Abstract
Objective To provide the initial installment of a living guideline that will provide up-to-date guidance on the pharmacological management of patients with rheumatoid arthritis (RA) in Canada.
Methods The Canadian Rheumatology Association (CRA) formed a multidisciplinary panel composed of rheumatologists, researchers, methodologists, and patients. In this first installment of our living guideline, the panel developed a recommendation for the tapering of biologic and targeted synthetic disease-modifying antirheumatic drug (b/ts DMARD) therapy in patients in sustained remission using the GRADE (Grading of Recommendations Assessment, Development, and Evaluation) approach, including a health equity framework developed for the Canadian RA population. The recommendation was adapted from a living guideline of the Australia & New Zealand Musculoskeletal Clinical Trials Network.
Results In people with RA who are in sustained low disease activity or remission for at least 6 months, we suggest offering stepwise reduction in the dose of b/tsDMARD without discontinuation, in the context of a shared decision, provided patients are able to rapidly access rheumatology care and reestablish their medications if needed. In patients where rapid access to care or reestablishing access to medications is challenging, we conditionally recommend against tapering. A patient decision aid was developed to complement the recommendation.
Conclusion This living guideline will provide contemporary RA management recommendations for Canadian practice. New recommendations will be added over time and updated, with the latest recommendation, evidence summaries, and Evidence to Decision summaries available through the CRA website (www.rheum.ca).
Rheumatoid arthritis (RA) is the most prevalent inflammatory arthritis, affecting an estimated 1.2% of Canadians aged 16 years and older.1 People with RA experience pain, fatigue, functional limitation, work loss and reduced quality of life. The economic burden of RA in Canada was estimated at CAD $5.7 billion (US $4.4 billion) annually in 2011, with a rising burden over time due to an aging population.2 The Canadian Rheumatology Association (CRA) developed initial treatment recommendations for RA in 2011–2012. Since then, the treatment landscape has changed considerably. Several new disease-modifying antirheumatic drugs (DMARDs) have been introduced, including targeted synthetic DMARDs (tsDMARDs) with novel mechanisms of action and biosimilars. New evidence continues to emerge to inform decisions between these treatments.
In the setting of continually emerging evidence and new treatments, maintaining up-to-date guidelines is challenging. In a typical guideline development cycle, the entire guideline is updated periodically, typically every 2 or 3 years or longer. In contrast, in a living guideline model, individual recommendations are developed or updated when necessary, ensuring recommendations stay up to date.3 Underpinning living guidelines are living systematic reviews.4 While living systematic reviews and guidelines may require more upfront effort to establish, they also provide efficiencies, as systematic review and guideline teams are continually active and engaged.3 The start-up efforts with each guideline cycle are removed, and the overall workload is diffused over time. Collaboration on systematic reviews across international groups can provide further efficiencies by avoiding duplication of effort.
The objective of these recommendations is to provide guidance for the pharmacological management of RA with DMARDs. The need for this guideline was approved by the Guidelines Committee of the CRA.
METHODS
This guideline was developed using the GRADE (Grading of Recommendations Assessment, Development, and Evaluation) approach, which provides a systematic process for appraising the certainty of evidence and grading the direction and strength of recommendations.5 Ethics approval was not required.
Target audience. The target audience is rheumatologists or other primary prescribers of RA medications and their patients with RA, in community and academic practice settings. Recommendations may also be of interest to other provincial and federal RA stakeholders and decision makers.
Target population. These recommendations apply to adult patients (age > 18 yrs) with RA. This includes patients whose RA began in childhood or adolescence as juvenile idiopathic arthritis and has persisted into adulthood.
Perspective. This guideline takes the perspective of treatment decisions made between the rheumatologist and the person living with RA.
Organization and panel composition. The CRA assembled a guideline panel that included rheumatologists, researchers, methodologists, and 2 people living with RA (Supplementary Material, available with the online version of this article). Methodological support was provided by Cochrane Musculoskeletal for evidence synthesis. All panel meetings were held virtually by video calls.
