The Individual and Interactive Effects of Alpha-Pinene and Delta-9-Tetrahydrocannabinol in Healthy Adults

Abstract

Introduction

Cannabis contains hundreds of chemical constituents beyond delta-9-tetrahydrocannabinol (Δ9-THC), which is thought to be the primary driver of most of its acute pharmacodynamic effects. The entourage effect theory asserts that the pharmacological and therapeutic effects of cannabis are not solely attributable to Δ9-THC but are influenced by other constituents, such as minor cannabinoids and terpenes, through distinct pharmacological action. However, empirical studies that have systematically evaluated this theory in humans remain limited. This study tested the hypothesis that the terpene α-pinene can attenuate the acute memory-impairing effects of inhaled Δ9-THC in humans.

Methods

Participants (N = 19; last cannabis use 39 days, on average, prior to first test session [SD = 84; range = 2–365]) completed six double-blind outpatient drug administration sessions during which they inhaled, using a Mighty Medic hand-held vaporizer, α-pinene alone (15 mg), Δ9-THC alone (30 mg), Δ9-THC and α-pinene together (30 mg Δ9-THC + 0.5 mg α-pinene; 30 mg Δ9-THC + 5 mg α-pinene; 30 mg Δ9-THC + 15 mg α-pinene), or placebo (ambient air) in a randomized order. Outcomes, which were collected up to 6 h post-drug exposure, included subjective drug effects, cognitive/psychomotor performance, and vital signs.

Results

Administration of 15 mg α-pinene alone produced no significant pharmacodynamic effects compared to placebo. Administration of 30 mg Δ9-THC alone elicited subjective, cognitive, and physiological effects consistent with acute Δ9-THC-dominant cannabis exposure, including impairment of cognitive performance and working memory ability compared to placebo. The co-administration of α-pinene with Δ9-THC did not mitigate Δ9-THC-induced memory impairment or significantly alter other acute subjective, cognitive, or physiological effects.

Conclusions

Inhaled α-pinene, at doses at and above those naturally found in cannabis flowers, did not mitigate Δ9-THC-induced cognitive impairments as hypothesized or influence other common acute effects of Δ9-THC in this sample of healthy adults. This result is inconsistent with some cannabis industry claims and speculation by some cannabis researchers. By systematically varying both Δ9-THC and terpene exposure and assessing their interaction across multiple pharmacodynamic domains, this work provides a model for future investigations into Δ9-THC-terpene interactions. As cannabis use continues to expand for both medicinal and non-medicinal purposes, more research is needed to better understand the acute effects of lesser studied chemical constituents of the plant and how they interact with predominant phytocannabinoids like Δ9-THC. This can inform cannabinoid drug development and product regulations.

Introduction

Cannabis is one of the most widely used psychoactive substances globally, with its consumption continuing to increase due to expanding legalization for both medicinal and non-medicinal (“recreational”) purposes [1–3]. In addition, the advent of legal retail sales has resulted in an increased diversity of product types and marketing about how characteristics of these products lead to a different user experience. The rising consumption and increased diversity in product composition and associated messaging highlights the importance of understanding the pharmacological and pharmacodynamic effects of individual chemical constituents of cannabis.

Delta-9-tetrahydrocannabinol (Δ9-THC), the primary psychoactive constituent of cannabis, has been extensively studied. Acting as a partial agonist at the cannabinoid type 1 (CB₁) and type 2 (CB₂) receptors, inhaled or orally administered Δ9-THC produces a range of behavioral effects [4, 5]. Its positive and/or purported therapeutic effects include feelings of euphoria, relaxed mood, as well as analgesic, antiemetic, and muscle relaxant effects. However, Δ9-THC can also induce negative and unwanted side effects, including dysphoria (panic, paranoia, acute psychosis), nausea/emesis, and cognitive impairment (working memory, divided attention, complex cognition), especially at acute doses above 10 mg (or 2 standard Δ9-THC units [STUs]) [6, 7].

Historically, Δ9-THC was regarded as the primary driver of the behavioral and psychoactive effects of cannabis, with other chemical constituents considered to be largely inconsequential. As a result, early research predominantly focused on Δ9-THC due to its psychoactive properties, while research on other phytocannabinoids (e.g., cannabinol [CBN], cannabigerol [CBG]) and terpenes (e.g., d-limonene, β-myrcene, α-pinene) commonly found in the cannabis plant has been far more limited [8]. In recent years, interest in “minor” cannabinoids and terpenes has grown, in part, due to an increased propagation of the so-called entourage effect theory [9]. This theory posits that the pharmacological and therapeutic effects of cannabis are not driven by Δ9-THC alone, but rather that many other cannabis constituents, including minor cannabinoids and terpenes, contribute meaningfully to these effects via direct and distinct pharmacological action.

The entourage effect theory has influenced how cannabis products are cultivated, marketed, and consumed. For example, cannabis plants are being selectively bred to have specific cannabinoid and terpene profiles, driven by the belief that certain ratios of these compounds produce specific desired effects for the user [10, 11]. It is also common practice for retail employees to offer specific recommendations of cannabis products to individuals based on health conditions they have, or desired effects from the drug [12]. Despite these emergent industry practices and anecdotal reports from cannabis consumers [13], rigorous scientific evidence to support or refute the validity of the entourage effect is largely lacking with respect to the specific effects of most minor cannabinoids/terpenes. To date, clinical studies on the interactive effects of Δ9-THC and other cannabinoids have primarily focused on cannabidiol (CBD), a compound believed by many to mitigate some of Δ9-THC’s psychoactive effects and reduce its adverse effect profile. While Δ9-THC-CBD interactions have been studied more extensively, findings are mixed [14] and recent work highlights that the route of administration significantly impacts the interaction between Δ9-THC and CBD [15]. Importantly, controlled research on Δ9-THC-terpene interactions remains understudied, representing a critical gap in our understanding of the behavioral pharmacology of cannabis.

