Cannabis Medicines: Guidance for the Selection, Purchase and Supply for Clinical Trials

Abstract

Cannabis medicines are in demand from the public for treating a range of diseases and symptoms; however, clinicians are reluctant to prescribe these products because of limited evidence and prescribing information. To generate this evidence, quality clinical trials of cannabis medicines must be undertaken, yet their design is a complex, often uncharted territory, and involves the cooperation and sharing of knowledge of multiple stakeholders. Before designing a clinical trial, researchers require a clear understanding of the potential therapeutic benefit cannabis medicines may have, the form and formulation of the product, and the dose to be investigated. Researchers must also be aware of the applicable pharmaceutical regulations in the country or jurisdiction where the research is to be undertaken, as well as manufacturing or licensing regulations that may be imposed at the source of the cannabis product. Importantly, collaborations with industry are a key to the successful outcome of cannabis medicines clinical trials. Without funding and sponsorship of clinical trials, the ability to generate quality data will be limited and the evidence for cannabis medicines to be registered as therapeutics lacking. Collaborations between researchers, industry, and regulators, working together in sharing knowledge, are therefore critical to generate high quality cannabis medicines research.

1. Introduction

Since the Federal legalization of cannabis medicines in Australia in 2016, cannabis has rapidly moved from being a recreational drug to a medicinal product. While it shows great therapeutic promise, one of the issues faced by clinicians in prescribing cannabis is the limited information available from high quality clinical trials across a broad range of indications. In order to generate this evidence, collaboration and sharing of knowledge between all stakeholders will be required to progress quality clinical trials of cannabis medicines. This paper discusses several issues that investigators have found when designing clinical studies using cannabis medicines as investigational products. These include cannabis medicine selection according to the indication(s) being studied, dosage form, dose range, drug-drug interactions, regulatory considerations, purchase and supply, purity and consistency of plant-based products, and industry engagement.

2. Selection of cannabis medicines for indications

Over 120 phytocannabinoids have been identified in Cannabis sativa L. that have a diverse range of molecular targets [1]. Consideration of the type of cannabinoid and its pharmacokinetics and pharmacodynamics is important when selecting products for clinical trials [2]. To date, research has focused heavily on cannabidiol (CBD) and ∆-9-tetrahydrocannabinol (THC). Cannabis medicine clinical trials can be located on clinical trial registries, such as the Australian New Zealand Clinical Trials Registry (ANZCTR), ClinicalTrials.gov and WHO International Clinical Trials Registry Platform (ICTRP) [3, 4, 5]. The selection of cannabis medicines for a particular indication in the context of a clinical trial requires researchers to consider several different factors.

Some trials have investigated a combination of cannabinoids. For example, in palliative care [6] and glioblastoma [7], THC and CBD have been trialed in the ratios of 1:1 and 4:1, respectively. Trials of CBD alone have been conducted in anxiety and schizophrenia [8] and refractory epilepsy (Dravet syndrome and Lennox-Gastaut syndrome) [9]. Nabiximols, an oral spray containing roughly equal parts CBD and THC and marketed under the trade name Sativex, has been studied in spasticity in multiple sclerosis [10], cannabis dependence [11], and neuropathic pain [12]. Dronabinol, an entirely synthetic form of THC has been trialed for anorexia and weight loss in patients with acquired immune deficiency syndrome [13] while dronabinol and nabilone, also a synthetic cannabinoid that is similar to THC but appears to be more potent, have both been studied in treatment-refractory chemotherapy-induced nausea and vomiting [14]. Dementia studies have trialed THC, dronabinol, and nabilone [15].

These are only a few examples of numerous trials for different clinical indications and disease states. Many of these studies have had mixed results suggesting that proposed clinical trials, especially for diseases not already studied will require a thorough examination of in vitro and preclinical work to inform the selection of the cannabinoid formulation most appropriate for the disease under study.

To this end, there is an abundance of in vitro and in vivo studies being conducted on a variety of cannabinoids and in different formulations that over time, will help identify the principal components required of a cannabinoid for the disease being targeted [1]. As always, a detailed literature search is essential.

3. Selection of the form of cannabis medicines

Several factors need to be considered when selecting a cannabis medicine for use in a clinical trial. Key features include the age of the patient cohort where participants may have limited dexterity in handling certain formulations, likely comorbid conditions that may affect drug absorption, and manufacturers’ ability to create a placebo that looks, tastes, and smells the same as the investigational product.

