Successful drug delivery to the intended target sites of action is dependent on multiple factors including the individual’s physiology and the drug’s physicochemical properties (solubility, dissolution, stability, permeability and metabolism). Poor bioavailability, which is dependent on these factors, generally leads to insufficient therapeutic efficacy and is more likely to produce high inter-individual variability in pharmacokinetic (PK) parameters [7,8]. In this context, the bioavailability of CBD varies greatly with route and mode of administration [9,10]. The oil/water partition coefficient (Log P) describes a drug preference which exists either in the water or oil part of a solution; higher Log P values mean more of the drug will be distributed in the oil part of the solution. Due to the highly lipophilic nature of CBD (Log P 6.3) [11], it is most commonly supplied as an oil or alcoholic formulation either in soft-gel capsules, liquid solution, sublingual drops, or as an oromucosal spray [12]. Studies examining oral and oromucosal delivery of CBD and THC at equimolar concentrations in humans show high inter/intra-individual variability [13,14].

Highly lipophilic drugs delivered orally in solution can precipitate in the gastrointestinal (GI) tract, resulting in an absorption rate slower than elimination [8]. The oral bioavailability of CBD is estimated at 6%, although data in this area have been noted to be considerably lacking [15]. Time to peak plasma concentration following oral delivery is slow (1–4 h), the C_max (maximum concentration) generated from 20 mg of orally delivered CBD/THC sprayed onto a gelatin capsule was 2.4 ng/mL of CBD, and the half-life of CBD was reported between 1.4 and 10.9 h after oromucosal spray [14,15]. Oil suspensions designed for oral and oromucosal routes of administration are currently favoured; Epidiolex^® is delivered orally in an oil solution, and Sativex as an oromucosal spray. The oromucosal route circumvents some of the problems associated with the oral route, and provides a more rapid onset of action. However, Itin and colleagues suggest that a substantial proportion of the oromucosal delivered dose may actually be absorbed through the GI tract [16]. Authors noted that the PK of Sativex is different when administered in a fasting state relative to a fed state (as is Epidiolex), but that this should not be the case as PK profiles should only be substantially influenced by a fed-state when consumed orally. Furthermore, in work published by Guy and colleagues, the PK parameters of Sativex were similar between oromucosal and oral routes of administration [14]. These observations are in agreement with further work by Itin and colleagues, demonstrating CBD’s lipophilic nature causes it to accumulate in the oral mucosal lining or enter the GI tract upon swallowing [17]. However, the authors suggest this route of administration remains viable, so long as exposure times to the oral mucosal lining are adequate, and a method of preventing washout of the drug by saliva is present.

A potential method to increase oral bioavailability is to administer CBD alongside a high-fat/high-calorie meal [16]. This has been demonstrated in healthy subjects with about four-fold increases in bioavailability compared to fasted, and in adult patients with refractory epilepsy with four- and fourteen-fold increases in area under the curve (AUC) and C_max, respectively, in the fed state compared to fasted [18,19,20]. This is likely due to increased micelle and chylomicron formation making more drugs available for lymphatic transport [21,22]. High fat meals also potentially inhibit the activity of drug efflux transporters present on the apical membrane of enterocytes, and stimulate the release of biliary secretion, which further inhibits efflux transporter activity [23,24]. Although lymphatic transport bypasses the liver into systemic circulation, CBD delivered orally is still subject to first pass metabolism. One study showed the AUC of 7-hydroxy-cannabidiol (7-OH-CBD) is about 40% the level of orally delivered CBD, however participants were also receiving concomitant treatment Clobazam, Valproate or Stripentol [25]. Indeed, impaired hepatic metabolism has been shown to cause an increase in the bioavailability of CBD, and indicates first pass metabolism represents a significant barrier in increasing the bioavailability of CBD [26].

Smoking or vaporisation can overcome some of the problems of CBD delivery. The bioavailability of five individuals after smoking 20 mg of deuterium-labelled CBD was reported at 31 ± 13% (range 11–45%) [27]. Smoking allows for rapid delivery of CBD, and peak plasma levels occurred within 3 min following inhalation (110 ± 55 ng/mL). A more recent study assessing the pharmacokinetics of vaporised CBD (100 mg) in healthy men (n = 6) and women (n = 6) showed a mean C_max of 125.4 ng/mL ± 95.2 in men and 83.7 ng/mL ± 8.8 in women [28]. This study also compared vaporised CBD to orally delivered CBD at the same dose, and found vaporisation produced around a ten-fold increase in C_max. Whilst vaporisation increases the bioavailability of CBD, large standard deviations indicate there is still substantial variability in inter-subject pharmacokinetics. Furthermore, the use of solubilisers required to create a vaporisable solution can result in irritation of the airways and may not be suitable for all patients. Smoking a cannabis-based medicinal product is also not currently allowed under legislation in the UK.

A number of alternative delivery methods by-passing the digestive tract and first pass metabolism are currently under investigation including transdermal and nasal routes, and eye drops, which have been reviewed elsewhere [29].