et al., 2012) (https://www.ncbi.nlm.nih.gov/books/NBK464183/; https://www.ncbi.nlm.nih.gov/books/NBK430875/). As there is no cure for these disorders, current treatments target individual symptoms, usually with a variety of antibiotics, corticosteroids, and immunosuppressive agents. Therefore, a medication that can safely address the varied underlying mechanisms of chronic inflammation may help improve quality of life and outcomes for individuals with these conditions.

Although the anti-inflammatory properties of CB1/CB2 agonists are thought to be peripherally mediated, these can also be centrally acting compounds, especially at the CB1 receptor. Therefore, it is important to evaluate their abuse potential in understanding their risk/benefit profile. Abuse potential can be defined as the tendency of a medication or drug to be used in a nonmedical or nontherapeutic manner to experience a psychologically or physiologically positive effect (https://www.fda.gov/media/116739/download). As defined by the Drug Enforcement Agency (DEA) in the United States, cannabis is a Schedule I substance considered to have “no currently accepted medical use and high potential for abuse” (https://www.dea.gov/drug-information/drug-scheduling). Although there have been recent efforts to reschedule cannabis, medications or agents that are structurally or pharmacologically similar to Schedule I substances necessitate careful evaluation of psychoactive properties and abuse potential. The US Food and Drug Administration (FDA) issued guidelines in 2017 for industry-funded studies examining the abuse liability of novel agents (https://www.fda.gov/media/116739/download). These guidelines pertain to medications or compounds that affect the central nervous system, produce psychoactive effects, or are chemically or pharmacologically similar to other drugs with known abuse potential.

The present study evaluated the abuse potential of lenabasum, a selective CB2 receptor agonist with 12-fold lower affinity for CB1 receptors and purportedly low psychoactive properties due to low CNS penetration (Tepper et al., 2014; Burstein, 2021). Lenabasum is the 9-carbon 1,1-demethylheptyl side-chain analog of THC-11-oic acid, THC’s terminal metabolite of tetrahydrocannabinol (THC). The effects of lenabasum on chronic inflammation and fibrosis appear to be a result of activation of CB2 receptors, which leads to the production of specialized proresolving lipid mediators, specifically increases in inflammation-resolving prostaglandins PG-D₂ and PGJ (Burstein, 2021). Selection of lenabasum doses for the present study was based on prior preclinical, phase I, and phase II trials evaluating its effects (see Dose Selections below). Preclinical data suggest significant anti-inflammatory and antifibrotic effects in dose ranges of 0.1 mg three times per week up to 10 mg/kg twice daily, with intravenous, oral, and gavage administration tested (Tepper et al., 2014; Burstein, 2021). Single oral doses of lenabasum up to 180 mg were safe and well tolerated in a phase 1 single ascending dose study in healthy volunteers (Burstein, 2018). Additional studies evaluating lenabasum suggest that it is safe and well tolerated in individuals with neuropathic pain and systemic sclerosis (Chmiel et al., 2021; Spiera et al., 2020, 2023; Tarique et al., 2020; Werth et al., 2022) (https://clinicaltrials.gov/study/NCT03093402). Pivotal phase 3 studies for registration in scleroderma and dermatomyositis were carried out at a dose of 20 mg twice daily.

Methods

Participants.

Individuals (n = 60) aged 18 to 55 years endorsing active recreational cannabis use (smoking cannabis ≥4 days per week for ≥3 months immediately prior to screening) were recruited from advertisements in New York and New Jersey and a proprietary database of Clinilabs, Inc. After a telephone interview, eligible participants were invited to initiate in-person screening procedures, which included collection of medical and social history, concomitant medications, vital signs, physical examination, laboratory safety tests (complete blood count, metabolic panel, hepatitis panel, HCG levels in women, and urine dipstick for blood, albumin/protein, and glucose), and a 12-lead electrocardiogram (ECG). Participants were required to be positive for THC (≥50 ng/ml) on urine drug screening, be physically healthy, and demonstrate the capacity to complete self-report questionnaires akin to those administered during laboratory sessions. Participants were excluded if they were actively seeking treatment of cannabis use disorder, met DSM-5 criteria for a substance use disorder other than cannabis use disorder, or had medical or psychiatric conditions that would be exacerbated by or interfere with the study drug. Participants were also excluded if they were medically unhealthy, pregnant or breastfeeding, or not practicing an effective method of birth control. All participants were enrolled between March 2018 and August 2019 and were required to reside in a clinical research unit for the duration of the inpatient study. Informed consent was obtained from all participants. Informed consent and study procedures were approved by the Institutional Review Board of the New York State Psychiatric Institute. The study was conducted in accordance with the January 2017 US Food and Drug Administration (FDA) Center for Drug Evaluation and Research Guidance for Industry Assessment of Abuse Potential of Drugs (https://www.fda.gov/media/116739/download).

