Management of substance abuse

The health and social effects of nonmedical cannabis use

New WHO publication on cannabis


Chapter 4. Neurobiology of cannabis use

4.1 What do we know?

4.1.1 The psychoactive components and neurobiology of cannabis use

The principal psychoactive component of Cannabis sativa, THC (Iversen, 2012), acts on specific receptors in the brain. These receptors also respond to naturally-occurring cannabinoids (known as endogenous cannabinoids or endocannabinoids) such as anandamide (Iversen, 2012). The endocannabinoids regulate the actions of neurotransmitters that play roles in human and animal cognition, emotion and memory (Cascio & Pertwee, 2012).

Two types of cannabinoid receptors have been identified on which THC acts: type 1 (CB1) and type 2 (CB2) cannabinoid receptors. CB1 receptors are found primarily in the brain where they are most concentrated in regions involved in memory (hippocampus), emotional responses (amygdala), cognition (cerebral cortex), motivation (the limbic forebrain) and motor coordination (cerebellum) (Hu & Mackie, 2015; Iversen, 2012). CB2 receptors are found primarily in the body where they seem to play a role in the regulation of the immune system (Iversen, 2012) and have multiple other functions, including acting on the gastrointestinal tract, liver, heart, muscle, skin and reproductive organs (Madras, 2015). The CB1 receptors play a key role in the psychoactive effects of cannabis. Drugs that block the actions of CB1 receptors block the cannabis high in humans and stop animals from self-administering cannabis (Huestis et al., 2001; Iversen, 2012).

The brain’s dopamine reward pathway contains both CB1 and CB2 receptors. Animal and human studies indicate that these receptors respond to THC by increasing dopamine release, an effect that probably explains the euphoric effects of cannabis. THC produces a smaller dopamine release than cocaine or methamphetamines, but dopamine release happens more quickly with cannabis because cannabis is typically smoked (Volkow, 2015). THC can be detected in plasma within seconds of smoking cannabis and it has a half-life of two hours. Peak plasma levels of THC are around 100 μg/L after smoking 10-15 mg of cannabis over a 5-7 minute period. THC is highly lipophilic and is distributed throughout the body (Moffat AC, Osselton MD, Widdop B, 2004).

Dopamine is involved in the control of cognition, attention, emotionality and motivation (Bloomfield et al., 2014). Cannabis alters time perception and coordination by acting on cannabinoid receptors in the basal ganglia, frontal cortex and cerebellum, which are brain regions involved in motor control and memory. Cannabis also affects psychomotor function. It impairs movement and coordination, manipulation and dexterity, grace, strength and speed. Evidence suggests that recent smoking and/or blood THC concentrations 25 ng/mL are associated with substantial driving impairment, particularly in occasional smokers (Hartman & Huestis, 2013) The effects of cannabis on the cerebellum probably explain the driving impairment produced by cannabis (Volkow et al., 2014a), which is described in detail in section 5.1. Animal and human studies show that cognitive and psychomotor functions are impaired directly after cannabis use (Iversen, 2012) and these impairments can persist for several days after use (Crean, Crane & Mason, 2011; Volkow et al., 2014a).

4.1.2 Neurobiology of long-term cannabis use

The daily use of cannabis over years and decades appears to produce persistent impairments in memory and cognition, especially when cannabis use begins in adolescence (Meier et al., 2012; Volkow et al., 2014a). The neurobiology of the cannabinoid system suggests that these effects may arise because chronic THC use reduces the number of CB1 receptors (i.e. “down-regulates” these receptors) in brain regions that are involved in memory and cognition (Iversen, 2012) Experimental studies suggest that animals exposed to THC during puberty may be more susceptible to these effects of cannabis (Schneider, 2012).

Brain imaging studies comparing school students who are regular long-term cannabis users and non-using students typically find poorer cognitive performance and large decreases in perfusion in the former using SPECT scans (Mena et al., 2013). These changes could partially explain the lower educational attainment and lower grades among chronic cannabis users (Volkow et al., 2014a) and are discussed in more detail in section 6.1.2