Guideline funding and management of conflicts of interest. The guideline development was supported by in-kind funding from the CRA, a nonprofit association that represents Canadian rheumatologists. The CRA also provides ongoing funding to Cochrane Musculoskeletal. Declarations of potential conflicts of interest (COI) were collected from all panelists using the International Committee of Medical Journal Editors (ICMJE) form. The chair (GSH) and co-chair (JPP), and all members of the voting panel were required to be free of any direct financial COI within the past 36 months, which meant no direct payments including research funding support from any manufacturers of RA therapeutics. Expert panel members with COI were allowed to participate in the discussion but did not vote on the direction or strength of the recommendation. All disclosure forms were reviewed and potential COI was adjudicated by an independent member of CRA who was not otherwise involved in the present guideline, and discussed with the chair and co-chair in the setting of ambiguity. The COIs are maintained over time and are available online (https://rheum.ca/resources/publications).
Formulating clinical questions. The initial clinical question for this guideline related to tapering of therapy was chosen by the panel for its importance to decision making and the availability of a recent (and living) Australian systematic review and guideline.6
Development of recommendation. The recommendation was developed using the GRADE-ADOLOPMENT approach, which provides a framework for efficient adoption or adaptation of existing guidelines or de novo development of recommendations.7 With GRADE-ADOLOPMENT, existing GRADE evidence profiles are used where possible. Evidence to Decision (EtD) profiles, which outline the evidence and rationale for the recommendation, are either generated or modified as necessary to contextualize the recommendation to a different healthcare context.
Prior to the panel meeting, a core team reviewed the published GRADE evidence profile (GSH and JPP) and EtD profile (GSH, NB, ChB, JEP), which were developed and are maintained in a living fashion by the Australia & New Zealand Musculoskeletal (ANZMUSC) Clinical Trials Network.6 No changes were made to the evidence profile, but the EtD profile was modified to contextualize it to a Canadian setting. Health equity was explicitly discussed and considered for each step in the EtD framework, following a recently published process that we developed for CRA guidelines,8 informed through stakeholder interviews.9 Within this framework, equity considerations relevant to RA guidelines for 6 populations at risk for inequities were generated and mapped to each step of the EtD framework. These populations included rural and remote residents, Indigenous peoples, elderly persons with frailty, minority populations of first-generation immigrants and refugees, persons with low socioeconomic status or who are vulnerably housed, and sex and gender populations.
The evidence profile and modified EtD framework were reviewed by panelists prior to the meeting and then discussed during the online panel meeting, leading to a consensus judgment for each step of the EtD framework/process. The direction (ie, to recommend or not) and strength of the final recommendation was discussed with all panelists and determined through a formal vote of the qualified voting panelists. A simple majority (> 50%) was required to determine the direction of the recommendation, and development of a strong recommendation required 80% agreement.8
How to read this guideline. In the GRADE approach, recommendations are categorized as strong or conditional.5 A strong recommendation means that all or almost all people with RA would choose that intervention. A conditional recommendation means that the majority of people with RA in this situation would want the suggested course of action, but many would not (Table 1).10
How to use this guideline. This recommendation is intended to help rheumatologists and patients make decisions regarding RA treatment and is not meant to replace clinical judgment. This recommendation is subject to change over time in a living fashion as new evidence emerges and should always consult the CRA website (https://rheum.ca/resources/publications) for the latest version.
Public commenting. Public commenting will be available through the CRA website. The public comments will be reviewed on an ongoing basis and may be considered in future updates.
Living guidelines. These guidelines will be maintained over time. New recommendations will be added, and existing recommendations may be modified in the setting of new evidence. Readers should consult the online version available (https://rheum.ca/resources/publications) for the latest version. This article will not be modified over time, but additional journal articles may be published to supplement the online living version and aid in knowledge translation.
RESULTS
Should biologic or targeted synthetic DMARDs be tapered in patients who are in sustained remission or low disease activity?
Recommendation. In people with RA who have been in sustained low disease activity or remission for at least 6 months, we suggest offering stepwise reduction in the dose of biologic (b)/tsDMARD without discontinuation, in the context of a shared decision, provided patients are able to rapidly access rheumatology care and reestablish their medications in case of a flare. (Conditional recommendation; moderate certainty of evidence.)
In patients where rapid access to care or reestablishing access to medications is challenging, we conditionally recommend against tapering. (Conditional recommendation; moderate certainty of evidence.)
