Cannabinoids and Pain: New Insights From Old Molecules.

Phytocannabinoids

There are about 100 different cannabinoids isolated from the cannabis plant (Andre et al., 2016). The main psychoactive compound is (-)-trans-Δ9-tetrahydrocannabinol (THC), which is produced mainly in the flowers and leaves of the plant. The THC content varies from 5% in marijuana to 80% in hashish oil. THC is an analog to the endogenous cannabinoid, anandamide (ananda is the Sanskrit word for bliss; arachidonoylethanolamide, AEA). It is responsible for most of the pharmacological actions of cannabis, including the psychoactive, analgesic, anti-inflammatory, anti-oxidant, antipruritic, bronchodilatory, anti-spasmodic, and muscle-relaxant activities (Rahn and Hohmann, 2009; Russo, 2011). THC acts as a partial agonist at cannabinoid receptors (CB1 and CB2) (Pertwee, 2008). A very high binding affinity of THC with the CB1 receptor appears to mediate its psychoactive properties (changes in mood or consciousness), memory processing, motor control, etc. It has been reported that a number of side effects of THC, including anxiety, impaired memory and immunosuppression, can be reversed by other constituents of the cannabis plant (cannabinoids, terpenoids, and flavonoids) (Russo and Guy, 2006; Russo, 2011; Andre et al., 2016).

The non-psychoactive analog of THC, cannabidiol (CBD), is another important cannabinoid found in the cannabis plant. It is thought to have significant analgesic, anti-inflammatory, anti-convulsant and anxiolytic activities without the psychoactive effect of THC (Costa et al., 2007). CBD has little binding affinity for either CB1 or CB2 receptors, but it is capable of antagonizing them in the presence of THC (Thomas et al., 2007). In fact, CBD behaves as a non-competitive negative allosteric modulator of CB1 receptor, and it reduces the efficacy and potency of THC and AEA (Laprairie et al., 2015). CBD also regulates the perception of pain by affecting the activity of a significant number of other targets, including non-cannabinoid GPCRs (e.g., 5-HT1A), ion channels (TRPV1, TRPA1 and TPRM8, GlyR), PPARs, while also inhibiting uptake of AEA and weakly inhibiting its hydrolysis by the enzyme fatty acid amide hydrolase (FAAH) (Russo et al., 2005; Staton et al., 2008; Ahrens et al., 2009; De Petrocellis et al., 2011; Burstein, 2015; Morales et al., 2017). It has been demonstrated that cannabidiol can act synergistically with THC and contribute to the analgesic effect of medicinal-based cannabis extract (Russo, 2011). At the same time, CBD displays an entourage effect (the mechanism by which non-psychoactive compounds present in cannabis modulate the overall effects of the plant), and is capable of improving tolerability and perhaps also the safety of THC by reducing the likelihood of psychoactive effects and antagonizing several other adverse effects of THC (sedation, tachycardia, and anxiety) (Russo and Guy, 2006; Abrams and Guzman, 2015). The differences in concentration of THC and CBD in the plant reflect the differences in the effects of different cannabis strains. Although CBD as a monotherapy in the treatment of pain has not been evaluated clinically, its anti-inflammatory (Ko et al., 2016) and anti-spasmodic benefits and good safety profile suggest that it could be an effective and safe analgesic (Wade et al., 2003).

Other phytocannabinoids that can contribute to the overall analgesic effects of medical cannabis are cannabichromene (CBC), cannabigerol (CBG), tetrahydrocannabivarin (THCV), and many others (Morales et al., 2017). Similarly to CBD, these compounds do not display significant affinities for cannabinoid receptors, but they have other modes of action. This is a new area of research that needs to be addressed (Piomelli et al., 2017).

Endocannabinoid System

This system seems to regulate many functions in the body, including learning and memory, mood and anxiety, drug addiction, feeding behavior, perception, modulation of pain and cardiovascular functions. The endocannabinoid system consists of cannabinoid receptors, endogenous cannabinoids (endocannabinoids), transport proteins and enzymes that synthesize or degrade the endocannabinoids.

Cannabinoid CB1 and CB2 receptors are 7-transmembrane G-protein coupled receptors (GPCRs). They play an important role in peripheral, spinal, and supraspinal nociception, including ascendant and descendent pain pathways (Hill et al., 2017). The signal transduction pathway of CB1 and CB2 involves inhibition of adenylyl cyclase, decreased cAMP formation, as well as an increase in the activity of mitogen-activated protein kinases (MAPK) (Ibsen et al., 2017). New evidence is emerging that different ligands can differentially activate these pathways, suggesting biased signaling through the cannabinoid receptors CB1 and CB2 (Ibsen et al., 2017).