Magnetic resonance imaging (MRI) studies have found structural differences between the brains of chronic adult cannabis users and the brains of non-using controls. Changes can be seen in the grey/white matter, in global brain measures (Batalla et al., 2013), and in connectivity (Lopez-Larson, Rogowska & Yurgelun-Todd, 2015). Structural brain abnormalities are seen in CB1-rich areas involved in cognitive functions. In addition, reduced hippocampal volume has been found in neuroimaging studies (Ashtari et al., 2011; Cousijn et al., 2012; Matochik et al., 2005; Yücel et al., 2008). In some studies these reductions persist after abstinence (Ashtari et al., 2011) and have been associated with impaired memory (Lorenzetti et al., 2015). Neuroimaging studies have also found reduced volumes in the amygdala, the cerebellum and frontal cortex in chronic cannabis users (Batalla et al., 2013; Yücel et al., 2008). In a large study population (1574 participants), in which cortical thickness was measured by MRI, an association was found between cannabis use in early adolescence and reduced cortical thickness in male participants with a high polygenic risk score. Adults who have smoked cannabis since adolescence show reduced neuronal connectivity in the prefrontal areas responsible for executive functioning and inhibitory control and in the subcortical networks that are responsible for habits and routines (Volkow et al., 2014a). The precuneous - a node involved in integration of various brain functions such as awareness and alertness – is particularly affected in frequent cannabis users. Long-term cannabis use is hazardous to the white matter of the developing brain, with evidence of axon connectivity damage in three fibre tracts: the hippocampus (right fimbria), the splenium of the corpus callosum, and commissural fibres (which connect the two halves of the cerebral hemispheres). Damage was higher with younger age of onset of regular cannabis use (Volkow et al., 2014a).

The fimbria is a part of the hippocampus involved in learning and memory (Zalesky et al., 2012). These findings are consistent with the observation that impaired memory is a common complaint among cannabis users seeking treatment (Hall, 2015). Recovery of hippocampal connectivity after long-term abstinence has been reported (Yücel et al., 2016). Atypical orbitofrontal functional connectivity patterns were observed in attentional/executive, motor and reward networks in adolescents with heavy cannabis use. These anomalies may be reflected in suboptimal decision-making capacity and increased impulsivity (Lopez-Larson, Rogowska & Yurgelun-Todd, 2015). Chronic cannabis use has also been shown to reduce the brain’s capacity to synthesize or release dopamine (Bloomfield et al., 2014), which could explain why cannabis users have higher scores on negative emotionality (Volkow et al., 2014b).

4.1.3 Neurobiology of prenatal cannabis exposure

Polydrug use makes it difficult to study the effects of cannabis on child development since the effects of other drugs, both illicit and legal, may influence study outcomes. A large multicentre study with over 10 000 pregnant women participating found that polydrug use was common among women who use drugs. Specifically, they found that 93% of all women who used, for instance, cocaine or opiates during pregnancy also used alcohol, tobacco and/or cannabis (Konijnenberg, 2015).

Nevertheless, accumulating evidence suggests that prenatal cannabis exposure may interfere with normal development and maturation of the brain. Children exposed to cannabis in utero demonstrate impaired attention, learning and memory, impulsivity and behavioural problems and a higher likelihood of using cannabis when they mature (Sonon et al., 2015; Noland et al., 2005; Goldschmidt, Day & Richardson, 2000; Goldschmidt et al., 2004; Goldschmidt et al., 2008; Day, Leech & Goldschmidt, 2011).

In animal studies, prenatal exposure to THC shows that it can make the brain’s reward system more sensitive to the effects of other drugs (DiNieri & Hurd, 2012). Human research has suggested that cannabis exposure in utero may alter regulation of the mesolimbic dopamine system in children (DiNieri et al., 2011). Children exposed to cannabis prenatally also have higher rates of neurobehavioural and cognitive impairments (Tortoriello, 2014) that may be related to the impaired formation of axonal connections between neurons during fetal development (Volkow, 2014a). Importantly, the negative effects of prenatal drug exposure may not become apparent until later in development. It is, therefore, essential to follow up cannabis-exposed children long into adolescence, and human research in this domain is still limited, which contrasts with nicotine or alcohol research.

4.1.4 Neurobiology of cannabis effects in adolescence

Accumulating evidence reveals that regular, heavy cannabis use during adolescence is associated with more severe and persistent negative outcomes than use during adulthood. As mentioned in section 3.1.2, the risk of dependence has been estimated at 16% in those who initiated cannabis use in adolescence (Anthony, 2006) and 33-50% in daily cannabis users (van der Pol et al., 2013).

The adolescent brain seems to be more vulnerable to cannabis than the adult brain, and early initiation of heavy use appears to disrupt the trajectory of normal brain development. Heavy or regular adolescent cannabis users manifest a range of cognitive deficits, including impairments in attention, learning and memory, and an inability to switch ideas or responses. These deficits are similar in adults, but in adolescents they are more likely to persist and may recover only after longer periods of abstinence (Fried, Watkinson & Gray, 2005). Earlier onset users show greater impairment in cognitive domains, including learning and memory, attention and other executive functions (Pope et al., 2003; Gruber et al., 2012). Decrements in cognitive function are correlated with initiation of cannabis use during adolescence (Pope et al., 2003).

A recent large-scale longitudinal study followed a large cohort from childhood to 38 years of age and assessed neuropsychological functioning at multiple time points. The study revealed that adolescents who used cannabis weekly, or harboured a cannabis-use disorder before the age of 18, showed larger neuropsychological decline and I.Q. reduction than those who became dependent during adulthood (Meier et al., 2012). The results are consistent with cross-sectional findings in adult populations, and reinforce the conclusion that sustained abstinence may not enable cognitive functional recovery if use was initiated during adolescence. A subsequent re-analysis showed that socioeconomic differences did not account for the sustained loss of I.Q. (Moffitt et al., 2013; Solowij et al., 2011).