Rationale and key remarks.
The panel judged that for reduction in therapy, given the moderate certainty evidence of the little negative impact on disease control and that most patients who flare can regain disease control promptly once medications are reestablished, a trial of treatment reduction (without complete discontinuation) would be appropriate for many patients to reduce medication burden and possible side effects. Given the increase in flares seen over relatively short follow-up in the trials of discontinuation of treatment, the panel made a conditional recommendation against complete discontinuation of advanced therapies.
Rapid access to care and ability to reestablish medications was highlighted as a particularly important consideration when deciding whether to taper. The panel felt that in situations where access to care is difficult, tapering would typically not be recommended. This, however, is also a conditional recommendation, meaning tapering may still be appropriate for some patients in the context of a shared decision.
The majority of the evidence relates to anti–tumor necrosis factor (TNF) therapy, although results for other mechanisms of action that have been studied appear similar, including rituximab,11 abatacept,12 tocilizumab,13 and baricitinib (trial studied the reduction of 4 mg/day dose to the approved dose in Canada of 2 mg/day).14
This recommendation applies to both biologic originator and biosimilar DMARDs.
While shared decision making (SDM) is implicit in a conditional recommendation, the panel felt it was important to highlight in the recommendation itself, given the wide variability in patient preferences around treatment tapering.
Implementation and practical information.
Implementation of the recommendation would be supported with models of care that allow rapid access to care from a rheumatology care team, including in populations at risk for inequity, and reimbursement policies that facilitate immediate reescalation of doses in case of a flare.
A flare management plan should be discussed with patients prior to tapering. While tapering, patients should be reassessed typically at 3 months. In the case of flare, a typical approach would be to increase the dose back to the previous effective dose.
Six months of adequate disease control was felt to be the minimum duration. The panel felt that sustained disease control (ideally remission according to a composite measure with no swollen joints) for 12 months would be ideal prior to tapering.
A tool to support SDM for this recommendation is available through the CRA website (https://rheum.ca/resources/publications).
Dose reduction may include extending the interval between doses or reducing the amount with each dose. A typical initial reduction would be 25% of the original effective dose, for example, by increasing the interval for adalimumab (ADA) from every 2 weeks to every 3 weeks. Further reductions (eg, extending the interval to every 4 weeks) may also be possible. Complete discontinuation of the b/tsDMARD is not routinely recommended, although it may be possible in some patients; in these patients, a conventional synthetic DMARD (csDMARD) should be continued.
In people taking csDMARD(s) in combination with b/tsDMARDs, the doses of csDMARD(s) should be kept stable during dose reduction of the b/tsDMARD.
Prior to reducing b/tsDMARDs, glucocorticoids should be discontinued, if possible.
Monitoring and evaluation. It will be important to monitor this recommendation in real-world practice, including the frequency of treatment tapering being offered, discussed, and initiated, and the resulting clinical outcomes. We support monitoring of this through existing Canadian RA registries and studies using administrative data, including in populations at risk for inequities.
Evidence to Decision profile. The following EtD profile was used in the development of this recommendation and is also available at (https://rheum.ca/resources/publications). The online version will be updated over time.
• Benefits and harms and certainty of evidence. The panel reviewed the GRADE Evidence Profiles of the ANZMUSC source guideline6 that summarized the evidence on benefits and harms and certainty of evidence for reduction of treatment (Table 2) and complete discontinuation of treatment (Table 3) and accepted them without modifications. There was moderate certainty evidence that reducing the dose of b/tsDMARD therapy was associated with little to no difference in disease control over 12 months, both in terms of the proportion of patients in remission (54 out of 100 with continuing treatment vs 49 out of 100 with reduction) and the proportion of patients with a flare (22 out of 100 with continuing treatment vs 27 out of 100 with reduction). There may be a small negative effect on function and the proportion of patients with a minimal amount of radiographic progression (Table 2). There was little to no difference in adverse events, although event rates were low.
There was moderate certainty evidence that discontinuing b/tsDMARD therapy (Table 3) was associated with a decrease in the proportion of patients with persistent remission (61 out of 100 with continuing treatment vs 34 out of 100 with discontinuation) and an increase in the proportion of patients with a flare (26 out of 100 with continuing treatment vs 49 out of 100 with discontinuation). Discontinuing therapy may also slightly increase the proportion of people with minimal radiographic progression, may lead to a slight deterioration in function, and may slightly worsen quality of life (Table 3). There was little to no difference in adverse events, although event rates were low.