Several different dosage formulations can be selected. Table 1 describes the potential advantages and disadvantages of common dosage forms of cannabis medicines that may be considered for use in a clinical trial. Less common dosage forms such as rectal, vaginal, and intravenous are not discussed as they are generally less practical forms for a clinical trial in comparison with the forms discussed below [16].

Route of delivery	Dosage Form	Advantages	Disadvantages
Inhaled	Smoked or vaporized	Rapid onset peak plasma concentrations 3–10 minutes) [2] Higher bioavailability (10–35%) [2, 17] May be suitable for those with difficulty administering other formulations Bypasses first pass metabolism [16]	Difficult to determine the total amount of product inhaled or absorbed, leading to unintended bias Difficult to reliably titrate dose when dose inhaled is uncertain [16] May be less portable depending on device used to vaporize product Bystanders exposed to inhaled product Matched placebo comparatively challenging to formulate Smell of the product is noticeable
Sublingual	Sprays, oils, tinctures, wafer	Rapid onset Higher bioavailability Liquid version amendable to fine titration Suitable for patients with swallowing difficulties Bypasses first metabolism [16]	Alcohol-based products may cause pain in participants with mucositis Participants may not be able to reliably hold sublingual product under tongue for an adequate duration Bottles and syringes required are challenging for the elderly who may have poor eyesight or arthritis Difficult to formulate an identical placebo Potential for spillage of liquid product Taste of product is noticeable Best absorption after meals [18]
Oral	Capsules, tablets, oils, tinctures	Easiest to create matched placebo Capsule or tablet version have minimal issues with taste Liquid version amenable to fine titration by volume Acid resistant coating for encapsulated product can avoid gastric degradation Can be packed into blister packs to aid administration for those with dexterity issues Measurable quantity is consumed reducing confounding from inadequate consumption	Slower onset of action [16, 19] Longer duration to peak plasma concentration (120 minutes) [2, 16, 19] Lower bioavailability [16, 19] Undergoes first-pass metabolism [16] Not suitable with gastrointestinal absorption issues (e.g. colitis, intestinal obstruction) or difficulty swallowing Packaging in tamper-proof bottles may cause difficulty for the elderly who may have dexterity issues Bottles and syringes required are challenging for those with poor eyesight or arthritis Potential for spillage of liquid product Taste of liquid product is noticeable
Topical	Creams, patches	Higher bioavailability [16] Steady dose delivery over a prolonged time period [16] Easy to administer [16] May be helpful for local symptoms (e.g. pain) Comparatively less challenging to create a matching placebo Suitable for most patients as comorbid conditions are less likely to affect absorption Bypasses first-pass metabolism [16, 20]	Possible skin reaction to drug or adhesive Patches may prematurely burst open Need for education on correct techniques of application (e.g. site rotation, avoiding unsuitable sites) Not suitable for participants with widespread skin conditions

Table 1.

Advantages and disadvantages of common dosage forms of Cannabis medicines when used in a clinical trial setting.

4. Dose selection

Currently, dosing information is only available for approved cannabis medicines, dronabinol, nabilone, nabiximols and Epidiolex (CBD). For unregistered cannabis medicines, there is no precise dosing recommendation. Pharmacokinetic variability of cannabinoids is very high both among and between cohorts and hence, dosing is highly individualized and dependant on the patient’s condition [21, 22]. Generally, the approach for cannabis dosing is to start low and go slow. The patient should be started on a low dose and gradually titrated until a therapeutic effect without any undesired side effect is achieved [23, 24, 25, 26].

Most patients take oral cannabis medicines 2–3 times per day. Epidiolex, a cannabis-derived form of CBD, is taken twice daily with a starting dose of 5 mg/kg/day up to a maximum dose of 20 mg/kg/day [27], while nabiximols is taken as 4 to 8 sprays/day (1 spray is equivalent to 2.7 mg THC and 2.5 mg CBD) up to a maximum of 12 sprays/day [28]. Frequency is dependent on the duration of action, which is in the order of 3–4 hours for inhaled products and 8–12 hours for oral products.

THC-dominant products can be taken at bedtime for days 1–2 to minimize undesirable daytime side effects such as dizziness or drowsiness and encourage tolerance of doses beginning at 2.5 mg of THC. If the dose is tolerated, the dosing can be doubled every 1–2 days until any undesired side effect(s) are experienced. In this event, patients are advised to revert to their previous dose [23].