The CB1 receptor is distributed throughout the nervous system. It mediates psychoactivity, pain regulation, memory processing and motor control. CB1 is a presynaptic heteroreceptor that modulates neurotransmitter and neuropeptide release and inhibits synaptic transmission. Activation of CB1 results in the activation of inwardly rectifying potassium channels, which decrease presynaptic neuron firing, and in the inhibition of voltage-sensitive calcium channels that decrease neurotransmitter release (Morales et al., 2017). The CB1 receptor is strategically located in regions of the peripheral and CNS where pain signaling is intricately controlled, including the peripheral and central terminals of primary afferent neurons, the dorsal root ganglion (DRG), the dorsal horn of the spinal cord, the periaqueductal gray matter, the ventral posterolateral thalamus and cortical regions associated with central pain processing, including the anterior cingulate cortex, amygdala and prefrontal cortex (Hill et al., 2017). The principal endogenous ligand for the CB1 receptor is AEA. CB1 receptors are observed more often on the gamma-aminobutyric acid (GABA) inhibitory interneurons in the dorsal horn of the spinal cord, and weakly expressed in most excitatory neurons (Hill et al., 2017). CB1 receptors are also present in multiple immune cells such as macrophages, mast cells and epidermal keratinocytes.

The CB2 receptor is found predominantly at the periphery (in tissues and cells of the immune system, hematopoietic cells, bone, liver, peripheral nerve terminals, keratinocytes), but also in brain microglia (Abrams and Guzman, 2015). The receptors are responsible for the inhibition of cytokine/chemokine release and neutrophil and macrophage migration and they contribute to slowing down of chronic inflammatory processes and modulate chronic pain (Niu et al., 2017). Both CB2 and CB1 receptors on mast cells participate in the anti-inflammatory mechanism of action of cannabinoids (Facci et al., 1995; Small-Howard et al., 2005). Also, activation of CB2 receptors on keratinocytes stimulates the release of β-endorphin, which acts at μ opioid receptors on peripheral sensory neurons to inhibit nociception (Ibrahim et al., 2005). Under basal conditions, CB2 receptors are present at low levels in the brain, the spinal cord and DRG, but may be upregulated in microglia where they modulate neuroimmune interaction in inflammation and after peripheral nerve damage (Hsieh et al., 2011). CB2 receptor activation inhibits adenylyl cyclase activity and stimulates MAPK activity, but the effect on calcium or potassium conductance is controversial (Rahn and Hohmann, 2009; Atwood et al., 2012). Stimulation of CB2 receptors does not produce cannabis-like effects on the psyche and circulation. The principal endogenous ligand for the CB2 receptor is 2-arachidonoylglycerol (2-AG) (Kano, 2014).

Endocannabinoids are arachidonic acid derivatives. AEA and 2-AG are synthesized separately, they have local (autocrine and paracrine) effects and are rapidly removed by hydrolysis by fatty acid amide hydrolase (FAAH) and monoacylglycerol lipase (MAGL), respectively (Pacher et al., 2006; Starowicz and Przewlocka, 2012; Howard et al., 2013). Beside AEA, FAAH inhibition significantly elevates the levels of other fatty-acid amides such as oleoylethanolamide (OEA) and palmitoylethanolamide (PEA) in the CNS and peripheral tissues (Lambert et al., 2002). Endocannabinoids, similarly to THC, appear to activate cannabinoid receptors. AEA and 2-AG are a partial and full agonist of CB receptors, respectively (Kano, 2014). They work as a part of a negative feedback loop that regulates neurotransmitter and neuropeptide release and thereby modulate various CNS functions, including pain processing (Vaughan and Christie, 2005).

The AEA is a full agonist at TRPV1 (AEA referred to as an ‘endovanilloid’) that activates TRPV1 which results in desensitization (Ross, 2003; Horvath et al., 2008; Starowicz and Przewlocka, 2012). AEA also activates GR55 (Ryberg et al., 2007), directly inhibits 5-HT3A receptors (Barann et al., 2002) potentiates the function of glycine receptors (Hejazi et al., 2006), inhibits T-type voltage-gated calcium channels (Chemin et al., 2001) and activates PPARs (Rockwell and Kaminski, 2004; Sun et al., 2007; Romano and Lograno, 2012; O’Sullivan, 2016).