∙ Equity considerations. There was no available evidence to conclude that the treatment effects or certainty of evidence would vary for populations facing inequities in rheumatology care and outcomes.
• Preferences and values. Recent reviews on RA patient preferences for down-titration15,16 identified several qualitative and mixed-methods studies.17-21 An additional Canadian study published since these reviews assessed preferences of patients and rheumatologists for tapering both biologic and nonbiologic therapy.22 Common themes identified in the qualitative work included a desire among some patients to reduce medication burden but also fear of flaring, and concern about the ability to successfully recapture disease control. Having a flare management plan, including the ability to rapidly access care and reescalate doses in the occurrence of a flare is particularly important. There is wide variability in preferences between patients, which is influenced by patients’ lived experiences, side effects, previous tapering experiences, disease trajectory (eg, severity of disease and number of previously failed therapies), remission duration, and current life roles. This supports the importance of SDM.
We did not identify any quantitative patient preference studies of treatment tapering in a recent systematic review,23 or an updated Pubmed search (“rheumatoid arthritis patient preference*”) to August 2021.
∙ Equity considerations. In other patient preference studies in patients with RA (not focused on treatment tapering), preferences are often associated with membership in populations at risk for inequities.23 These preferences will reflect both individual and population beliefs and values, informed by population membership, which should be explored in an SDM strategy for tapering.
• Resource use. A Cochrane review24 in 2019 identified 2 trials in Europe that studied the cost effectiveness and costs of guided tapering of anti-TNF agents etanercept and ADA with anti-TNF continuation.25,26 Both studies found little to no difference in quality-adjusted life-years, but resulted in considerable cost savings. Modeling studies have projected similar results in the long term.27 In Canada, it is assumed that dose reduction of b/tsDMARDs will reduce costs for payers (governments and insurers) and may reduce out-of-pocket costs for patients. Out-of-pocket costs will vary by Province, whether patients have supplementary medical insurance, and depending on their province of residence, age, income, and other concomitant prescription drugs. Approximately 8% of Canadians who received a prescription did not take the drug as prescribed because of cost.28
∙ Equity considerations. Populations facing inequities in rheumatology care and outcomes will have intersecting limitations in available resources to access bDMARDs; therefore, tapering may have specific advantages. Specific populations, however, may not entertain tapering due to the consequences of this choice (eg, insurers may limit the ability to reescalate therapy in case of a flare).
• Acceptability and feasibility. The acceptability of reducing bDMARDs or tsDMARDs is expected to vary widely between patients. The feasibility of reducing bDMARDs or tsDMARDs may change based on insurance coverage. This is different between provinces and represents a barrier to implementation.
∙ Equity considerations. Acceptability of tapering likely differs by population membership. Differences in funding reimbursement may affect the ability to reescalate treatment for particular populations (eg, federal Non-Insured Health Benefits formulary, seniors insurance plans), and when there appears to be nonadherence to the recommended dose of an expensive medication, it theoretically may not be reimbursed when renewed.
DISCUSSION
This guideline represents the first installment of CRA living treatment recommendations for RA. This marks the transition to a living mode of guideline development, where individual recommendations will be updated and maintained over time.
This is also the first implementation of our recently published equity framework,8 which was informed through stakeholder interviews among Canadian patients and rheumatology providers.9 In the context of this recommendation, equity considerations led to a conditional subgroup recommendation against treatment tapering in situations where rapid access to care is challenging. Importantly, and in line with our equity framework, this subgroup recommendation was linked to the underlying factor (in this case, barriers to accessing care), rather than the population itself. Equitable implementation of this recommendation can be supported through models of care that allow for access to appropriate care for all patients. The populations identified in our equity framework where access to care is systematically different, includes, but may not be limited to, people living in rural/remote locations, Indigenous peoples, refugee and first-generation immigrant populations, and persons of low socioeconomic status and vulnerably housed.8,9
SDM features prominently in this recommendation. For some patients, the risk of flaring will outweigh the potential benefits. This will depend both on an individual’s risk of flare as well as the impact of a flare on their life. While validated predictive tools for an individual’s flare risk are not in widespread use, this is an active area of research. Patient preferences for tapering may change over time, so the decision should be revisited. To support SDM for this recommendation, a decision aid has been developed. In the living guideline, we will continue to develop tools to support SDM for preference-sensitive recommendations.