In vitro receptor binding studies have shown mixed results, perhaps due to different methodological approaches, but there is evidence that some terpenes commonly found in cannabis have a CB₁-mediated mechanism of action and/or have additive or even synergistic effects on CB₁ activity when combined with Δ9-THC [16–19]. Additional evidence of meaningful Δ9-THC-terpene interactions comes from a recent human laboratory study that showed a dose-dependent attenuation of Δ9-THC-induced anxiety by d-limonene [20]. In that study, inhalation of 30 mg Δ9-THC and 15 mg d-limonene via vaporization was associated with significantly lower ratings of “anxious/nervous” and “paranoid” compared with 30 mg Δ9-THC alone, without altering the magnitude or time course of any other subjective, cognitive, or cardiovascular effects. These findings suggest that terpenes may selectively modulate the acute effects of Δ9-THC under some conditions, which has medical implications. Because adverse effects of Δ9-THC (e.g., anxiety, cognitive impairment) are often dose-dependent [21], co-administration of terpenes that can mitigate these effects, such as limonene’s attenuation of Δ9-THC-induced anxiety [20], may help widen the therapeutic window of Δ9-THC, which could improve its clinical utility. However, further studies are needed to explore how other terpenes may influence Δ9-THC’s pharmacological and psychoactive properties.

α-Pinene, one of the most abundant terpenes found in cannabis and various other plant species, has gained interest due to its purported neuroprotective and cognitive-enhancing properties [22]. α-Pinene acts as a competitive acetylcholinesterase inhibitor, potentially enhancing cognitive function by preserving acetylcholine, a neurotransmitter crucial for learning and memory [9, 23]. Preclinical studies provide some evidence for α-pinene having cognitive benefits. For instance, administration of an essential oil containing α-pinene (6.07%) was associated with a 72% increase in memory performance on a passive avoidance test compared with vehicle in mice [24]. Similarly, administration of a terpene-rich (e.g., α-pinene, β-pinene, d-limonene, linalool) lemon essential oil improved performance on novel object recognition and Morris water maze tests (suggestive of improved recognition and working memory) in amyloid precursor protein/presenilin 1 (APP/PS1) transgenic mice (a model of Alzheimer’s disease) compared to untreated mice [25]. In human studies, oral consumption of a sage essential oil, which included terpene constituents such as α-pinene (5.6%), improved performance on secondary memory tasks, reduced mental fatigue, and increased alertness in healthy participants [26].

In contrast to the cognitive-enhancing properties of α-pinene, Δ9-THC is well-documented for its acute deleterious effects on memory, particularly short-term and working memory impairments [27], with numerous studies indicating that acute use of Δ9-THC or cannabis can impair memory functioning [28, 29]. This raises the important question of whether α-pinene, a terpene with some demonstration of neuroprotective and cognitive-enhancing properties, could counteract short-term memory deficits induced by Δ9-THC. However, as described above, most prior research evaluated products that contained α-pinene in combination with other substances and no human studies have investigated directly whether α-pinene can modulate the effect of Δ9-THC on memory performance, representing a critical gap in understanding this concept. The present controlled laboratory study evaluated whether, and to what extent, α-pinene modulates the acute memory-impairing effects of inhaled Δ9-THC. It was hypothesized that inhalation of α-pinene would attenuate Δ9-THC-induced impairment of working memory and verbal recall. Secondary endpoints included evaluation of other common acute subjective, cognitive performance, and physiological effects of Δ9-THC as well as assessing whether α-pinene alone differed from placebo on any pharmacodynamic assessments.

Methods

Study Design

The present study utilized a double-blind, within-subjects crossover design. All participants completed a total of six outpatient drug administration sessions during which they inhaled α-pinene alone (15 mg), Δ9-THC alone (30 mg), Δ9-THC and α-pinene together (30 mg Δ9-THC + 0.5 mg α-pinene; 30 mg Δ9-THC + 5 mg α-pinene; 30 mg Δ9-THC + 15 mg α-pinene), or placebo (ambient air). The six experimental sessions were completed in a randomized order to minimize possible order effects and reduce bias [30]. All sessions were separated by at least 48 h. This study was conducted at the Johns Hopkins University (JHU) Cannabis Science Laboratory, which is part of the Behavioral Pharmacology Research Unit (BPRU). The protocol was approved by the JHU School of Medicine Institutional Review Board (IRB00085652), approved by the US Food and Drug Administration (IND140339), and was registered on ClinicalTrials.gov (NCT03609853).

Screening and Experimental Procedures

Participants were recruited via internet advertisements and word-of-mouth communication. Interested individuals first completed a telephone or web-based screening questionnaire. Those who were eligible based on the initial screening were invited for an in-person evaluation consisting of written informed consent followed by a detailed medical history review and physical examination. A blood sample was taken for clinical chemistry testing and pregnancy testing for female volunteers, a urine sample was taken to test for recent illicit drug use (i.e., methamphetamine, PCP, cocaine, MDMA, oxycodone, morphine, amphetamine, methadone, benzodiazepines, barbiturates, buprenorphine, and Δ9-THC), a breath sample was taken to test for recent alcohol use, and the Timeline Follow-Back was used to assess non-medicinal drug and alcohol use over the past 90 days [31].