Study Design.

The study was designed as a randomized, double-blind, placebo- and active-controlled crossover study and consisted of a study qualification phase (day 1 to day 6), study treatment phase (day 7 to day 23), and safety follow-up visit (day 30 ± 2 days or 7 ± 2 days after withdrawal in the case of early discontinuation).

At prespecified time points during the study qualification and treatment phases, participants completed a visual analog scale (VAS) designed to measure subjective responses to each dose. The VAS consists of eight questions (Drug Liking, Take Drug Again, High, Good Drug Effects, Bad Drug Effects, Any Drug Effects, Desire for Marijuana, and Stoned) presented in the same order with a corresponding bipolar (Liking, Take Again, Desire) or unipolar (Good Effects, Bad Effects, Any Drug Effects, High, and Stoned) 100-mm scale per question. Participants also completed the Addiction Research Center Inventory–Marijuana scale (ARCI-M) (Bozarth, 1987), a 12-item yes/true–no/false questionnaire designed to assess the presence/absence of psychotropic effects after drug administration. The ARCI-M was administered prior to each dose administration and after each VAS administration during the study treatment phase.

During the study qualification phase, participants were randomized to receive a single oral dose of either 6 mg nabilone or placebo on day 1. On day 4, after a washout period of at least 72 hours, participants were crossed over to the alternate treatment. Participants completed a VAS questionnaire at 10 time points on dosing days during the qualification phase at 0.5, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 5.0, and 6.0 hours after dose. This qualification phase ensured that participants could distinguish the effects of active versus placebo nabilone and that they endorsed liking the effects of the 6-mg nabilone dose (E_max >60 mm and ≥15 mm closer to “Strong Liking” than the E_max to placebo on the bipolar 100-mm Drug Liking VAS). Vitals were taken prior to each dose and 1 and 2 hours after dosing, and participants were closely monitored for adverse events (AEs) or changes in concomitant medications.

During the study treatment phase, participants received a single oral dose of six different treatments (lenabasum 20 mg, 60 mg, or 120 mg; nabilone 3 mg or 6 mg; or placebo) in a randomized, six-way crossover sequence with a 6 × 6 Williams square design. Each dose included six capsules with study treatment prepared using a double dummy method and separated by at least 72 hours. The double dummy design was selected to reduce observer and participant bias and to reduce placebo effects. On dosing days (days 7, 10, 13, 16, 19, and 22), participants completed the VAS at 13 time points after dose (0.5, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 5.0, 6.0, 8.0, 12.0, and 24.0 hours). VAS question 2 (Take Drug Again) was only presented at hours 12.0 and 24.0 after dose, and questions 3 and 8 (High and Stoned, respectively) were asked before dose and after dose during both the qualification and treatment phases. Subjects also completed the ARCI-M questionnaire before dose and after each administration of the VAS questionnaire. On dosing days, blood samples were collected at 14 timepoints (before dose and 0.5, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 5.0, 6.0, 8.0, 12.0, and 24.0 hours after dose) to determine plasma lenabasum concentrations. Throughout the duration of the study treatment phase and the entire inpatient stay, vital signs were closely monitored and AEs and changes to concomitant medications assessed on an ongoing basis. Laboratory safety tests, urine drug testing, and ethanol breath testing was done on day 14. The safety follow-up visit consisted of assessments of adverse events (AEs), vital signs, concomitant medications, and a urine pregnancy test for females of child-bearing potential.

Dose Selections.