Endocannabinoids, which are produced in neural and non-neural cells in the physiological response to tissue injury or excessive nociceptive signaling, suppress inflammation, sensitization and pain (Piomelli and Sasso, 2014; Maccarrone et al., 2015). Inhibitors of FAAH lead to elevated AEA levels and are intended for therapeutic use (Hwang et al., 2010). N-acylethanolamines such as PEA and OEA do not belong to endocannabinoids as they do not bind to cannabinoid receptors; they exhibit anti-inflammatory action via PPARs, and also inhibit pain through TRPV1 receptors. They are of interest to the field of cannabinoid pain research as they elevate levels of AEA through substrate competition at FAAH (Lambert et al., 2002).

There is a constant active exchange of substrates and metabolites between endocannabinoid and eicosanoid pathways. The enzyme FAAH breaks down AEA to arachidonic acid and ethanolamine or, alternatively, AEA can be directly transformed by cyclooxygenase-2 (COX-2) into proalgesic prostaglandins. As such, AEA may contribute to the analgesic properties of COX-2 selective NSAIDs. It was established that the metabolite of paracetamol combines with arachidonic acid by the action of FAAH to produce an endocannabinoid, which is a potent agonist at the TRPV1 and a weak agonist at both CB1 and CB2 receptors and an inhibitor of AEA reuptake (Bertolini et al., 2006).

Synthetic Cannabinoids

At present, there are two synthetic cannabinoids on the market, dronabinol and nabilone, which may be of benefit in the treatment of pain (Abrams and Guzman, 2015). In general, their use in pain treatment is off-label. Dronabinol is a generic name for the oral form of synthetic THC (Marinol^®). It is approved for the treatment of chemotherapy-associated nausea and vomiting, and anorexia associated with human immunodeficiency virus infection. Nabilone, a generic name for the orally administered synthetic structural analog of THC (Cesamet^®), is approved for the treatment of chemotherapy-associated nausea and vomiting. Their medical use is mostly limited by their psychoactive side effects, as well as their limited bioavailability (Huestis, 2007).

Neuropathic Pain

Cannabinoids have been studied in various types of neuropathic pain in animals, including chronic nerve constriction traumatic nerve injury, trigeminal neuralgia, chemotherapy- and streptozotocin-induced neuropathy, etc.

Both CB1 and CB2 receptors have been found to be upregulated in nervous structures involved in pain processing in response to peripheral nerve damage (Lim et al., 2003; Zhang et al., 2003; Hsieh et al., 2011), and this may explain the beneficial effects of cannabinoid receptor agonists on neuropathic pain. It has been shown that increased CB2 expression is accompanied by the appearance of activated microglia (Zhang et al., 2003). Both microglial activation and neuropathic pain symptoms can be suppressed by CB2 agonists (Wilkerson et al., 2012). Consistent with this, CB2 knockout mice and transgenic mice overexpressing CB2 are characterized by enhanced and suppressed reactivity of microglia and neuropathic pain symptoms, respectively (Racz et al., 2008). TRPV1 expression is also increased in glutamatergic neurons of the medial prefrontal cortex in a model of spared nerve injury (SNI) in rats (Giordano et al., 2012).

In different neuropathic pain conditions, systemic administration of synthetic mixed cannabinoid CB1/CB2 agonists produces antinociceptive effects similar to those of THC (Herzberg et al., 1997; Pascual et al., 2005; Liang et al., 2007). The CB2 selective agonists given intrathecally or systemically are also effective in several animal models of neuropathic pain (Yamamoto et al., 2008; Kinsey et al., 2011), but their antinociceptive effects are without development of tolerance, physical withdrawal and other CNS side effects that accompany CB1 agonism (Deng et al., 2015).

When given early in the course of diabetes, CBD attenuates microgliosis in the ventral lumbar spinal cord of diabetic mice, as well as tactile allodynia and thermal hyperalgesia. However, if given later in the course of the disease, CBD has a little effect on pain-related behavior (Toth et al., 2010).

A controlled cannabis extract containing numerous cannabinoids and other non-cannabinoid fractions such as terpenes and flavonoids demonstrated greater antinociceptive efficacy than the single cannabinoid given alone, indicating synergistic antinociceptive interaction between cannabinoids and non-cannabinoids in a rat model of neuropathic pain (Comelli et al., 2008). The anti-hyperalgesic effect did not involve the cannabinoid receptors but was mediated by TRPV1 and thus it most probably belongs to CBD.

