Therapeutic use of cannabis, cannabis-derived products and synthetic cannabinoids for rheumatoid arthritis: Protocol for a systematic review

PROTOCOL AND REGISTRATION

This review is part of a larger project of multiple parallel systematic reviews for any condition for which cannabis, cannabis-derived products or synthetic cannabinoids have been postulated as a therapeutic alternative. The protocol of the larger project is registered on PROSPERO and was allocated the registration number CRD42018097382 (40). The protocol was adapted to the specificities of the question assessed in this review in line with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols (PRISMA-P) (41) and submitted to PROSPERO (awaiting ID allocation). The PRISMA-P checklist is given as supplement (see Supplement 1: PRISMA-P checklist archive).

AMENDMENTS

In the event that an amendment to this protocol is necessary, we will record the date of each amendment along with a description of the change and the rationale.

ELIGIBILITY CRITERIA

ELECTRONIC SEARCHES

We will conduct sensitive electronic searches of the following databases (without any date, publication status or language restriction):

• Cochrane Central Register of Controlled Trials (Wiley).

• MEDLINE (PubMed).

• EMBASE (Elsevier).

The MEDLINE strategy was adapted to the syntax and subject headings of the other databases (see Supplement 2: Search strategy for electronic databases). The search will cover the period until the day before submission to a journal and will be will be updated toward the end of the review.

SEARCHING OTHER SOURCES

An expanded search will be performed to identify articles potentially missed through the database searches and in order to identify ‘grey literature’ and unpublished studies. This includes the following:

1. We will manually search reference lists of all included studies.

2. We will screen the reference lists of selected guidelines and narrative reviews on the topic.

3. We will conduct searches in Epistemonikos database (www.epistemonikos.org) and L·OVE (https://iloveevidence.com/) to identify other systematic reviews on the topic, scan their reference lists and evaluate in full text all the articles they include.

4. We will review Medwave (46), a peer-reviewed international general medical journal, in order to find FRISBEEs (FRIendly Summary of the Body of Evidence using Epistemonikos) that address our clinical question. FRISBEEs are rapid evidence summaries created from an evidence matrix in Epistemonikos (47). The evidence matrix is meta-analysed using the included primary studies and a summary of findings table is created using the GRADE method. Finally, the FRISBEE is drafted, peer-reviewed and published in Medwave (47).

5. We will create a matrix of evidence in Epistemonikos with all the existing systematic reviews identified, in order to identify other reviews that share primary studies with the index reviews.

6. We will conduct a cross-citation search in Google Scholar and Microsoft Academic, using each included study as the index reference.

7. We will hand search journals specialized in cannabis medicine: Cannabis, Cannabis and Cannabinoid Research.

8. We will review international conferences related to cannabis medicine: Scientific Meeting of the Research Society on Marijuana, International Cannabinoid Research Society Symposium on the Cannabinoids and Cann.

9. We will screen international conferences related to rheumatoid arthritis: ACR (https://acrabstracts.org/) and EULAR (https://www.eular.org/).

10. We will review websites from pharmaceutical companies producing cannabis-based products and other selected sites such as GW pharmaceuticals (https://www.gwpharm.com/), International Association for Cannabinoid Medicines (https://www.cannabis-med.org/?lng=en).

11. We will search the websites of regulatory agencies: USA Food and Drug Administration-MedWatch (https://www.fda.gov/safety/medwatch-fda-safety-information-and-adverse-event-reporting-program), European Medicines Evaluation Agency (https://www.ema.europa.eu/en), Australian Adverse Drug Reactions Bulletin (https://www.tga.gov.au/adr/aadrb.htm), and United Kingdom Medicines and Healthcare products Regulatory Agency pharmacovigilance and drug safety updates (https://www.gov.uk/government/organisations/medicines-and-healthcare-products-regulatory-agency).

12. The WHO International Clinical Trials Registry Platform Search Portal (http://apps.who.int/trialsearch/), the European Union Clinical Trials Register (https://www.clinicaltrialsregister.eu/), the United States National Institutes of Health Ongoing Trials Register (https://clinicaltrials.gov/), the ISRCTN registry (https://www.isrctn.com/), Current Controlled Trials (http://www.controlled-trials.com) and Natural Medicines (https://naturalmedicines.therapeuticresearch.com) will be searched to identify additional published or unpublished data and ongoing trials.