Study inclusion criteria were: (1) aged 18–55; (2) good health status as determined by in-person screening (e.g., medical exam, vital signs, blood clinical chemistry); (3) tested negative for drugs of misuse other than cannabis at screening and prior to each study visit; (4) not pregnant or breastfeeding; (5) body mass index (BMI) between 18 and 36 kg/m²; (6) screening and baseline blood pressure not >150 mm Hg systolic blood pressure or >90 mm Hg diastolic blood pressure; (7) no allergies to study drugs; (8) demonstrated competency on cognitive performance measures at screening visit. Study exclusion criteria included: (1) self-reported non-medicinal use of psychoactive drugs other than cannabis, nicotine, alcohol, or caffeine in the 30 days prior to randomization; (2) history or current evidence of a significant medical condition that could put the participant at risk (e.g., seizure or cardiac disorder); (3) use of prescription or over-the counter medications (including supplements or vitamins) that could interfere with study results or participant safety; (4) use of dronabinol in the past 30 days; (5) enrollment in another clinical trial in the past 30 days; (6) having sought medical attention (e.g., ER visit) to manage adverse events (AEs) from cannabis in the past; (7) having anemia or having donated blood in the past 30 days; and (8) use of cannabis, on average, more than twice per week in the past 3 months.

Participants that met the above criteria were occasional cannabis smokers and were scheduled to complete the 6 drug exposure sessions. Upon arrival for each session, participants provided a urine sample to test for pregnancy/recent drug use, completed an alcohol breathalyzer, and self-reported their use of drugs/alcohol since their last visit via the Timeline Follow-Back (sessions were not conducted if the participant tested positive for pregnancy, alcohol use, or use of drugs aside from cannabis). Next, participants received a standard low-fat breakfast of toast and jam followed by baseline assessments of subjective effects, cognitive performance, and vital signs.

After baseline assessments, participants were taken to a ventilated negative pressure chamber for drug self-administration under staff supervision. Participants inhaled either placebo (ambient air), α-pinene, Δ9-THC, or a combination of Δ9-THC and α-pinene using the Mighty Medic hand-held vaporizer (Storz and Bickel^®; see below for drug preparation details). The Mighty Medic was set to a temperature of 210°C in each study condition. Participants were given 15 min to inhale the assigned study drug(s) ad libitum (i.e., at their own pace) but were required to take a minimum of 15 puffs (as in a prior experiment [20], 15 puffs was sufficient to exhaust a 30 mg Δ9-THC dose for most participants). An ad libitum puff procedure was used to allow participants to dose comfortably and to allow for dose titration if adverse drug effects began to emerge during drug administration. After each puff, participants exhaled into a handheld smoke filter (Sploofy; City of Industry, CA, USA) to preserve the study blind, as pre-study testing showed that vapor visibility differed across drug conditions. After the 15th puff, participants exhaled into the open air to determine if the dose had been depleted. If a visible vapor was still observed after the 15th puff, participants were instructed to continue taking puffs until they no longer exhaled a visible vapor. Additionally, a specialized adapter was used to capture puff topography (e.g., volume and intensity of each puff) during drug administration. Following drug administration, pharmacodynamic outcome measures were collected at 15–60 min intervals for 6 h (see Outcome Measures below). Participants were permitted to eat lunch and snacks as needed in between assessments during the experimental sessions.

Study Drug and Materials

All study drugs were prepared and dispensed by the Johns Hopkins University (JHU) Behavioral Pharmacology Research Unit (BPRU) Pharmacy. Synthetic Δ9-THC (>99% purity) was obtained from Δ9-THC Pharm GmbH (Frankfurt Am Main, Germany) and was dissolved in pharmacy-grade ethanol (190 proof; Spectrum Chemical, Gardena, CA, USA) to create a solution that was approximately 10% Δ9-THC/90% ethanol. Botanically derived α-pinene (>99% purity) was obtained from True Terpenes (Hillsboro, OR, USA). Prior to each experimental session, a pharmacist applied a precise amount of α-pinene and/or Δ9-THC using a micropipette to a steel wool dosing pad, which fit into a small dosing capsule (or “pod”) that was placed inside a hand-held vaporizer (Mighty Medic, Storz & Bickel, GmbH). Dosing pods containing Δ9-THC were kept out of the dosing chamber for approximately 30 min to allow the ethanol to dissipate prior to loading into the vaporizer.

Because there was no published safety or efficacy data related to the direct inhalation of α-pinene at the outset of the experiment, we conducted testing to ensure that α-pinene was effectively vaporized and not chemically altered by the vaporization process (GC/MS analysis of multiple test doses captured by a carbon filter). Doses of α-pinene were selected based on chemical analysis of 107 samples of cannabis, representing 29 unique cannabis cultivars, being sold by a licensed medical cannabis producer in Canada. Specifically, the two lower doses of α-pinene used in this study (0.5 mg and 5 mg) reflect the mean and max α-pinene doses contained in 1 g of cannabis flower based on the analyses of those cannabis samples. One gram is the amount of plant material commonly used to make a single cannabis cigarette (“pre-roll,” “joint,” or “blunt”). In addition to the 0.5 mg and 5 mg α-pinene doses, a 15 mg α-pinene dose was added into the study after pilot testing demonstrated safety at the two lower doses and a lack of overt impact of those doses on acute Δ9-THC exposure. Thus, doses used in this study were ecologically relevant and pilot testing demonstrated safety with respect to acute dosing via direct inhalation.

Outcome Measures

Subjective drug effects, subjective ratings of mood, vital signs, and cognitive performance were assessed at baseline, immediately following drug exposure (i.e., time “0”), and 0.25, 0.5. 0.75, 1, 2, 3, 4, 5, and 6 h after drug exposure. Study measures were completed in the same order each time.

Subjective Drug Effects

A Drug Effect Questionnaire (DEQ [32]) was administered that included 21 items presented individually on a 100-mm visual analog scale ranging from 0 (“not at all”) to 100 (“extremely”) [21]. This questionnaire assessed the overall magnitude of drug effect (e.g., “feel drug effect”), as well as positive (e.g., drug “liking”) and negative/adverse effects (e.g., “unpleasant,” “sick”). One item on the DEQ assessed acute self-reported memory impairment (i.e., “trouble with memory”).