As noted in section 4.1.1, the CB1 and CB2 receptors are expressed in the brain and peripheral tissues (Mackie, 2005). In the brain, CB1 receptors are the most abundant of the G-protein coupled receptors and mediate most, if not all, of the psychoactive effects of THC in cannabis. CB2 receptors in the brain also modulate the release of chemical signals primarily engaged in immune system functions (e.g. cytokines). Brain imaging has generally revealed changes in the brains of adolescents or adults who initiated cannabis use during adolescence (Lorenzetti et al., 2013; Bossong et al., 2014; Jacobus & Tapert, 2014). Frequent cannabis use is associated with smaller whole brain and hippocampus size, reduced cortical grey matter, and insular cortical thickness that varies in accordance with level of use (Churchwell, Lopez-Larson & Yurgelun-Todd, 2010; Lopez-Larson et al., 2011). Some studies found correlations between brain changes and deficits in learning and memory (Ashtari et al., 2011). Age of onset of cannabis use is apparently not as important in causing hippocampal shrinkage as the amount or frequency of use (Lorenzetti et al., 2014). Changes in cortical volume may predate and predispose individuals to use cannabis, but this is unlikely for changes in the hippocampus (Cheetham et al., 2012) which seems vulnerable to heavy cannabis use, regardless of age.

Studies in rodents have shown that long-term exposure to cannabinoids during adolescence decreases dopamine release in the brain’s reward regions (Pistis et al., 2004; Schneider, 2012). The effects of early cannabis use on dopamine pathways could possibly, in addition to environmental risk factors, explain the role of cannabis as an apparent “gateway drug” – i.e. a drug whose early use increases the risk of later use of other illicit drugs (see also section 6.1.3.2). Early alcohol and nicotine use can also function as gateways for cannabis use by priming the brain to produce elevated dopamine responses to cannabis and other drugs, though alternative explanations based on the overall susceptibility to drug-taking behaviour and higher accessibility of marijuana cannot be excluded (Volkow, 2014b).

4.1.5 Modifiers of risk: the interplay between genetics and environment

The acute and long-term effects of cannabis use depend on interactions between genetic predispositions and environmental factors (Danielsson et al., 2015). Individuals with certain personality profiles may be more likely to use cannabis – particularly those who score higher on sensation-seeking (Muro & Rodríguez, 2015), extraversion and neuroticism, or on adolescent aggression scales, and those who engage in antisocial behaviour (Hayatbakhsh et al., 2009). See also section 2.1.4 on risk and protective factors.

A meta-analysis of twin studies estimated that, among males, 51% of problematic cannabis use could be attributed to shared genes, 20% to a shared environment and 29% to an unshared environment. Among females, 59% was attributed to genetics, 15% to a shared environment, and 26% to an unshared environment (Verweij et al., 2010).

A gene variant of cannabinoid receptor 1 (CNR1) has been associated with cannabis-related problems among frequent users. This variant appears to moderate the relationship between trait impulsivity and cannabis-related problems. Individuals who frequently use cannabis and who have the CNR1 risk variants have higher trait impulsivity and have a higher risk of developing problems related to their cannabis use (Bidwell et al., 2013).

Gerra and colleagues found that serotonin transporter (5-HTT) gene variants were related to cannabis initiation but the environment played a larger role via the stressful effects of perceived parental neglect, a factor consistently related to initiation of cannabis use (Gerra et al., 2010). Lack of parental control and support increases the probability of cannabis initiation by interacting with emotional stability and extraversion (Creemers et al., 2015).

4.1.6 Areas that require more research

Much of the research on the neurobiological effects of cannabis involves persons who are either still heavy cannabis users or who have only recently stopped using cannabis. This makes it difficult to know if the neurobiological effects, and specifically the cognitive impairments, found in these users improve after a year or more of abstinence. The available limited evidence is mixed. Some studies have found persistent impairments while others have found that impairments improve significantly with prolonged abstinence (Solowij & Pesa, 2012; Meier M et al., 2012).

·      Better studies are needed to assess the degree of cognitive recovery in regular cannabis users after sustained abstinence, as a function of age of onset of use, THC potency, frequency of use and similar parameters.

·      “Reverse translational” research is needed to verify, in animals, whether observed changes in human brain structure or function (e.g. dopamine release) can be replicated using cannabis or THC.

·      Both human and animal studies require confirmation by multiple groups using sufficiently large numbers of subjects to yield robust statistical significance. In research on the effects of prenatal cannabis exposure, it is essential to follow up cannabis-exposed children long into adolescence.

·       There is need for large-scale longitudinal research on adolescents, beginning prior to drug initiation and continuing long into adulthood.