Nabilone is a synthetic CB1 receptor agonist that is FDA approved for the treatment of chemotherapy-induced nausea and vomiting. It was selected as a positive comparator based on prior work suggesting that it reliably produces dose-dependent increases in positive subjective ratings among individuals who use cannabis (Lile et al., 2011; Bedi et al., 2013). Both 3-mg and 6-mg doses of nabilone have demonstrated statistically significant increases in VAS ratings of Drug Liking and Take Again compared with placebo (Lile et al., 2011; Bedi et al., 2013). Lenabasum doses were selected based on efforts to include therapeutic (20-mg) and supratherapeutic (60-mg and 120-mg) doses and based on phase 1 and phase 2 studies of lenabasum in healthy volunteers and individuals with systemic sclerosis and cystic fibrosis that suggested pharmacological activity and an acceptable safety profile (Spiera et al., 2020, 2023; Chmiel et al., 2021). Lenabasum dose selection was also guided by input from the FDA on development of a phase 3 study of lenabasum in individuals with cystic fibrosis.

Statistical Analyses

Pharmacodynamic Endpoints.

Statistical analyses were performed using SAS software. Analyses presented here include all participants who completed the study qualification phases, received all study treatments, and completed all treatment periods. An a priori power analysis was conducted to determine required sample size to achieve 90% power to detect a statistically significant difference (15 points; a = 0.05, one-sided) between conditions. A required sample size of 54 participants was determined, with intent to randomize 60 participants to account for attrition.

The primary pharmacodynamic (PD) endpoint was peak Drug Liking (E_max) on a bipolar VAS scale, with eleven pairwise comparisons modeled. A linear mixed-effects model for a crossover study design was used to assess differences in E_max, with fixed effects for sequence, period, and treatment arm and a random effect for subject. An unstructured covariance matrix was used. Least-squares means and 90% confidence intervals were computed for each treatment arm and for all pairwise comparisons. Predose baseline was included as a covariate for tests administered before dose. Pairwise comparisons included: nabilone 3 mg versus placebo, nabilone 6 mg versus placebo, lenabasum 20 mg versus nabilone 3 mg, lenabasum 60 mg versus nabilone 3 mg, lenabasum 120 mg versus nabilone 3 mg, lenabasum 20 mg versus nabilone 6 mg, lenabasum 60 mg versus nabilone 6 mg, lenabasum 120 mg versus nabilone 6 mg, lenabasum 20 mg versus placebo, lenabasum 60 mg versus placebo, and lenabasum 120 mg versus placebo. Secondary PD analyses included time to E_max Drug Liking and seven additional VAS items (Take Again, High, Good Effect, Bad Effect, Any Effect, Desire for Cannabis, and Stoned). Secondary endpoints were analyzed using linear mixed-effects models nearly identical to the model described above but with 90% confidence intervals and two-sided tests.

Pharmacokinetic Endpoints.

The primary pharmacokinetic (PK) endpoints were maximum plasma concentration (C_max), time to maximum plasma concentration (T_max), and total plasma exposure [area under the curve (AUC) 24 hours after dose] after single oral doses of 20 mg, 60 mg, and 120 mg of lenabasum. The linear trapezoidal method, which is the standard method for calculation of area under the curve (AUC) in PK analysis, was used to compute each outcome.

Safety Endpoints.

Treatment-emergent adverse events (TEAEs) were summarized by treatment condition. Treatment emergent was defined as any event not documented or present prior to initiation of treatment or any event that worsened in either intensity or frequency after exposure to the treatment. All TEAEs were summarized by relationship to study drug and by severity by treatment. The proportion of subjects experiencing a TEAE was analyzed for each dose condition.

Results

Participant Demographics.

One hundred eighty-seven participants signed informed consent, 85 failed screening, and 102 entered the qualification phase (QP) and received nabilone 6 mg and/or placebo. Two of these participants discontinued due to adverse events, one had mild tonsillitis unrelated to study medication or procedures, and another had severe dizziness, moderate somnolence, nausea, dry eyes, and vomiting related to nabilone 6 mg. Sixty participants from the qualification phase continued into the treatment phase (TP) and received study medication; four of these participants discontinued early due to withdrawal of consent. No participant discontinued study participation in the TP due to an adverse event, and 56 participants completed all study procedures. Participants who completed the trial had a mean age of 29.5 years (S.D. 7.24), and a mean BMI of 26.4 kg/m2 (S.D. 3.52). The sample was majority male (78.6%, n = 44) and predominantly African American (80.4%, n = 45) and non-Hispanic (82.1%; n = 46; Table 1).

TABLE 1.