The 35-item Profile of Mood States (POMS [33]) questionnaire was used to assess state-dependent changes in mood before and after drug administration on 7 different domains: anger-hostility, confusion-bewilderment, depression-dejection, fatigue-inertia, tension-anxiety, vigor-activity, and friendliness. The confusion-bewilderment sub-scale score was an outcome of interest for this study.

Cognitive Performance Tasks

Cognitive performance was assessed using three tasks, the International Shopping List Test (ISLT [34], the Paced Serial Addition Task (PASAT [35]), and the Digit Symbol Substitution Task (DSST [36]). Participants were trained on these tasks during the screening visit and performance on both the PASAT and DSST was repeated until a stable baseline was established to minimize practice effects during experimental sessions.

The ISLT is a simple delayed verbal recall task that has been shown to be sensitive to acute cannabis exposure in prior controlled studies [37]. In this task, participants were shown a 12-item grocery list prior to cannabis exposure and then asked to freely recall as many items from that list as possible at the end of the 0-h time point (approximately 15 min postdrug administration; near peak drug effects). This test assessed the impact of drug exposure on memory retrieval. Following the free recall of the initial pre-dose shopping list, participants were presented with a new list of 12 non-grocery items and were asked to memorize this new list. Recall of the second list was then assessed at the end of the experimental session (6 h-post dosing). This second list tested the impact of drug exposure on memory encoding/consolidation as well as memory retrieval. Participants were presented with new lists at each experimental session.

The PASAT is a task that predominantly assesses working memory, but also includes aspects of psychomotor ability and higher order cognition. In the task, participants viewed a string of single-digit numbers on a computer screen and were instructed to select the sum of the two numbers most recently presented on the computer screen by clicking on squares containing potential sum integers using the computer mouse. Numbers were presented at intervals ranging from 2.4 to 2.8 s for a total of 90 trials. The primary outcome for the PASAT was the total number of correct responses and secondary outcomes included percent correct responses and reaction time to both correct and incorrect responses.

The DSST is a task that predominantly assesses psychomotor ability, but also includes a working memory component of performance. In this task, participants were presented a series of patterns on the computer screen that needed to be replicated using the numerical keys on the computer keyboard as quickly and accurately as possible over a 90 s test period. The primary outcome for the DSST was the total number of correct responses and secondary outcome was the percent of correct responses.

Vital Signs and Pharmacokinetics

Heart rate (HR), systolic blood pressure, and diastolic blood pressure were measured in the seated position using an automated monitor. Participants were required to have been seated for 5 min prior to vitals measurement.

Data Presentation and Analysis

Demographic characteristics are presented using descriptive statistics, including means and standard deviations. Change from baseline and peak change from baseline scores were calculated for subjective effects, cognitive performance, and vital signs. Change from baseline scores for each outcome were analyzed using repeated-measures analysis of variance (ANOVA), with the within subject factors of dose (placebo, 15 mg α-pinene, 30 mg Δ9-THC, 30 mg Δ9-THC + 0.5 mg α-pinene, 30 mg Δ9-THC + 5 mg α-pinene, 30 mg Δ9-THC + 15 mg α-pinene) and time (each post-dose assessment) (see online suppl. Table 1; for all online suppl. material, see https://doi.org/10.1159/000547014 for ANOVA results). Planned comparisons (Fisher’s LSD tests) were used to compare peak change from baseline data for: (1) α-pinene alone (15 mg) to placebo; (2) Δ9-THC alone (30 mg) to placebo; and (3) Δ9-THC alone (30 mg) to the corresponding Δ9-THC/α-pinene combination conditions (i.e., 0.5, 5, and 15 mg α-pinene + 30 mg Δ9-THC). Peak change from baseline scores were constrained to assessments conducted within the first 3-h post-drug exposure, which captures the time window when peak pharmacodynamic effects typically occur after inhaled Δ9-THC exposure. Larger changes outside of 3 h likely would have been impacted by nondrug-related factors (e.g., boredom and/or fatigue). Moreover, because 4 of 6 total dose conditions included an identical dose of Δ9-THC, which may have reduced the power to detect a main effect of dose, a priori planned comparisons were conducted for the main variables of interest for each measure (see Table 1, asterisked variables) independent of a statistically significant main effect. This approach allowed us to confirm the validity of the drug manipulation (i.e., ensuring that Δ9-THC alone produced expected impairment of memory ability), which is essential for interpreting any potential modulatory effects of α-pinene. For all other variables, planned comparisons were conducted only when main effects of dose or dose × time interactions were significant. For all analyses, statistical significance was defined as an alpha level of <0.05. All statistical analyses were conducted and graphs were made using R [38].

Table 1.

Mean peak change from baseline values for pharmacokinetic measures by THC dose and planned comparisons

