Using real world evidence to optimize care: the case of medical cannabis

KEYWORDS:

• Cannabis based medicinal products (CBMPs)
• real-world evidence (RWE)
• data registries
• T21

1. Introduction

Cannabis has been used medicinally for millennia [1] and, given the recent policy and legal changes in multiple jurisdictions internationally, cannabis-based medicinal products (CBMPs) are being increasingly and widely used to treat an array of medical conditions [2]. These changes have occurred despite limited and sometimes low-quality evidence from clinical trials regarding their efficacy [3,4].

Recognizing gaps in the available evidence on safety and efficacy of CBMPs and capitalizing on the widespread, legal use of these products, multiple research groups have established registry and other observational data collections to explore their safety and effectiveness in real world settings [5–8].

The features of real world data collections make them ideally suited for studying CBMPs and may also help explain the apparent discrepancy between randomized controlled trial (RCT) and real world evidence (RWE) derived conclusions [9]. Specifically, it has been noted that results on the efficacy/effectiveness of medicinal cannabis derived from observational studies seem more promising than those derived from clinical trials [4]. While RCTs are typically considered the ‘gold standard’ for medical research [10,11], their utility for assessing the efficacy of CBMPs may be limited by the following factors:

2. Exclusion criteria

They typically exclude many potential participants based on various factors including age, comorbid conditions, the use of other prescribed medications and prior or current use of cannabis. However, evidence is mounting that these exclusion criteria are likely to exclude a large percentage of patients who might receive CBMPs outside clinical trials. For example, data from T21, a UK-based registry, indicates that between 59% and 93% of patients receiving CBMPs would likely be excluded from trials due to comorbid conditions [9]. Similarly, 65% reported that they were currently using cannabis (primarily for medicinal reasons) and 71% reported using other prescribed medications, both factors which, in some trials, have been used to exclude potential participants.

These results closely mirror findings from a systematic review and meta-analysis of studies estimating, across a range of disorders, the percentage of ‘real world’ patients who would be excluded from specific trials [12]. This analysis identified 50 studies which reported exclusion rates from 305 trials and estimated a median exclusion rate of 77%. The combined effect of these exclusion criteria is that the majority of the target population are likely to be excluded from clinical trials, questioning the extent to which results from clinical trials have adequate external (real world) clinical validity.

3. Manner in which cannabis is used medicinally

In their simplest form RCTs typically compare differences in outcome between individuals receiving a specified dose of a single cannabis product and those receiving placebo. However, evidence from T21 indicates that individuals often use multiple products concurrently and there are also considerable variations in doses: 32% used a single product, 46% used two and 23% used three or more prescribed cannabis products. Similarly, many trials examining the efficacy of prescribed cannabis products have used relatively short periods of treatment, including as little as seven days or less, while relatively few have extended follow-up beyond three months. In contrast, observational studies can and have followed patients receiving CBMPs for one year or longer. This is an essential contribution to the study of CBMPs as many of the safety concerns (such as the development of dependence) can only be assessed over an extended time.

4. Feasibility of RCTs

The cost of RCTs of CBMPs may be prohibitively expensive given potential uncertainties about the security of intellectual property claims. Additionally, it may be increasingly difficult to recruit participants into a trial including a placebo arm in jurisdictions where there are other, legal routes of access to these treatments.

Together this evidence suggests that RCTs of CBMPs may become gradually less likely and practicable especially where there are also concerns about the external validity of this method when applied to CBMPs. In contrast RWE has, to date, made important contributions to our understanding of CBMPs. In particular, they provide accumulating evidence that, as used in real world medicine, these drugs are both safe and effective in reducing symptoms and improving quality of life across a range of primary conditions with chronic pain and anxiety being the most common conditions for which people seek CBMP treatment [13].

As more and larger real-world registries accumulate data, research is entering a second phase in which more refined hypotheses can be tested. Research of this nature includes findings that use of CBMPs is increasing in those aged 65 years or older [14] (who are traditionally excluded from RCTs [9]) and that, in some instances, the presence of comorbid conditions, which would traditionally be used as an exclusion criterion in clinical trials, may be associated with increased effectiveness of and benefits from CBMPs [15].