Participant characteristics

	Completer Population (N = 56)
	N (%)
Sex
Male	44 (78.6%)
Female	12 (21.4%)
Race
African American or Black	45 (80.4%)
White	10 (17.9%)
Native Hawaiian or Pacific Islander	1 (1.8%)
Other: Black – Biracial	1 (1.8%)
Ethnicity
Hispanic or Latino	10 (17.9%)
Non–Hispanic or Latino	46 (82.1%)
	Mean (S.D.)
Age	29.5 (7.24)
BMI	26.4 (3.52)
Baseline Cannabis Use (grams per day)	3.28 (3.22)

Primary Pharmacodynamic Endpoint.

Least-squares (LS) mean E_max values of Drug Liking were significantly greater for both active doses of nabilone compared with placebo (P < 0.003), confirming the study’s validity. LS mean area under the effect (AUE) values for both doses of nabilone were significantly greater than placebo (P < 0.00001).

LS mean E_max values for Drug Liking at 20 mg and 60 mg lenabasum were numerically different from E_max values for placebo, but the difference did not reach the prespecified threshold of a ≥11-point difference and were therefore not statistically significant. LS mean E_max values for Drug Liking for 120 mg lenabasum were statistically significantly greater than placebo. In comparing lenabasum to nabilone, all three doses of lenabasum had significantly lower E_max Drug Liking values than 3 mg and 6 mg nabilone (P < 0.001; Table 2). AUE_{0–2 hours} Drug Liking values for all lenabasum doses were indistinguishable from placebo. AUE_{0–2 hours} for Drug Liking values for lenabasum 20, 60, and 120 mg were lower than AUE_{0–2 hours} Drug Liking values for nabilone 3 and 6 mg.

TABLE 2.

E_max subjective VAS measures across study doses

“Drug Liking,” “Take Again,” and “Desire for Cannabis” were presented on a 100-mm bipolar scale (anchors for each item include: Liking: “Strong Disliking,” “Neither Like nor Dislike,” “Strong Liking”; Take Again: “Definitely Would Not,” “Do Not Care,” “Definitely Would”; Desire for Cannabis: “Definitely Would,” “Definitely Would Not”). All other items were presented on a 100-mm unipolar scale ranging from 0 (“not at all”) to 100 (“extremely”). Values in bold type are significantly different from placebo (P < 0.01). Values highlighted in gray are significantly different from 3 mg nabilone (P < 0.01). Underlined values are significantly different from 6 mg nabilone (P < 0.01).

Peak Effect	Placebo LS Mean (S.E.)	20 mg Lenabasum LS Mean (S.E.)	60 mg Lenabasum LS Mean (S.E.)	120 mg Lenabasum LS Mean (S.E.)	3 mg Nabilone LS Mean (S.E.)	6 mg Nabilone LS Mean (S.E.)
Drug Liking	56.1 (2.08)	53.7 (2.08)	62.6 (2.08)	70.8 (2.08)	79.8 (2.08)	86.5 (2.08)
High	14.5 (4.00)	11.6 (4.00)	28.3 (4.00)	45.7 (4.00)	63.5 (4.00)	79.2 (4.00)
Good Drug Effect	17.1 (4.08)	13.9 (4.08)	28.1 (4.08)	45.8 (4.08)	65.3 (4.08)	79.4 (4.08)
Bad Drug Effect	6.6 (2.88)	6.5 (2.88)	8.1 (2.88)	14.3 (2.88)	19.2 (2.88)	26.4 (2.88)
Any Drug Effect	17.0 (4.24)	16.5 (4.24)	32.2 (4.24)	51.9 (4.24)	73.0 (4.24)	82.9 (4.24)
Take Drug Again	49.6 (3.05)	48.6 (3.05)	58.8 (3.05)	65.5 (3.05)	69.7 (3.05)	76.2 (3.05)
Desire for Cannabis	66.3 (4.70)	72.6 (4.70)	62.7 (4.70)	58.1 (4.70)	51.1 (4.70)	42.2 (4.70)
Stoned	13.4 (4.29)	8.8 (4.29)	23.2 (4.28)	34.7 (4.29)	43.4 (4.28)	57.8 (4.28)

The median time to peak effect on Drug Liking (tE_max) for 20, 60, and 120 mg lenabasum was 2.0, 3.0, and 3.0 hours, respectively, compared with 2.5 hours for both nabilone doses and 2.0 hours for placebo (Fig. 1).

Fig. 1.