13. We will email the contact authors of all of the included trials to ask for additional publications or data on their studies, and for other studies in the topic.

STUDY SELECTION

The results of the literature search will be uploaded to the software CollaboratronTM (48). References will be de-duplicated by an algorithm comparing unique identifiers (database ID, DOI, trial registry ID), and citation details (i.e. author names, journal, year of publication, volume, number, pages, article title and article abstract). We will not anonymize the studies in any way before assessment.

At least two authors will independently screen each title and abstract identified by the search to decide upon their potential inclusion. We will obtain the full reports for all titles that appear to meet the inclusion criteria or require further analysis to decide on their inclusion. Then, full text articles will be screened against the inclusion criteria by at least two independent reviewers.

We will resolve disagreements through discussion or through a third reviewer if the discrepancy could not be solved. The reasons for excluding trials in any stage of the search will be recorded. We will outline the study selection process in a PRISMA flow diagram (see Figure 1).

• Download figure
• Open in new tab

Figure 1.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram— A schematic of the processes of the systemic review. FRISBEEs: FRIendly Summary of the Body of Evidence using Epistemonikos.

DATA EXTRACTION AND MANAGEMENT

Using standardized forms, two reviewers will extract data independently from each included trial. The following data will be extracted, where possible:

• Study design.

• Study eligibility criteria.

• Length of follow-up.

• Sample size.

• Rheumatoid arthritis diagnosis (clinical diagnosis or diagnosis using classification criteria such as ACR (42) or ACR/EULAR (43)).

• Characteristics of study population, including age, female/male ratio, geographic region, socioeconomic status (e.g., education and income), smoking habit, nutritional status, comorbidities (e.g. cardiovascular disease, malignancies, infections, gastrointestinal disease, osteoporosis, depression), disease duration (early versus established or years of evolution), severity and activity, structural damage (erosive rheumatoid arthritis) and previous treatments

• Details of the administered intervention(s) including dose, therapeutic scheme, route of administration and treatment duration.

• Details of any co-interventions, including current synthetic DMARDS, biologic DMARDS and steroids use.

• Method used to seek adverse reactions.

• Outcomes and timing of outcome assessment.

• Losses to follow up, exclusions and their reasons.

• Source of study funding.

• Conflicts of interest disclosed by the investigators.

• Risk of bias assessment for each individual study.

The raw data will be extracted for outcomes of interest. Whenever possible, we will use results from intention-to-treat analyses.

Differences in data extraction will be resolved by referring back to the original articles and establishing a consensus. One arbiter will adjudicate unresolved disagreements.

We will contact study authors to resolve any uncertainties.

ASSESSMENT OF RISK OF BIAS IN INCLUDED STUDIES

At least two reviewers independently will assess risk of bias using a ‘risk of bias’ tool (RoB 2.0: a revised tool to assess risk of bias in randomized trials) (49). We will consider the effect of assignment to the intervention for this review. Two reviewers will independently evaluate five domains of bias for each outcome result of all reported outcomes and time points (bias due to the randomization process, deviations from intended interventions (effects of assignment to interventions at baseline), missing outcome data, and measurement of the outcome and selection of reported results). Answers to signaling questions and supporting information collectively will lead to a domain-level judgment in the form of ‘Low risk of bias’, ‘Some concerns’, or ‘High risk of bias’. These domain-level judgments will inform an overall ‘risk of bias’ judgment for each result. In case reported details in a study were insufficient, the original study investigators will be contacted for more information. Disagreements will be resolved first by discussion and then by consulting a third author for arbitration.

We will compute graphic representations of potential bias within and across studies using RevMan 5.3.5 (50).

MEASURES OF TREATMENT EFFECTS

For dichotomous outcomes, we will express the estimate of treatment effect of an intervention as risk ratios (RR) or odds ratios (OR) together with 95% confidence intervals (CIs). If the outcome is a rare event (approximately less than 10%), we will use Peto odds ratio (Peto OR) instead of RR.