CBD dominant products can be used at higher concentrations than THC products because they produce fewer adverse effects. Doses of CBD between 1 and 50 mg/kg/day improve psychotic symptoms, seizures, and anxiety [29]. An average CBD dose of 15 mg/kg/day showed positive significant reductions of seizure while CBD between 150 and 600 mg/day produced therapeutic effects in social anxiety disorder and insomnia. The maximum tolerated dose for CBD in humans is 1500 mg/day [30]. This data shows that CBD-dominant products have a higher therapeutic index than THC-dominant products. Patients are advised to keep a journal of their cannabis medicine dosing together with a record of their symptoms to aid in determining the optimal CBD dose for their particular condition.

5. Drug-drug interactions

There is variable evidence indicating other drugs interact with cannabis medicines, ranging from hypothetical concepts to documented clinical trial evidence [31, 32]. Interactions among drugs is particularly relevant in trials where a variety of pharmacologic products is being investigated. Drug-drug interactions may increase the active concentration of cannabinoids, enhancing the possibility of adverse effects, or they may decrease cannabinoid concentrations, compromising their physiologic effects [33]. The converse effect might be expected from the competing drug.

Examples of drug-drug interactions include blood pressure-lowering medications, warfarin [33], antiepileptic drugs (e.g. clobazam) [34], and central nervous system depressant medications, including opioids and benzodiazepines [35]. Researchers should keep in mind that drug-drug interactions can occur with cannabis medicines.

6. Regulatory, purchase and supply

Clinical trials place stringent demands upon the availability of precise and reproducible formulations of the medicines under study. Unfortunately, the quality and consistency of supply of many cannabis medicines are not of a standard that would facilitate their use in most clinical trials. Indeed, while plant genetics can be tightly regulated through cloning techniques to minimize variability amongst cannabis plants, multiple factors can still alter the phytochemical profile of plant-based medicines, including environmental factors, time of harvest, manufacturing processes and storage conditions (reviewed in [36]). In Australia, to overcome this the Therapeutic Goods Administration (TGA) has introduced guidelines for cannabis medicines and more recently the U.S. Food and Drug Administration (FDA) has also introduced guidelines [37, 38]. Cannabis medicines used in clinical trials must align with the existing framework for the use of medications in clinical trials, and meet the requirements for human use of cannabis medicines set by the local regulator e.g. FDA, European Medicines Agency (EMA), TGA [39, 40, 41]. Local pharmaceutical, prescribing and holding regulations, approvals, and requirements for labelling, transport, and storage of investigational medicinal product for use in a clinical trial must also be confirmed prior to selecting and purchasing a product for a clinical trial [42, 43, 44].

Australia’s TGA approves the use of medicines in a clinical trial through their Clinical Trial Notification (CTN) scheme [41, 45]. Prescribed cannabis medicines must conform to Therapeutic Goods Orders No. 93 (Standard for Medicinal Cannabis) (TGO 93) and TGO 100 (Microbiological quality of medicinal cannabis products), among others [37, 46, 47]. Adherence to Good Manufacturing Process (GMP), and to federal and state pharmacy regulations, policies, and their respective drugs and poisons legislation is required [48, 49, 50, 51, 52]. Where the manufacture of cannabis medicine products will form part of a clinical trial, a license authorising manufacture for clinical trial use must be in place between the manufacturer and the TGA [44, 53].

Following product feasibility, risk review, and final selection, supply contracts should be developed for the purchase of the product(s) for the clinical trial, to document agreed requirements, roles, and responsibilities including:

• Formulation (dose, volume, and form)

• Packaging and labeling requirements (if required) must comply with Pharmaceutical Inspection Convention and Pharmaceutical Inspection Co-operative Scheme (PIC/S) Annex 13 [43, 44]

• Consistency of supply (i.e. volume, consistency and purity of product supplied across the duration of the clinical trial) with an agreement to provide certificates of analysis with each batch to ensure purity and consistency in phytochemical profile

• Availability of stock

• Cost

• Any regulatory fees, import permits as required, shipping costs

• Insurance and indemnity

• Access to data if appropriate (this may be an ethics committee decision)

• Pharmacovigilance reporting responsibilities

• Confirmation that all products under trial will meet and be maintained in compliance with International Conference for Harmonisation of technical requirements for pharmaceuticals for human use Good Clinical Practice (ICH- GCP) R6E2 requirements

An essential consideration of any cannabis clinical trial is the stability of the cannabis medicines and the robustness both of their supply chain and the methods used to assess their stability [54]. It is imperative that any medication under study is available via a reliable and continuous supply for the duration of the study and for any ethically approved post-study period. This is particularly important when utilizing plant-based sources that can have inherent variability compared to chemically-synthesized medicines of mainstream pharmaceutical products.