In animals with neuropathic pain, increased levels of endocannabinoids (AEA and 2-AG) have been detected in different regions of the spinal cord and brain stem (Mitrirattanakul et al., 2006; Petrosino et al., 2007). However, they appeared to be differentially regulated in different models of neuropathic pain, depending on the characteristic of the pain and the affected tissues (Starowicz and Przewlocka, 2012). Genetic or pharmacological inactivation of FAAH/MAGL resulting in the elevation of endocannabinoid (AEA/2-AG) levels in the spinal cord and brain stem (Lichtman et al., 2004; Schlosburg et al., 2009; Long et al., 2009; Adamson Barnes et al., 2016) show promise for suppressing both neuropathic and inflammatory pain. In general, the antinociceptive effect of endocannabinoids is sensitive to antagonists of CB1 and CB2 receptors, TRPV1 channels and PPARα antagonism, indicating that multiple targets could be involved in the mechanism of their action (Kinsey et al., 2010; Caprioli et al., 2012; Piomelli, 2014; Adamson Barnes et al., 2016). The reduction in the side effects that accompany CB1 agonism, such as motor incoordination, catalepsy, sedation and hypothermia, suggests that mainly TRPV1, but not a cannabinoid receptor-dependent mechanism, mediate the analgesic properties of exogenously and endogenously elevated levels of AEA in neuropathic pain. In a rat chronic constriction injury (CCI) model, depending on the dose of URB597 (FAAH inhibitor) used, lower or higher elevation of endogenous AEA levels and CB1- or TRPV1-mediated analgesia were achieved, respectively (Starowicz et al., 2012). It has been suggested that endocannabinoids can increase the excitability of nociceptive neurons by reducing synaptic release of inhibitory neurotransmitters via CB1 receptors on dorsal horn neurons (Pernía-Andrade et al., 2009), as well as by agonist activity on TRPV1 (Ross, 2003).

Monoacylglycerol lipase inhibitors demonstrated CB1-dependent behavioral effects, including analgesia, hypothermia and hypomotility (Long et al., 2009). In a mouse model of neuropathic pain both CB1 and CB2 were engaged in the anti-allodynic effects of FAAH inhibitors, while only CB1 was involved in the anti-allodynic effect of the MAGL inhibitor (Kinsey et al., 2010). Also, unlike FAAH inhibitors, the persistent blockade of MAGL activity leads to desensitization of brain CB1 receptors and loss of the analgesic phenotype (Chanda et al., 2010) and physical dependence (Schlosburg et al., 2009). However, a new highly selective MAGL inhibitor, KML29, exhibited antinociceptive activity without cannabimimetic side effects (Ignatowska-Jankowska et al., 2014).

In CCI in mice, JZL195, a dual inhibitor of FAAH and MAGL, demonstrated greater anti-allodynic efficacy than selective FAAH or MAGL inhibitors, and a greater therapeutic window (less motor incoordination, catalepsy and sedation) than WIN55212, a cannabinoid receptor agonist (Adamson Barnes et al., 2016).

Co-administration of sub-threshold doses of FAAH inhibitor, PF-3845 and the non-selective COX inhibitor, diclofenac sodium, produced enhanced antinociceptive effects in rodent models of both neuropathic (CCI) and inflammatory pain (intra-plantar carrageenan) (Grim et al., 2014). Combined FAAH inhibition/TRPV1 antagonism is also an attractive therapeutic strategy because FAAH inhibition only produced biphasic effects, with antinociception via CB1 at low levels of AEA, and when AEA levels were higher, pronociceptive effects via TRPV1 (Maione et al., 2007; Malek and Starowicz, 2016).

Cannabinoids may attenuate neuropathic pain by peripheral action via both CB1 and/or CB2 receptors (Fox et al., 2001; Elmes et al., 2004). The peripherally acting cannabinoid agonist AZ11713908 reduced mechanical allodynia with a similar efficacy to WIN55,212-2, an agonist that entered the CNS (Yu et al., 2010). In addition, URB937, a brain impermeant inhibitor of FAAH, elevated anandamide outside the brain and controlled neuropathic pain behavior without producing CNS side effects (Clapper et al., 2010).