For continuous outcomes, we will use mean difference and standard deviation (SD) to summarize the data along with 95% CIs. Whenever continuous outcomes were measured using different scales, the treatment effect will be expressed as a standardized mean difference (SMD) with 95% CI. When possible, we will multiply the SMD by a standard deviation that is representative from the pooled studies, for example, the SD from a well-known scale used by several of the studies included in the analysis on which the result is based. In cases where the minimally important difference (MID) is known, we will also present continuous outcomes as MID units or inform the results as the difference in the proportion of patients achieving a minimal important effect between intervention and control (45). Then, these results will be displayed on the ‘Summary of Findings Table’ as mean difference (45).

DEALING WITH MISSING DATA

When possible, we will contact the original authors of the study to obtain any missing data. If important numerical missing data could not be obtained, an imputation method will be used (51). When available, we will extract data from graphs by using data extraction software.

ASSESSMENT OF HETEROGENEITY

We will assess the variations in treatment effect from the different trials by means of the I2 statistic. A rough guide to the interpretation of the I2 statistic given in the Cochrane Handbook is: 0–40% might not be important, 30–60% may represent moderate heterogeneity, 50–90% may represent substantial heterogeneity and 75–100% considerable heterogeneity. However, its importance will depend on magnitude and direction of effects and strength of evidence for heterogeneity (52).

ASSESSMENT OF META-BIASES

We will investigate the presence of publication bias visually with the use of funnel plots. We will base evidence of asymmetry on p<0.10, and presented intercepts with 90% CIs. Other reporting biases, including selective non-reporting of an outcome, will be evaluated through discrepancies between the registered protocol and the final publication. If we could not find the record of a study in clinical trial registries, we will contact the authors for more information.

STRATEGY FOR DATA SYNTHESIS

We will only conduct meta-analysis if the included studies are sufficiently homogeneous in terms of design, population, interventions and comparators reporting the same outcome measures.

The results for clinically homogeneous studies will be meta-analyzed using RevMan 5 (50), with the inverse variance method with random effects model. Separate meta-analyses will be presented for specific populations or interventions if statistically significant heterogeneity is explained by some of these, or if a convincing subgroup effect was found.

For any outcomes where insufficient data is found for a meta-analysis, a narrative synthesis will be presented.

SUBGROUP ANALYSIS AND INVESTIGATION OF HETEROGENEITY

We will conduct our main analyses combining all the interventions, but when possible we will also conduct the following subgroup analyses:

• Patients’ age (< 65 years versus ≥ 65 years).

• Gender (male versus female).

• Duration of rheumatoid arthritis (≤ 1 year versus > 1 year).

• Rheumatoid arthritis disease activity (e.g. remission/low disease activity versus moderate/high disease activity).

• Intervention category (herbal cannabis, cannabis-derived products and synthetic cannabinoids).

• Individual intervention (i.e. specific drugs).

• Route of administration (e.g. inhaled).

If we identify significant differences between subgroups (test for interaction <0.05) we will report the results of individual subgroups separately.

If an unexpected clinical or methodological heterogeneity is found due to obvious reasons, we will state hypotheses regarding these for future versions of this review. We will not anticipate undertaking additional analyses in this version.

SENSITIVITY ANALYSIS

We will use sensitivity analyses to assess the impact on the treatment effects of inclusion of trials with ‘high risk of bias’ versus ‘low overall risk of bias’ for each comparison and outcome. If the primary analysis effect estimates and the sensitivity analysis effect estimates significantly differ, we will either present the low risk of bias — adjusted sensitivity analysis estimates — or report the primary analysis estimates downgrading the certainty of the evidence due to risk of bias.

If we find substantial clinical heterogeneity on how the condition (rheumatoid arthritis) was defined we will explore this using sensitivity analysis.

ASSESSMENT OF CERTAINTY OF EVIDENCE

The certainty of the evidence for all outcomes will be judged using the Grading of Recommendations Assessment, Development and Evaluation working group methodology (GRADE Working Group) (53).The certainty of the evidence will be assessed across the domains of risk of bias, consistency, directness, precision and reporting bias. Certainty will be adjudicated as high, moderate, low or very low.

For the main comparisons and outcomes, we will prepare Summary of Findings (SoF) tables and also interactive Summary of Findings (iSoF) tables (http://isof.epistemonikos.org/). A SoF table with all the comparisons and outcomes will be presented as an appendix.