​	0 mg THC+		30 mg THC+
	0 mg α-pinene	15 mg α-pinene	0 mg α-pinene	0.5 mg α-pinene	5 mg α-pinene	15 mg α-pinene
	mean (SD)	mean (SD)	mean (SD)	mean (SD)	mean (SD)	mean (SD)
Subjective measures
DEQ
Drug effect	−1.2 (16.9)1	5.1 (19.7)	67.2 (27.4)1	66.1 (35.1)	61.5 (37.0)	62.9 (38.0)
Unpleasant	−0.1 (5.0)1	0.1 (3.0)	19.4 (24.3)1	24.2 (30.9)	20.4 (26.8)	18.1 (21.8)
Pleasant	3.1 (24.6)1	11.2 (28.2)	59.4 (36.9)1	58.6 (36.0)	59.4 (38.9)	55.7 (36.4)
Like	6.3 (25.6)1	13.3 (29.2)	57.1 (35.2)1	56.8 (32.2)	53.9 (36.2)	50.1 (42.0)
Sick	−0.8 (4.4)	−0.1 (4.7)	6.6 (11.4)	9.8 (14.6)	5.8 (13.2)	6.4 (11.3)
Heart racing	−1.8 (11.3)1	−4 (12.1)	23.3 (22.0)1	22.0 (22.2)	19.2 (24.3)	26.1 (28.4)
Anxious/nervous	2.2 (18.8)	−3.7 (6.1)	11.9 (18.7)	16.1 (26.5)	18.5 (22.6)	16.7 (19.6)
Relaxed	−7.2 (32.0)	−13.8 (25.4)	−9.5 (41.0)	6.4 (35.5)	3.1 (38.1)	4.2 (37.2)
Paranoid	−0.1 (3.7)1	−0.3 (2.0)	7.8 (12.5)1	8.5 (15.7)	10.0 (17.1)	12.5 (19.7)
Sleepy	6.3 (29.6)1	10.4 (31.3)	28.6 (39.9)1	28.5 (32.3)	23.8 (33.2)	26.2 (37.0)
Alert	−5.4 (26.7)	−12.0 (28.6)	−19.4 (34.0)	−21.1 (33.4)	−14.8 (45.2)	−24.8 (38.5)
Irritable	−0.37 (8.3)	4.8 (13.8)	6.4 (16.9)	5.7 (14.2)	3.4 (7.04)	9.6 (17.6)
Vigorous/motivated	−8.1 (25.2)	−7.2 (20.7)	−8.2 (38.3)	−9.7 (40.2)	−19.6 (38.6)	−18.4 (31.9)
Restless	2.5 (10.3)	2.6 (21.2)	3.0 (20.8)	13.1 (22.2)	20.5 (21.4)	12.5 (21.2)
Hungry/munchies	7.9 (31.8)1	13.2 (27.3)	38.4 (35.5)1	43.7 (35.7)	42.6 (31.5)	36.6 (38.4)
Craving	7.6 (22.3)	0.7 (19.1)	10.0 (23.6)	7.8 (13.2)	7.8 (18.0)	7.7 (20.8)
Dry mouth	−1.5 (8.7)1	−0.5 (8.6)	34.6 (28.9)1	37.2 (22.7)	36.4 (21.4)	36.7 (27.9)
Dry/red eyes	−1.5 (11.0)1	0.0 (12.7)	20.7 (22.5)1	19.9 (20.8)	20.2 (28.3)	20.3 (32.1)
Trouble with memory2	−0.6 (7.9)1	1.7 (8.4)	24.5 (28.3)1	28.2 (26.2)	29.3 (26.2)	21.6 (29.7)
Throat irritation/coughing	−2.0 (6.6)1	1.6 (9.1)	22.7 (22.1)1	25.6 (26.9)	20.0 (21.8)	21.9 (25.5)
Difficulty performing routine tasks	0.2 (6.2)	0.1 (5.5)	30.8 (34.4)	32.2 (28.9)	27.3 (29.2)	24.6 (32.6)
POMS
Anger/hostility	0.1 (1.0)	0.3 (0.8)	0.4 (1.1)	0.4 (1.0)	0.5 (0.9)	0.6 (1.3)
Confusion/bewilderment2	0.6 (1.6)1	0.4 (0.9)	2.2 (2.7)1	2.4 (2.5)	1.8 (2.1)	2.9 (3.3)
Depression/dejection	0.3 (0.7)	0.2 (0.7)	0.5 (1.1)	0.5 (1.5)	0.4 (0.8)	0.7 (1.3)
Fatigue	0.6 (1.8)	1.2 (1.8)	2.6 (5.2)	2.5 (3.6)	1.8 (3.5)	3.1 (4.6)
Tension/anxiety	0.1 (1.3)	−0.7 (1.5)	1.2 (2.2)	1.4 (2.8)	1.8 (2.5)	1.9 (2.3)
Vigor/activity	−1.0 (2.9)	−1.5 (3.2)	−1.6 (5.8)	−1.7 (4.7)	−2.4 (4.5)	−2.4 (5.1)
Total mood disturbance	2.8 (5.1)	2.5 (5.7)	8.4 (13.0)	7.6 (8.7)	7.2 (7.6)	10.4 (12.1)
Cognitive measures
ISLTa
Retrieval2	7.4 (2.5)1	7.8 (2.2)	5.3 (2.9)1	6.6 (3.0)	6.3 (3.2)	5.6 (3.2)
Encoding2	4.4 (2.4)1	4.2 (2.3)	2.8 (2.2)1	3.7 (1.8)	2.9 (2.6)	2.6 (2.2)
PASAT
Total correct2	1.6 (10.9)1	−4.2 (10.4)	−12.6 (16.9)1	−11.2 (13.2)	−8.2 (11.9)	−11.1 (15.8)
% correct	1.8 (12,1)1	−4.6 (11.6)	−14.0 (18.7)1	−12.4 (14.7)	−9.1 (13.2)	−12.3 (17.6)
Reaction time correct, ms	−88.1 (148.0)	−23.8 (150.4)	60.1 (237.3)	4.1 (399.7)	26.5 (184.2)	75.6 (164.5)
Reaction time incorrect, ms	−54.7 (1,175.6)	142.0 (1,334.3)	256.7 (1,149.8)	324.9 (1,047.5)	571.3 (1,185.6)	33.3 (1,174.6)
DSST
Total correct2	1.8 (10.3)	2.5 (12.6)	−8.7 (13.7)	−5.1 (14.7)	−5.7 (11.8)	−7.5 (12.9)
% correct	−2.0 (13.6)	4.0 (24.1)	−11.1 (24.6)	−8.8 (22.7)	−10.4 (21.4)	−9.6 (21.9)
Physiological measures
HR, beats/min2	1.9 (18.7)1	−2.3 (13.0)	29.9 (17.3)1	31.1 (19.1)	34 (17.1)	21.5 (25.2)
DBP, mm Hg	0.6 (14.3)	−9.3 (9.9)	−3.5 (14.9)	−4.4 (17.3)	1.3 (20.4)	−2.2 (13.3)
Systolic blood pressure, mm Hg	4.8 (13.5)	−2.3 (17.1)	−7.5 (19.5)	−12.9 (26.5)	−4.6 (18.1)	−1.9 (15.9)

SD, standard deviation; DEQ, drug effect questionnaire; DSST, digit symbol substitution task; PASAT, paced serial addition task; ISLT, international shopping list test.