Mean drug effects by treatment and time. Note: “Drug Liking” was presented on a 100-mm bipolar scale (anchors for Liking included: “Strong Disliking” at 0 mm, “Neither Like nor Dislike” at 50 mm, and “Strong Liking” at 100 mm). “Good Drug Effects” was presented on a 100-mm unipolar scale ranging from 0 (“not at all”) to 100 (“extremely”).

Secondary Pharmacodynamic Endpoints.

With regard to secondary PD endpoints, 20 mg lenabasum did not significantly differ from placebo on any VAS items (High, Good Effect, Bad Effect, Any Effect, Take Again, Desire for Cannabis, and Stoned). Similarly, both 3-mg and 6-mg doses of nabilone were associated with significantly higher VAS ratings on all items compared with 20 mg lenabasum. On the other hand, supratherapeutic (60 mg and 120 mg) doses of lenabasum were associated with significantly different VAS ratings when compared with placebo. Specifically, compared with placebo, a 60-mg dose of lenabasum was associated with significantly higher ratings of High, Good Drug Effect, Any Drug Effect, Take Again, and Stoned. Compared with placebo, a 120-mg dose of lenabasum was associated with higher ratings of High, Good Drug Effect, Bad Drug Effect, Any Drug Effect, Take Again, Desire for Cannabis, and Stoned. Both nabilone doses were associated with the highest E_max scores on all secondary VAS endpoints. Specifically, when comparing 60 mg lenabasum to 3 mg nabilone, nabilone was associated with significantly higher ratings on all VAS items. Compared with 120 mg lenabasum, 3 mg nabilone was associated with significantly higher ratings of High, Good Drug Effect, and Any Drug Effect. Finally, compared with 60 mg and 120 mg lenabasum, 6 mg nabilone was associated with significantly higher ratings on all secondary VAS items (Fig. 1; Table 2).

The 20-mg lenabasum dose had no effect on ARCI-M scores compared with placebo, and results were generally consistent with those observed for the Drug Liking parameter. At 2.5 hours after dose, which was the approximate median time to E_max on Drug Liking VAS, the proportion of subjects with an increase in ARCI-M score ≥5 was lowest for 20 mg lenabasum (1.8%) and placebo (3.6%), followed by 60 mg lenabasum (10.7%), then 120 mg lenabasum (21.4%), and lastly 3 mg nabilone (35.7%) and 6 mg nabilone (44.6%). The highest ARCI-M scores were observed in the nabilone 6-mg treatment group.

Pharmacokinetic Effects.

Plasma concentrations rose rapidly with a peak of 2.5 hours for 20 mg lenabasum and 2.9 hours for 60 and 120 mg lenabasum (Table 3). Plasma concentration time curves after the absorption phase declined in a manner consistent with first-order linear kinetics. Mean AUC_0–24 increased approximately 2.7-fold from 20 mg to 60 mg lenabasum and approximately 1.7-fold from 60 mg to 120 mg lenabasum. Mean C_max increased approximately 2.3-fold from 20 mg to 60 mg lenabasum and approximately 1.5-fold from 60 mg to 120 mg lenabasum. Median T_max was 2.5 hours for all three lenabasum doses, which was within ±30 minutes of median tE_max (Fig. 2; Tables 3 and 4).

TABLE 3.

Summary of lenabasum plasma pharmacokinetic parameters

Nabilone plasma levels were not measured in this study. Nabilone manufacturer package insert states that peak plasma concentrations are achieved within 2 hours.

Parameter	Statistic	20 mg Lenabasum (N = 56)	60 mg Lenabasum (N = 56)	120 mg Lenabasum (N = 56)
AUC0–24 (hours*ng/ml) Cmax (hours*ng/ml) Tmax (hours)	Mean (CV) Mean (CV) Median Q1, Q3 Min, Max	3710 (36.9) 748 (33.2) 2.5 2.0, 2.5 1.5, 8.0	9980 (35.3) 1750 (32.0) 2.5 2.1, 2.8 1.5, 12.0	16,700 (30.8) 2580 (28.3) 2.5 2.5, 3.0 1.0, 12.0

Max, maximum; Min, minimum; Q1, first quartile (25th percentile); Q3, third quartile (75th percentile).

TABLE 4.