After identification of allosteric binding site(s) on the CB1 GPCR (Price et al., 2005), several CB1-positive allosteric modulators have been developed and tested in animals. They attenuated both inflammatory and neuropathic pain behavior without producing the CB1-mediated side effects of orthosteric CB1 agonists but did not produce tolerance after repeated administration (Khurana et al., 2017; Slivicki et al., 2017).

Inflammatory Pain

Different classes of cannabinoids (i.e., CB1 agonists, CB2 agonists, mixed CB1/CB2 agonists, endocannabinoids and endocannabinoid modulators) all suppressed pain behavior in various animal models of inflammatory pain (Clayton et al., 2002; Burgos et al., 2010; Starowicz and Finn, 2017). Since inflammatory pain is a characteristic of several chronic diseases, including cancer, arthritis, inflammatory bowel disease, sickle-cell disease, etc., cannabinoids appear to promise the lessening of severe pain in these diseases (Fichna et al., 2014; Abrams and Guzman, 2015; Turcotte et al., 2016; Vincent et al., 2016).

It is well known that CB2 receptor expression increases in microglia in response to inflammation and serves to regulate neuroimmune interactions and inflammatory hyperalgesia (Dhopeshwarkar and Mackie, 2014). However, the extent of CB2 expression in neurons is a subject of controversy (Atwood and Mackie, 2010; Atwood et al., 2012). It has been suggested that peripheral inflammation, unlike peripheral nerve injury, does not induce CB2 receptor expression in the spinal cord (Zhang et al., 2003). In contrast, Hsieh et al. (2011) demonstrated that the CB2 receptor gene is significantly upregulated in DRG and paws ipsilateral to inflammation induced by injection of complete Freund’s adjuvant (CFA).

Systemic or local peripheral injection of the CB2-selective agonist was reported to reduce nociceptive behavior and swelling in different animal models of inflammation (Quartilho et al., 2003; Elmes et al., 2005; Kinsey et al., 2011). In addition, the CB2-selective agonist did not produce hypothermia or motor deficit that are CB1-mediated side effects (Kinsey et al., 2011). Therefore, a CB2 receptor selective agonist is expected to have less psychomimetic side effects and lower abuse potential as compared to the available non-selective or CB1-selective cannabinoid agonists. In animal models of inflammatory disease, CB2 agonists slow the progression of diseases (Turcotte et al., 2016). In a murine model of rheumatoid arthritis, collagen-induced arthritis (CIA), CB2-selective agonists did not prevent the onset of arthritis, but did ameliorate established arthritis (Sumariwalla et al., 2004). JWH133, a selective CB2 agonist, inhibited in vitro production of cytokines in synoviocytes and in vivo reduced the arthritis score, inflammatory cell infiltration and bone destruction in CIA (Fukuda et al., 2014). Another CB2-selective agonist, HU-308, was shown to reduce swelling, synovial inflammation and joint destruction, in addition to lowering circulating antibodies against collagen I in CIA (Gui et al., 2015). Although approved in a range of preclinical models of pain, LY2828360, CB2 agonist, failed in a trial of patients with osteoarthritic knee pain (Pereira et al., 2013).

It was shown that formalin administration to the hind paw of rats induced AEA release into the periaqueductal gray matter (Walker et al., 1999). FAAH knockout mice and mice that express FAAH exclusively in nervous tissue, displayed anti-inflammatory and anti-hyperalgesic effects in both the carrageenan and CIA models, and the effects were prevented by administration of a CB2 but not a CB1 antagonist (Lichtman et al., 2004; Kinsey et al., 2011). FAAH inhibition may also reduce nociceptive behavior induced by lipopolysaccharide injection into the rat hind paw, and examination of the mechanism showed that both CB1 and CB2 were involved, but not TRPV1, PPARs, or opioid receptors (Booker et al., 2012). Oral administration of PF-04457845, a highly efficacious and selective FAAH inhibitor, produced potent antinociceptive effects in the CFA model of arthritis in rats, and it was shown that both CB1 and CB2 receptors were implicated in this effect (Ahn et al., 2011). In contrast to animal data, PF-04457845 failed to demonstrate efficacy in a randomized placebo and active-controlled clinical trial on pain in osteoarthritis of the knee (Ahn et al., 2011; Huggins et al., 2012). The possible explanations are development of tolerance to chronically elevated endocannabinoids or sensitization of TRPV1 receptors. A pronociceptive phenotype has been recently documented in FAAH knockout mice after administration of a challenge dose of TRPV1 agonist capsaicin (Carey et al., 2016). The increased nociceptive response was attenuated by antagonists of CB1 and TRPV1 receptors.