¹Planned comparisons showed placebo condition significantly differed from THC alone condition.

²Primary outcome measures.

^aThe ISLT was administered only once per session, preventing calculation of peak-change from baseline scores; therefore, raw mean differences across dose conditions were examined.

Results

Participants

A total of 48 individuals provided informed consent and were screened for the study. Of these individuals, 33 (17 males and 15 females, and 1 non-binary individual) were eligible and randomized. Of those randomized, 19 (10 males, 8 females, and 1 non-binary individual) completed all six study conditions and were included in data analyses. Of the 14 who were excluded from data analyses, six participants completed pilot testing for initial safety and did not receive all final dose conditions), two participants were discontinued due to adverse effects associated with the study drug (both after 30 mg Δ9-THC + 15 mg α-pinene), four were lost to follow-up during study participation, one was discontinued due to reporting anxiety prior to dosing upon arrival for the 2nd study session, and one was discontinued due to failure to perform adequately on cognitive tests at baseline after repeated task instruction.

The racial/ethnic breakdown of the final study sample (n = 19) was: 58% Caucasian/Non-Hispanic, 32% African American/Non-Hispanic, 5% Middle Eastern/Non-Hispanic, and 5% Asian/Non-Hispanic. On average, participants smoked 1 tobacco cigarette/day (SD = 5) and had 2 standard alcoholic drinks/day (SD = 2). Participants, on average, had not used cannabis for 39 days (SD = 84; range = 2–365) prior to their first test session. Their mean (SD) BMI was 29 kg/m² (SD = 4), their mean weight was 188 lbs (SD = 31) or 85 kgs (SD = 14), and their mean age was 30 years old (SD = 9; range: 22–53 years old). Analysis of dose inhalation behavior showed no difference between dose conditions on vaporizer puff topography (see Table 2).

Table 2.

Mean topography values by THC dose

​	0 mg THC +		30 mg THC +
	0 mg α-pinene	15 mg α-pinene	0 mg α-pinene	0.5 mg α-pinene	5 mg α-pinene	15 mg α-pinene
	mean (SD)	mean (SD)	mean (SD)	mean (SD)	mean (SD)	mean (SD)
Topography outcomes
Mean Puff duration, s	4.9 (2.4)	5.1 (1.9)	3.8 (1.3)	3.9 (1.7)	3.9 (1.8)	3.9 (1.8)
Mean flow rate, mL/s	57.7 (13.8)	54.9 (10.1)	56.2 (11.4)	55.6 (9.0)	56.0 (11.8)	51.6 (7.8)
Mean IPI, s	12.3 (6.4)	13.7 (8.2)	21.3 (13.2)	21.1 (13.8)	19.6 (10.8)	17.6 (9.0)
Total puffs	15.3 (2.3)	15.3 (0.6)	20.6 (6.8)	22.5 (7.5)	23.6 (6.7)	23.2 (6.2)
Total inhaled volume, mL	4,183.9 (1,999.9)	4,219.5 (1,397.5)	4,211.9 (1,858.1)	4,613.1 (2,143.9)	5,112.8 (3,332.4)	4,606.4 (2,431.4)
Mean Puff volume, mL	281.4 (144.8)	277.0 (94.2)	213.3 (89.5)	212.3 (83.7)	219.4 (114.6)	198.8 (89.7)
Maximum puff volume, mL	393.2 (201.4)	368.3 (117.0)	327.5 (143.1)	336.6 (145.6)	328.4 (177.3)	291.8 (124.4)

IPI, Inter-puff interval.

Subjective Drug Effects

Outcomes of interest for subjective drug effects related to memory impairment included the DEQ item “trouble with memory” and the POMS domain “confusion/bewilderment.” Planned comparisons revealed a significant difference in peak-change from baseline scores between the placebo and Δ9-THC alone conditions for both items but did not detect any significant differences between the placebo and α-pinene conditions or the Δ9-THC alone and Δ9-THC + α-pinene combination conditions (see Table 1). Figure 1 shows the (1) time course for mean subjective ratings of and (2) mean peak change from baseline scores for DEQ “trouble with memory” and POMS “confusion/bewilderment” composite score across dose conditions.

Fig. 1.

Time course for mean (SEM) subjective ratings of the Drug Effects Questionnaire (DEQ) item “Trouble with Memory” and Profile of Mood States (POMS) item “Confusion/Bewilderment”. Mean (SEM) peak change from baseline ratings for the DEQ item “Trouble with Memory” and the POMS item “Confusion/Bewilderment.” DEQ scores ranged from 0 (not at all) to 100 (extremely). POMS scores ranged from 0 (not at all) to 20 (extremely). Gray bars indicate conditions with active THC administration.

In addition to these two items, planned comparisons revealed a significant difference in peak change from baseline scores between the placebo and Δ9-THC-alone conditions for 10 DEQ items, which included drug effect, pleasant drug effect, drug liking, heart racing, paranoia, sleepiness, increased appetite, dry mouth, dry eyes, and throat irritation (see online suppl. Table 1 for repeated-measures ANOVA results, which determined the planned comparisons conducted). However, no significant differences were observed on any additional POMS domains. Further, there were no differences between the placebo and α-pinene alone conditions or the Δ9-THC alone and Δ9-THC + α-pinene combination conditions for any of these variables (see Table 1).