Summary of time to peak effect (tE_max) for Drug Liking VAS

Parameter	Statistic	Placebo	20 mg Lenabasum	60 mg Lenabasum	120 mg Lenabasum	3 mg Nabilone	6 mg Nabilone
tEmax (h)	Median Q1, Q3 Min, Max	2.0 1.0, 3.8 0.5, 24.0	2.0 0.8, 3.5 0.5, 24.0	3.0 1.5, 5.0 0.5, 12.0	3.0 1.8, 5.0 0.5, 12.0	2.5 1.5, 4.0 0.5, 24.0	2.5 1.8, 3.8 0.5, 12.0

h, hours; Max, maximum; Min, minimum; Q1, first quartile (25th percentile); Q3, third quartile (75th percentile).

Fig. 2.

Mean (+S.D.) plasma lenabasum concentration time profiles by dose.

Safety Endpoints.

Single oral doses of lenabasum 20, 60, and 120 mg were safely administered and well tolerated in the current sample. There were no TEAEs resulting in discontinuation for any of the active lenabasum doses.

The proportion of participants reporting at least one TEAE was similar among treatment groups (Table 5). The most common TEAEs reported in at least one subject in any treatment group or more than one subject across treatment groups are presented in Table 5. Mild- or moderate-intensity headache was the most common TEAE reported in all treatment groups, including placebo. Mild to moderate dizziness was reported in all treatment groups except 20 mg lenabasum. Gastrointestinal AEs, including nausea, upper abdominal pain, and constipation, were reported in placebo, lenabasum, and nabilone groups. Dry mouth and vomiting were reported in lenabasum and nabilone groups. Palpitations were reported in placebo, 120 mg lenabasum, and 6 mg nabilone groups. There were no clinically significant adverse events, changes in safety laboratories, ECGs, physician examination findings, or vital signs during the study.

TABLE 5.

Summary of most common treatment-emergent adverse events (TEAEs)

	Placebo (N = 60) n (%)	20 mg Lenabasum (N = 58) n (%)	60 mg Lenabasum (N = 57) n (%)	120 mg Lenabasum (N = 58) n (%)	3 mg Nabilone (N = 58) n (%)	6 mg Nabilone (N = 58) n (%)
Participants with at least one TEAE	14 (23.3)	9 (15.5)	7 (12.3)	15 (25.9)	11 (19.0)	12 (20.7)
Most common TEAEs a
Headache	4 (6.7)	5 (8.6)	2 (3.5)	2 (3.4)	1 (1.7)	4 (6.9)
Dizziness	1 (1.7)	0	1 (1.8)	2 (3.4)	3 (5.2)	2 (3.4)
Dry mouth	0	0	1 (1.8)	2 (3.4)	0	4 (6.9)
Nausea	2 (3.3)	0	1 (1.8)	2 (3.4)	1 (1.7)	1 (1.7)
Vomiting	0	1 (1.7)	0	1 (1.7)	1 (1.7)	3 (5.2)
Upper abdominal pain	1 (1.7)	2 (3.4)	0	0	1 (1.7)	0
Constipation	1 (1.7)	1 (1.7)	1 (1.8)	1 (1.7)	0	0
Palpitations	3 (5.0)	0	0	3 (5.2)	0	2 (3.4)
Fatigue	1 (1.7)	1 (1.7)	0	2 (3.4)	0	0
Contact dermatitis	1 (1.7)	0	0	2 (3.4)	0	0
Hypoesthesia	2 (3.3)	0	0	0	0	0

^aMost common TEAE defined as occurring in more than one subject in any treatment group or in at least one subject in multiple treatment groups (e.g., constipation occurred in one subject in four treatment groups).

Discussion

The present study evaluated the abuse potential of lenabasum, a selective CB2 receptor agonist compared with a positive control (3 mg and 6 mg nabilone) and placebo in a double-blind, placebo-controlled, randomized crossover design. The present study recruited individuals endorsing active recreational cannabis use, and eligible participants were enrolled into a two-phase study. Findings suggest that a target therapeutic dose (20 mg) of lenabasum had no effect on Drug Liking compared with placebo and had no distinguishable effect on other PD endpoints (Take Drug Again, High, Good Drug Effect, Bad Drug Effect, Desire for Cannabis, and Stoned) compared with placebo. A dose-dependent relationship was observed with regard to Drug Liking such that 60- and 120-mg lenabasum doses produced significantly higher ratings of Drug Liking and other VAS items compared with placebo. Although lenabasum 60 and 120 mg produced greater ratings of Drug Liking and other VAS ratings compared with placebo, Drug Liking and all other VAS outcomes were highest for nabilone 3 mg and 6 mg, suggesting that at the doses that were tested, lenabasum has less abuse potential than 3 mg and 6 mg of nabilone. Although PK curves for nabilone doses were not computed in the current sample, published data suggests that maximum plasma concentration occurs within 2 hours after nabilone dosing (Lemberger et al., 1982; McGilveray, 2005; Tsang and Giudice, 2016). This suggests a shorter average time to peak concentration for nabilone compared with doses of lenabasum studied in the current trial.