As the legal production of GMP-certified cannabis medicines that are suitable for clinical trials is limited to certain countries, securing appropriate supply may involve transportation across international and state borders. Export and import logistics, licensing, permits, and quarantine requirements may be a consideration to ensure adequate supply in a timely manner for the trial.

7. Engagement with industry

A diverse stakeholder cohort underpins all clinical trials. In cannabis research, most stakeholders—clinicians, researchers, regulatory bodies, and the medicinal cannabis industry—support the need for robust clinical evidence that informs the use of cannabis derived therapeutics. In Australia as in other countries, cooperation amongst stakeholders is essential to ensure the challenges associated with a complex regulatory environment are appropriately addressed. Furthermore, stakeholders have the responsibility of managing industry expectations and delivering a level of patient recruitment that leads to successful clinical trials as well as fostering more efficient and effective collaborations.

A lack of global consistency on quality standards presents a different challenge. Establishing collaborations and implementing common agreements with local industry provides a framework to share knowledge especially when accessing the GMP cannabis medicines produced in Australia. Moreover, importation of products for clinical trials requires authorization, first from the country to which the product will be imported and subsequently from the country from which the product will be exported. Industry stakeholders may be highly responsive and supportive, but response timelines of regulators on both ends of the importing process can be prolonged and unreliable and may impact continuity of supply for longer-term trials. In addition, the requirement to comply with local standards that are geographically specific and often unique to cannabis medicines, means that the supplier must be both willing and capable of meeting those standards. In practical terms, this requirement can constrain a research team to source cannabis medicines only from companies with an established local operation with experience in the current regulations. Without this experience the impost of ensuring compliance shall fall either on the supplier or the research project and is likely to be cost-prohibitive.

8. Conclusion

The design of clinical trials for cannabis medicines is a complex process. Before embarking on a clinical trial, researchers need to have a clear understanding of the potential therapeutic benefit cannabis medicines may have. This understanding will inform the selection of cannabis product, the formulation of the product and the dose to be tested. Researchers must also be aware of regulations that are applicable where the research is to be undertaken as well as regulations that may be imposed at the source of the cannabis product. The phytochemical consistency of plant-derived products can be solidified with further research and may assist the approval of additional botanical products to the cannabis medicines market, increasing options for clinicians and patients, as plant-based products may be preferred to chemically-synthesized or bioengineered medicines [55]. Importantly, collaborations with industry are key to the successful outcome of cannabis medicines clinical trials. Without significant investment and sponsorship of clinical trials, the ability to generate quality data will be limited and the evidence for cannabis medicines to be registered as therapeutics lacking. Collaborations between researchers, industry and regulators, working together in sharing knowledge will generate high quality cannabis medicines research.

Acknowledgments

We would like to thank the members of the Australian Centre for Cannabinoid Clinical and Research Excellence (ACRE) Guidance Writing Group for Clinical Trials in Cannabinoids: Selection, Purchase and Supply of Cannabis Medicines as Study Drugs in Clinical Trials members Prof Janet Hardy (Mater Cancer Care Centre) and A/Prof Phillip Good (Mater Cancer Care Centre) for their contribution to the development of this paper.

Author Disclosure Statement

Dr Peter Galettis, Dr Rachel Galettis and Mrs Courtney Hill are part of the Australian Centre for Cannabinoid Clinical and Research Excellence (ACRE), which is funded through the National Health and Medical Research Council’s Centres of Research Excellence Program. ACRE also receives funding through NSW Health’s Clinical Cannabis Medicines Program.

John Barlow is a consultant to Applied Cannabis Research.

Jaroslav Boublik is Chief Science Officer of LeafCann Group Pty Ltd, a privately held Australian federally licensed medicinal cannabis production company.

Myfanwy Graham is funded through NSW Health’s Clinical Cannabis Medicines Program.

Melinda Thompson is an alternate board member for the Medicinal Cannabis Industry Australia (MCIA)

Aaron Wong is Primary Investigator in a medicinal cannabis trial which is funded by the Victorian Cancer Agency.

SC, JSa, JS, PM, RR, LT, KW, & BZ – No competing financial interests exist