Cognitive Performance

Outcomes of interest for cognitive performance included ISLT retrieval and encoding, PASAT total number of correct trials, and DSST total number of correct trials. Planned comparisons showed that word recall in the Δ9-THC alone condition was worse compared with the placebo condition for both lists, but no differences were detected between the placebo and α-pinene alone conditions or the Δ9-THC alone and Δ9-THC + α-pinene combination conditions (see Table 1). Figure 2 illustrates retrieval and encoding scores on the ISLT across dose conditions.

Fig. 2.

Data for retrieval and encoding scores on the International Shopping List Test (ISLT). Gray bars indicate conditions with active THC administration.

Planned comparisons revealed a significant difference between the placebo and Δ9-THC alone conditions for peak change from baseline total number of correct trials on the PASAT, but not the DSST (see Table 1). No significant differences were observed between the placebo and α-pinene conditions or the Δ9-THC alone and Δ9-THC/α-pinene combination conditions for the total number of correct trials on the DSST or PASAT. Planned comparisons revealed no significant differences between any dose conditions for peak change from baseline reaction time to correct or incorrect responses on the PASAT (see online suppl. Table 1 for repeated-measures ANOVA results). Figure 3 illustrates the mean peak change from baseline in total correct scores for the DSST and PASAT across dose conditions.

Fig. 3.

Mean (SEM) peak change from baseline ratings for total correct on the Digit Symbol Substitution Task (DSST) and the Paced Serial Addition Task (PASAT). A decrease in total correct indicates an impairment of cognitive/psychomotor function. Gray bars indicate conditions with active THC administration.

Vital Signs

Planned comparisons revealed a significant increase in HR after Δ9-THC alone compared with placebo, but not between the placebo and α-pinene alone conditions or the Δ9-THC alone and Δ9-THC + α-pinene combination conditions (see Table 1). Figure 4 illustrates the mean peak change from baseline data (i.e., beats per minute) for HR across conditions. There were no main effects or dose × time interactions on blood pressure assessments (see online suppl. Table 1).

Fig. 4.

Mean (SEM) peak change from baseline data (i.e., beats per minute) for HR. Gray bars indicate conditions with active THC administration.

Adverse Effects

In total, there were 12 AEs spontaneously reported by participants in the study, none of which were considered unanticipated or serious. Participants experienced dizziness/lightheadedness (7 instances; three in the 30 mg Δ9-THC condition, two in the 30 mg Δ9-THC + 15 mg α-pinene condition, one in the 30 mg Δ9-THC + 5 mg α-pinene, and one in the 30 mg Δ9-THC + 0.5 mg α-pinene) and/or anxiety-like effects (7 instances; three in the 30 mg Δ9-THC condition, two in the 30 mg Δ9-THC + 15 mg α-pinene condition, one in the 30 mg Δ9-THC + 5 mg α-pinene, and one in the 30 mg Δ9-THC + 0.5 mg α-pinene). Two participants were withdrawn from the study due to AEs; one who experienced dizziness, shallow breath, and sedation following drug administration and the other who experienced dizziness, nausea, and emesis following drug administration (both in the 30 mg Δ9-THC + 15 mg α-pinene condition).

Discussion

Despite the widespread popularization of the “entourage effect” theory, little controlled clinical research has systematically evaluated the interactions between Δ9-THC and the specific terpenes frequently highlighted in cannabis industry marketing and product labeling. Results from this study showed that Δ9-THC alone induced cognitive and memory impairments compared to placebo, consistent with previous literature [27–29, 39], but the co-administration of α-pinene with Δ9-THC did not attenuate Δ9-THC-induced cognitive impairments as hypothesized. Additionally, α-pinene alone did not elicit any pharmacodynamic effects at 0.5–15 mg doses, indicating no significant cognitive-enhancing properties, when compared with placebo. While no prior clinical or preclinical studies have examined the co-administration of pure α-pinene and Δ9-THC on memory or cognitive effects, these results differ from studies showing cognitive benefits from α-pinene or α-pinene-containing essential oil administration in both human and animal models [24–26]. Furthermore, regarding the co-administration of terpenes and Δ9-THC, these findings differ from a similar terpene/Δ9-THC co-administration study [20], which found that simultaneous administration of vaporized d-limonene (a terpene purported to have anxiolytic properties) and Δ9-THC attenuated Δ9-THC-induced anxiety in a dose-orderly manner compared with Δ9-THC alone.

The null findings from the co-administration of α-pinene and Δ9-THC in this study have important implications for cannabis products, including their marketing (i.e., advertising of terpene profile), and formulation and usage recommendations. Notably, the doses of α-pinene tested in this study ranged from amounts that would be expected to be found in retail cannabis flower to amounts far exceeding those found in retail cannabis flower [40, 41]. These findings suggest that even when using doses much higher than those found in retail products, α-pinene did not mitigate Δ9-THC-induced cognitive impairments. This is particularly important given that retailers often promote cannabis products to consumers based on their terpene profiles [12], such as suggesting that α-pinene may have cognitive-enhancing properties. Although α-pinene plays no direct role in legal determinations of impairments, such claims may lead individuals to believe erroneously that products containing α-pinene may reduce impairment and consequently engage in risky behaviors, such as driving. These considerations are important when informing regulations surrounding the development and marketing of cannabis products.

