The endocannabinoid system (ECS) is a neuro-modulatory system that influences many physiological functions, including pain processing and modulation ^[1][9]. ECS is composed of: endocannabinoids (eCBs), their receptors (CB) and their synthetic and catabolic enzymes. Arachidonoylethanolamide (AEA) and 2-arachidonoyl-glycerol (2-AG) are the most studied eCBs and act primarily on two isoforms of CB: type 1 (CB1) and type 2 (CB2) ^[2][3][10,11]. CB1 is primarily expressed in neurons, whilst CB2 is mainly expressed in immune cells ^[4][12]. It also seems plausible that other eCBs receptors are involved, particularly type 3 CB (CB3), known as GPR55 ^[5][13], and the transient receptor potential vanilloid 1 (TRPV1) ion channel ^[6][14]. AEA binds both receptor subtypes, with higher selectivity for CB1 than for CB2 ^[7][15], whilst 2-AG activates both receptor subtypes as a full agonist ^[8][16]. AEA and 2-AG are synthesized from lipid precursors and subsequently released from postsynaptic neurons into the synaptic space ^[9][17]. The synthesis of AEA is catalyzed by N-acylphosphatidylethanolamine-phospholipase D (NAPE-PLD) ^[10][18], whilst that of 2-AG is catalyzed by sn-1-specific diacylglycerol lipase (DAGL) ^[11][19]. After their release, eCBs are retrieved via an endocannabinoid membrane transporter (EMT), and AEA is degraded by fatty acid amide hydrolase (FAAH), whereas 2-AG is degraded by monoglyceride lipase (MAGL) ^[12][20]. After their release from postsynaptic neurons, eCBs stimulate presynaptic CB1, balancing GABAergic inhibitory activity and glutamatergic excitatory activity. In addition, eCBs may act autocrinally on CB1 and interact with other neurotransmitters, such as dopamine, regulating intrinsic neuronal activity ^[12][20]. In 2006, Russo suggested the “clinical endocannabinoid deficiency” (CED), as low levels of eCBs had been reported in painful conditions such as fibromyalgia and migraine ^[13][21]. Evidence from several preclinical studies seems to indicate that the dysregulation of ECS, with reduced eCB activity, plays a role in migraine. The key publication by Akerman et al. ^[14][22] demonstrated that AEA decreases trigeminovascular system excitability (primarily involved in a migraine attack) in nitroglycerin (NTG)-induced migraine models; in contrast, the most recent preclinical studies have mainly focused on blocking FAAH and MAGL activities, but a complete dissertation of these publications is beyond the scope of this paper (see [23,24,25] for an overview). Clinically, Cupini et al. demonstrated increased activity of FAAH and EMT in platelets in female migraineurs, but not in males ^[15][26]. This increased activity, which is not found in tension-type headache sufferers, drove a reduction in the level of AEA and may suggest an imbalance in eCB degradation in women affected by migraine without aura. The same group subsequently observed a significant reduction in FAAH and EMT activity in chronic migraine (CM) and medication overuse headache (MOH) sufferers, the latter being a complication resulting from the frequent use of medicines to treat migraine ^[16][27]. However, as the researcheuthors had previously observed ^[15][26], FAAH and EMT activities were higher in female sufferers of episodic migraine (EM) than in healthy controls and the CM-MOH group. This may be attributable to an adaptive response induced by chronic headache and/or drug overuse ^[16][27]. These results were later confirmed by another study, showing that AEA and 2-AG levels were significantly lower in the peripheral platelets of CM sufferers compared to healthy controls ^[17][28]. The reduced levels were more evident in females than in males. However, Gouveia-Figueira et al. detected no significant variations in the plasma levels of AEA in EM sufferers ^[18][29]. In the cerebrospinal fluid, Sarchielli et al. found that AEA was lower in CM sufferers than healthy controls. In addition, palmitoylethanolamide (PEA) was significantly higher in CM sufferers than in healthy controls, suggesting that higher levels of PEA might represent a compensatory response to the reduced ECS tone in CM ^[19][30]. In a positron emission tomography (PET) study, Van der Schueren et al. demonstrated that the binding of a specific CB1 ligand ((18F)MK-9470) to CB1 was augmented in pain-modulating brain areas in the interictal period in female migraineurs compared to controls, suggesting an eCB deficiency ^[20][31]. Perrotta et al. found that FAAH activity was significantly reduced after the withdrawal of painkillers, coinciding with clinical improvement. These data were interpreted as indicative of a relationship between AEA levels and the anti-nociceptive effect ^[21][32]. Migraine genome-wide association studies did not find specific genetic variants within the ECS ^[22][23][33,34], but Juhasz et al. identified an association between CB1 gene variants and headache with nausea, especially in patients subjected to recent stressful events, indicating the possible role of ECS in patients suffering from life-stress-triggered migraine attacks ^[24][35]. Greco et al. found higher CB1 and CB2 levels in mononuclear cells of EM and CM-MOH sufferers compared to healthy controls ^[25][36], in accordance with the results on CB1 binding activity in EM ^[20][31]. FAAH gene expression was lower in both EM and CM-MOH compared to healthy controls. These data are similar to those reported by Cupini et al. for CM-MOH patients ^[16][27] but dissimilar to those reported for EM patients ^[15][16][26,27], even though lower levels of FAAH were detected in CM-MOH in the latter study when compared to the EM group. This indicates a possible dynamic compensatory mechanism to maintain higher AEA levels in a challenged system. NAPE-PLD and DAGL mRNAs were increased in EM and CM-MOH sufferers vs. controls. NAPE-PLD and DAGL mRNA levels were also higher in CM-MOH vs. EM subjects. All of these findings suggest a compensatory mechanism to relieve an eCB deficiency. MAGL mRNAs were also increased in EM and CM-MOH patients: this result, apparently surprising, may be indicative of a higher turnover of 2-AG. Interestingly, all of these changes in the gene expression of different components of the ECS were associated with migraine days. This supports the notion that a dysregulation of the ECS is present in migraine and correlates with the seriousness of migraine. Last year, the plasma levels of AEA and PEA were evaluated in a double-blind, parallel-group clinical study of migraine provocation after receiving sublingual nitroglycerin (0.9 mg).

Phytocannabinoids are a group of substances that display a cannabinoid structure and are found in the Cannabis sativa plant. To date, about 60 different phytocannabinoids have been described, and their number is still increasing ^[26][7], with Δ9-tetrahydrocannabinol (THC) and cannabidiol (CBD) being the most studied ones. Phytocannabinoids have similar chemical formulas but distinct properties that separate them from one another. Despite not having any conclusive scientific evidence, medical cannabis is frequently used by migraine sufferers as a last-resort self-treatment ^[27][38]. In a survey-based study conducted in 9003 patients, 121 patients claimed to use cannabis for migraine relief. Interestingly, most of these patients inhaled cannabis, often without informing their general practitioner ^[28][39]. In a survey of 145 patients, Aviram et al. found that medical cannabis resulted in a long-term reduction in migraine frequency in >60% of treated patients and was associated with reduced medication intake and less disability ^[29][40]. Another study conducted in 589 adult cannabis users reported that migraine sufferers experienced significant migraine relief using medical cannabis ^[30][41]. Rhyne et al. retrospectively evaluated the effects of medical cannabis in 121 EM sufferers attending two medical marijuana specialty clinics in Colorado (United States), reporting a global decrease in migraine frequency ^[31][42]. However, most of these patients used different formulae of marijuana, even on the same day, and through different routes of administration ^[31][42]. Another study explored the effect of different oral formulae of phytocannabinoids in CM sufferers. Patients reported a reduction in pain severity and analgesic consumption after 3 and 6 months of use compared to the baseline, but there were no changes in the number of headache days. The researcheuthors concluded that, considering the tonic regulatory role of the ECS, this result may indicate that phytocannabinoids are more useful in pain intensity, rather than frequency ^[32][43]. An online survey conducted in 1429 medical cannabis users found that the consumption of phytocannabinoids for migraine treatment often occurred without a physician’s supervision ^[33][44]. Despite the reported benefits of cannabis, its therapeutic effects on migraine are influenced by its formulae as well as its route of consumption ^[34][45]. Moreover, different cannabinoid formulae may also have different pharmacokinetics, even if taken by the same route ^[35][46]. In the above-mentioned study, patients took differently titrated cannabis forms through different routes, thus making it almost impossible to understand which phytocannabinoid is really effective on migraine. The only study that focused on a single route of administration and on three standardized/titrated cannabis forms is the one conducted by Baraldi et al., but the small sample size affected the results ^[32][43]. Additionally, the possibility of developing a tolerance exists, since Cuttler et al. found that migraine patients who inhaled medical cannabis used higher doses over time ^[36][47]. Another question yet to be answered is if cannabis consumption can lead to the development of MOH. In a preclinical migraine model, the infusion of a cannabinoid receptor agonist, such as THC or WIN55,212-2, seemed to induce latent trigeminal sensitization, thus raising the possibility of MOH development ^[37][48]. Moreover, a retrospective study conducted through an electronic chart review found a slight but significant association between MOH and cannabis consumption ^[38][49]. No randomized clinical trials investigating phytocannabinoids have been reported in migraine patients. Two trials have been initiated (NCT03972124 and NCT04360044) ^[39][40][50,51], but no results have been published yet.