The development of this recommendation was possible through the publication of Australian living recommendations, along with full EtD tables and an EtD framework.6 Future recommendations will be supported by ongoing Cochrane living systematic reviews of DMARD therapy,29 also a collaborative effort. In the living guideline model, we will also continue to make use of other guidelines, through the GRADE-ADOLOPMENT approach. Ideally, international guideline groups in rheumatology would collaborate on living systematic reviews for common clinical questions, saving considerable duplication of effort. Different groups can then contextualize the recommendation to their setting. Published EtD tables aid this process, by presenting the evidence and judgments behind the recommendation according to the structured GRADE process. In our tapering recommendation, our overall recommendation was the same as the source guideline (conditional recommendation for tapering), with some differences in the wording to reflect the importance of SDM and equity concerns regarding access to care. Currently, European Alliance of Associations for Rheumatology guidelines state that tapering can be considered, especially if the treatment is combined with a csDMARD.30 American College of Rheumatology guideline provides a conditional recommendation against tapering, although it also recommends gradual reduction vs abrupt discontinuation in patients where tapering is being tried.31
In summary, we present an initial recommendation on tapering of b/tsDMARDs in patients with RA. Readers should consult the online version for the latest version of the recommendation.
ACKNOWLEDGMENT
The authors thank Sue Ranta and the CRA for administrative support during the development and dissemination of the guideline, and the Australia & New Zealand Musculoskeletal Clinical Trials Network (ANZMUSC) for their work on developing and maintaining the source recommendation that we used for adaptation.
Footnotes
Funding for this guideline was provided by the Canadian Rheumatology Association (CRA).
JPP received funding from the CRA to Cochrane Musculoskeletal to provide methodological support for guideline development. C. Barnabe received honoraria for advisory boards (Gilead, Sanofi, Celltrion) and speaker fees (Sanofi) in the past 3 years. LP and DPR are volunteer vice presidents of the Canadian Arthritis Patient Alliance, which receives the majority of its funding from independent grants from pharmaceutical companies. LP has received payment and travel support in the last 3 years from Lilly Canada (sharing patient perspectives at an event). DPR is an employee of Five02 Labs Inc., which has received payment in the last 3 years from Novo Nordisk Canada (speaking on living with arthritis and arthritis advocacy) and Lilly Canada (sharing patient story, participating in an advisory board). PA receives honoraria for advisory boards (Janssen, Eli Lilly, Pfizer, Sandoz, Celltrion, AbbVie, Roche). C. Bombardier receives consulting fees for Novartis and Samsung; and honoraria for AbbVie, Janssen, Merck, Pfizer, GSK, Mylan. VB receives consulting fees from Amgen, BMS, Gilead, Genzyme, Regeneron, UCB; and research funding from Amgen, BMS, Genzyme, Pfizer, Sanofi Aventis, UCB. SJ receives honoraria for advisory boards (AbbVie, Amgen, BMS, Boehringer, Celgene, Celltrion, Eli Lilly, Fresenius Kabi, Gilead, Janssen, Merck, Novartis, Pfizer, Roche, Sandoz, Sanofi, Teva, UCB). MK receives consulting for Amgen, AbbVie, Celgene, Merck, Novartis, Pfizer, Gilead. JCT receives consulting for Celgene, AbbVie, Merck, Pfizer, Sandoz. JEP receives consulting for AbbVie, Amgen, BI, BMS, Celltrion, Fresenius Kabi, Galapagos, Gilead, Janssen, Eli Lilly, Medexus, Merck, Mitsubishi Tanabe Pharma, Novartis, Pfizer, Roche, Sandoz, Samsung, Sanofi, Sobi, Teva, UCB, VIatris. The remaining authors declare no conflicts of interest relevant to this article.
- Accepted for publication June 24, 2022.
- Copyright © 2022 by the Journal of Rheumatology