There are several limitations of the present study’s design that are important to consider in light of the observed null findings. First, this study used pure Δ9-THC and α-pinene rather than whole-plant cannabis, which often contains varying concentrations of additional terpenes and minor cannabinoids beyond α-pinene. It is possible that α-pinene alone may not exert significant effects in isolation but may instead contribute to synergistic interactions within whole-plant or “full-spectrum” cannabis products [10, 42]. Therefore, future studies are needed that examine the effects of α-pinene in combination with other naturally occurring chemical constituents in the cannabis plant to determine whether other terpenes or compounds enhance its potential to mitigate Δ9-THC-induced cognitive deficits. Second, this study used vaporized Δ9-THC and α-pinene, which may differ from other routes of administration (e.g., smoked, sublingual, or oral ingestion). This is particularly relevant as research indicates that people who use cannabis medicinally most commonly do so via ingested oils, gummies, baked goods, and/or beverages [43] and typically prefer oral ingestion over inhalation [44, 45]. Another important consideration is that the present study required participants to exhale (following Δ9-THC/α-pinene inhalation) into a handheld smoke filter to obscure the visibility of the resultant vapor and preserve the study blind; this practice, while important for maintaining the study blind, may have prevented olfactory drug exposure or other sensory cues typically present during cannabis inhalation in real-world settings that are important for memory promoting effects of α-pinene. It is important to acknowledge that this study had a small sample size (n = 19), making it more difficult to draw conclusions about populations with different demographics, consumption patterns, or cannabis use histories. Additionally, the laboratory setting may not accurately reflect real-world cannabis consumption patterns and could have influenced expectancy effects. All that said, the strengths of the study include a rigorous within-subjects design, controlled dosing, and expected impacts of Δ9-THC on pharmacodynamic outcomes. Therefore, we are confident of the validity of the present results, but recommend additional research to examine the interactions between α-pinene and Δ9-THC in naturalistic settings, with larger/different samples, and perhaps alternative doses, product formulations, and/or routes of administration.

In summary, the present controlled human laboratory study was the first to examine whether the cannabis terpene α-pinene attenuated Δ9-THC-induced cognitive impairment in a dose-orderly fashion. Although Δ9-THC alone demonstrated cognitive impairments compared to placebo, these findings indicate that α-pinene did not mitigate Δ9-THC-induced cognitive impairments or influence other common acute subjective, cognitive, or physiological effects of Δ9-THC in this sample of healthy adult occasional cannabis users. Moreover, when inhaled alone, α-pinene did not produce any acute effects that differed from placebo, but this study did establish the safety of direct acute inhalation of α-pinene doses up to 15 mg in healthy adults using a vaporizer. Future research is needed to evaluate the modulatory effects of α-pinene under conditions that explore its interactions with other naturally occurring cannabis constituents, assess alternative methods of administration, include diverse user groups (e.g., regular users, abstinent individuals, users of synthetic cannabis or CBD), and incorporates more naturalistic use behavior (e.g., without masking exhaled vapor). Although α-pinene did not attenuate Δ9-THC-induced cognitive impairments in this study, the findings highlight the importance of using a controlled, dose-ranging approach to systematically assess entourage effects. By varying Δ9-THC and terpene exposure and examining their interaction on validated pharmacodynamic outcomes, this methodology provides a model for future research into Δ9-THC-terpene combinations and their potential therapeutic applications. As the use of cannabis expands for both medicinal and non-medicinal purposes, a deeper understanding of how individual cannabis constituents influence Δ9-THC’s effects – either by enhancing its safety profile (e.g., reducing acute cognitive and memory impairments), exacerbating adverse effects, or having no effect – is important for advancing the development of cannabinoids in medicine and protecting public health.

Acknowledgments

We thank the research, medical, nursing, and pharmacy support staff of the Johns Hopkins University Behavioral Pharmacology Research Unit. We also thank True Terpenes, THC Pharm GmbH, and Storz and Bickel GmbH for material support.

Statement of Ethics

This study protocol was approved by the JHU School of Medicine Institutional Review Board (IRB00085652), approved by the US Food and Drug Administration (IND140339), and was registered on ClinicalTrials.gov (NCT03609853). Written informed consent was obtained from participants to participate in the study.

Conflict of Interest Statement

Dr. Tory Spindle has served as a consultant for Canopy Health Innovations Inc. and has received research funding from Cultivate Biologics. Dr. Ryan Vandrey has served as a consultant or received honoraria from Mira1a Therapeutics Inc., Jazz Pharmaceuticals, Charlotte’s Web, Syqe Medical Ltd., and WebMD. Dr. Ethan Russo is the founder and CEO of CReDo Science and a scientific advisor to True Terpenes. The study was conceived and designed by Drs. Vandrey and Russo. Drs. Kumar, Spindle, Vandrey, and Zamarripa had full access to the data and worked together to write the first draft of the manuscript and each take responsibility for the integrity of the reported data. Remaining authors contributed to the manuscript by reviewing the completed draft and providing critical feedback and edits. Johns Hopkins played no role in the production of this manuscript. The remaining authors have no conflicts of interest to declare.

Funding Sources

This research was supported by the National Institute on Drug Abuse (NIDA; R01 DA043475, T32DA07209 and P50 DA056408). The opinions, findings, conclusions, and recommendations expressed in this publication are those of the authors and do not necessarily reflect those of NIDA.

Author Contributions

Lakshmi Kumar: writing – original draft, writing – review and editing, formal analysis, and visualization. Tory R. Spindle: writing – review and editing, writing – original draft, conceptualization, and supervision. C. Austin Zamarripa: writing – review and editing and formal analysis. Harrison J. Elder: writing – review and editing. Ethan B. Russo: writing – review and editing. George Bigelow: writing – review and editing, funding acquisition, and conceptualization. Ryan Vandrey: writing – review and editing, writing – original draft, funding acquisition, conceptualization, and supervision.

Funding Statement

This research was supported by the National Institute on Drug Abuse (NIDA; R01 DA043475, T32DA07209 and P50 DA056408). The opinions, findings, conclusions, and recommendations expressed in this publication are those of the authors and do not necessarily reflect those of NIDA.

Data Availability Statement

All data generated or analyzed during this study are included in this article and its supplementary material files. Further inquiries can be directed to the corresponding author.