Despite these advances, there are still many unknowns while the nature of CBMPs also offer unique challenges to RWE approaches. These relate both to the nature of cannabis itself and to the manner in which it is typically used. Cannabis is a complex plant. Besides the well-known Cannabidiol (CBD) and Tetrahydrocannabinol (THC) components, more than 100 different cannabinoids have been identified [16]. These compounds can exist in many different ratios, and product combinations. Patients also use variable dosing over different health conditions and time frames. The combined consequence of these factors is that there is a need for large samples (an order of magnitude higher than those currently available), coupled with precision in the assessment of what products are being used and their precise constituents.

The second of these is helped in jurisdictions such as the UK, Australia and Canada where only CBMPs manufactured according to Good Medical Practice (GMP) or to specific governmental regulations are available; their constituents are well documented and variations between batches of these products should be minimized. Even with this possibility the number and variety of products available, their use in various combinations and highly variable dosing indicates the need for samples containing tens of thousands – or more – patients to be able to adequately assess the relative safety and effectiveness of a specific product or combination of products for a particular condition.

RWE has already and will continue to make important contributions to knowledge about the safety and effectiveness of CBMPs. However, it needs to be considered alongside findings from other approaches, including preclinical, translational research and RCTs. These research approaches will increase our understanding of CBMPs mechanism of action and of patient characteristics which may help predicting who will respond better to these products. Thus, RWE is not the sole or necessarily optimal approach across all research questions but, especially given widespread availability and use of these products outside clinical trials, has an important role to play. These approaches have already made substantial contributions and as sample sizes and level for precision regarding constituents of specific CBMPs coupled with characteristics of the people who use them they will likely become more valuable.

5. Expert opinion

RWE has provided important information about the characteristics of people seeking treatment with CBMPs and the safety and effectiveness of these drugs as they are used in legal markets. With increasing samples sizes and length of follow-up RWE can contribute to refining our understanding of optimal variants and dosing for specific conditions. While the number and precision of such studies will increase in common years they face multiple challenges, particularly related to numerous constituents in cannabis and the resultant wide range of different products that are available for use. Nonetheless, regulatory authorities such as the FDA, EMA and MHRA need to give greater weight to RWE when making decisions about the licensing of new medications. Given existing evidence from RWE supported by preclinical findings it is likely that CBMPs will play an increasing role in the management of chronic illness. Perhaps one of the most common benefits of these medications, which is applicable to their use across a range of chronic conditions is their potential to improve quality of life and well-being among people living with chronic illness.

Declaration of interest

DJ Nutt is Chair of the charity Drug Science. M Lynskey is Chief Research Officer of Drug Science. AK Schlag is Acting CEO and Head of Research of Drug Science. She is also scientific advisor to Evolve and Somai. Drug Science receives an unrestricted educational grant from a consortium of medical cannabis companies to further its mission, that is the pursuit of an unbiased and scientific assessment of drugs regardless of their regulatory class. All Drug Science committee members, including the Chair, are unpaid by Drug Science for their effort and commitment to this organization. None of the authors would benefit from the wider prescription of medical cannabis in any form. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Reviewer disclosures

A reviewer on this manuscript has disclosed that previously they served as an expert advisor for several licensed producers of medical cannabis in Canada and maintain a Pro bono association with Empower Pharma a company that will produce CBD tablets. Peer reviewers on this manuscript have no other relevant financial relationships or otherwise to disclose.

Acknowledgments

We thank all T21 patients who are continuing to provide their date to develop the scientific evidence based on medical cannabis. We would like to express gratitude to our partners whose generosity is enabling T21 patients to received their CBMPs at a reduced rate: Somai Pharmaceuticals, Blackpoint Biotech, Ethypharm and 4CLabs. We also thank our previous partners Alta Flora, Cellen Biotech LtD., JMCC Group, Khiron Life Sciences Corp., and Lyphe Group.

Additional information

Funding

This paper was not funded.