Although this study offers novel findings in a well powered, highly controlled inpatient design, it is not without limitations. First, although consistent with 2017 FDA guidance that recommends testing doses of a proposed medication that are two to three times the therapeutic dose, the present study examined a lenabasum dose (120 mg) that was six times the expected therapeutic dose (20 mg). Therefore, we cannot determine if higher doses of lenabasum (e.g., 200 mg) elicit greater effect or have abuse potential similar to or greater than nabilone (3 mg and 6 mg). However, it is worth noting that testing a dose six times the expected therapeutic dose conforms with FDA guidelines on evaluating abuse potential, and to our knowledge reports of nabilone misuse (since its approval in 2010) are quite rare. To date, there is no widespread evidence that nabilone is commonly misused. Further, most human abuse potential studies only use one active comparator dose, so use of both 3 mg and 6 mg nabilone is notable. Still, there remains a possibility that individuals seeking reinforcing effects may use lenabasum in significantly higher doses than those tested in the current trial, which is a limitation of the current design. Second, although efforts were made to recruit a generalizable sample of participants, the complete population of the current sample is majority male (78.6%), and majority African American (80.4%), limiting our ability to analyze subgroup differences and potentially limiting generalizability to other groups. Finally, although a strength of the present study was the inpatient format, which increases experimental control, it is possible that this inpatient design failed to capture more naturalistic drug use patterns (especially in the current context in the United States of state-regulated recreational cannabis use), which may impact abuse potential.

Overall, the current trial suggests that lenabasum is safe and well tolerated in a sample of recreational cannabis users. A 20-mg dose of lenabasum was not associated with an increase in Drug Liking compared with placebo and therefore appears to have a low risk for abuse. The 120-mg dose of lenabasum was associated with increased Drug Liking compared with placebo and both 60-mg and 120-mg doses of lenabasum were associated with increased secondary VAS endpoints compared with placebo. However, compared with nabilone, these doses of lenabasum produced significantly lower ratings of Drug Liking, which may be in part due to lower brain penetration of lenabasum as well as subtle mechanistic differences related to its interaction with CB1 as well as a longer absorption time with a larger dose and more capsules administered. Although drugs with faster onsets of action are generally better liked than drugs with slower onsets of action, the observed differences here are minor and may be more likely related to differences in degree of receptor engagement. Therefore, the current trial supports the therapeutic use of 20 mg lenabasum as a safe and well tolerated compound with low potential for abuse. Although there is some concern for abuse potential at supratherapeutic doses, it is important to note that even in this sample of recreational cannabis users, supratherapeutic doses of lenabasum were rated as less well liked than 3-mg and 6-mg doses of nabilone, which is currently approved by the FDA. Further work examining the therapeutic efficacy of lenabasum for inflammatory conditions may provide additional insight into the risk/benefit ratio of this compound.

Abbreviations

AE

adverse event

ARCI-M

Addiction Research Center Inventory–Marijuana scale

AUC

area under the curve

AUE

area under the effect

CB1

cannabinoid receptor 1

CB2

cannabinoid receptor 2

CBD

cannabidiol

CNS

central nervous system

E_max

peak effect

FDA

US Food and Drug Administration

LS

least-squares

PD

pharmacodynamic

PK

pharmacokinetic

TEAE

treatment-emergent adverse event

tE_max

time to peak effect

THC

delta-9-tetrahydrocannabinol

T_max

time to maximum plasma concentration

VAS

visual analog scale

Authorship Contributions

Participated in research design: Comer.

Conducted experiments: Comer.

Wrote or contributed to the writing of the manuscript: Luba, Madera, Schusterman, Kolodziej, Hodgson, Comer.