In a recent phase I trial, the FAAH inhibitor BIA-102474 caused death in one and severe neurological damage in five participants (Kaur et al., 2016; Moore, 2016). It has been suggested that specificity and non-selectivity of this molecule and several errors in the design of the study were responsible for its toxicity, and not targeting of FAAH per se (Huggins et al., 2012; Pawsey et al., 2016). More research is necessary to characterize both the efficacy and safety profiles of endocannabinoid-directed therapeutic strategies.

An increase in local endocannabinoid levels by inhibition with local peripheral administration of URB597 (an irreversible FAAH inhibitor) induced analgesia in a model of carrageenan-induced inflammation in rats that was inhibited by a PPARα antagonist but not by a CB1 receptor antagonist (Sagar et al., 2008). However, local administration of URB597 into osteoarthritic knee joints reduced pain via CB1 receptors [monosodium iodoacetate (MIA)-induced osteoarthritis in rats and the model of spontaneous osteoarthritis in Dunkin-Hartley guinea pigs] (Schuelert et al., 2011). A peripherally restricted FAAH inhibitor, URB937, also reduced inflammatory pain in rodents via CB1 receptors (Clapper et al., 2010).

It was shown that inhibition of fatty acid binding proteins (FABPs) reduced inflammatory pain in mice. This effect was associated with an upregulation of AEA and the effect was inhibited by antagonists of CB1 or PPARα receptors (Kaczocha et al., 2014).

Recent animal findings suggest that cannabinoids may have beneficial effect on affective-emotional and cognitive aspect of chronic pain (La Porta et al., 2015; Neugebauer, 2015; Kiritoshi et al., 2016). In mice with MIA-induced arthritis, selective agonists of both CB1 and CB2 receptors ameliorated the nociceptive and affective manifestations of osteoarthritis, while a CB1-selective agonist improved the memory impairment associated with arthritis (La Porta et al., 2015; Woodhams et al., 2017). This is in agreement with human studies of cannabinoids that indicate a significant improvement in secondary outcome measures, such as sleep and mood (Lynch and Ware, 2015).

The combined FAAH/COX inhibitor ARN2508 demonstrated efficacy against intestinal inflammation and was without gastrointestinal side effects (Sasso et al., 2015) because AEA, which is similar to prostanoids, has protective actions on the gastrointestinal mucosa.

Efficacy of Cannabis/Cannabinoids in the Treatment of Chronic Pain

Until recently, there was no consensus about the role of cannabinoids for the treatment of chronic pain. Several years ago, the European Federation of Neurological Societies recommended cannabinoids (THC, oromucosal sprays 2.7 mg delta-9-tetrahydrocannabinol/2.5 mg cannabidiol) as the second or third line of treatment of central pain in MS (Attal et al., 2010). More recently, the Canadian Pain Society supported their use as the third-line option for the treatment of neuropathic pain, after anti-convulsives, anti-depressants, and opioids (Moulin et al., 2014). In addition, Health Canada provided preliminary guidelines for prescribing smoked cannabis in the treatment of chronic non-cancer pain (Kahan et al., 2014). At the same time, the Special Interest Group on neuropathic pain of the International Association for the Study of Pain provided “a weak recommendation against the use of cannabinoids in neuropathic pain, mainly because of negative results, potential misuse, abuse, diversion and long-term mental health risks particularly in susceptible” (Finnerup et al., 2015).

There is a growing body of evidence to support the use of medicinal cannabis in the treatment of chronic pain. At present, there is a scientific consensus on the medicinal effects of cannabis for the treatment of chronic pain that is based on scientific evidence. The National Academy of Sciences, Engineering and Medicine (NASEM, 2017) has evaluated more than 10,000 scientific abstracts and established that there is “conclusive or substantial evidence” for the use of cannabis in treating chronic pain in adults. Also, there is “moderate evidence” that cannabinoids, in particular nabiximols, are effective in improving short-term sleep outcomes in patients with chronic pain (NASEM, 2017). The expert report NASEM supports more research to determine dose–response effects, routes of administration, side effects and risk-benefit ratio of cannabis/cannabinoid use with precision and make possible evidence based policy measure implementation. At the same time, the PDQ Integrative Alternative and Complementary Therapies Editorial Board (2018) states that pain relief is one of the potential benefits of cannabis/cannabinoids for people living with cancer (in addition to its anti-emetic effects, appetite stimulation, and improved sleep).