What is the best treatment for early rheumatoid arthritis?

This editorial refers to ‘Effectiveness of different combinations of DMARDs and glucocorticoid bridging in early rheumatoid arthritis: two-year results of CareRA’, by Veerle Stouten et al., on pages 2284–94. Different combinations of DMARDs in early RA (ERA) were studied by Stouten et al. using 2-year data from the CareRA trial, and they have reported their results in this issue of Rheumatology [1]. The CareRA trial was a pragmatic trial in ERA that compared three treatment arms for poor prognosis (high-risk) ERA [MTX 15 mg/week. SSZ 2 g/d and tapering high-dose prednisone starting with 60 mg/d (COBRA-Classic), MTX 15 mg/week and tapering prednisone starting at 30 mg/day (COBRA-Slim), and MTX 15 mg/week, LEF 10 mg/d and tapering prednisone starting at 30 mg/d (COBRA-Avant-Garde)]. In the low-risk ERA group, there was randomization to COBRA-Slim or MTX 15 mg/week with no steroids (Tight Step Up). In all groups, prednisone was to be discontinued by week 34 and DMARDs tapered at week 40 to MTX monotherapy [2]. In year 2, treatment was at the discretion of the rheumatologist, using a treat-to-target approach. Although at year 2, clinical disease activity index remission was better in COBRA Avant-Garde (the treatment that included LEF with MTX and steroids), there were no differences with respect to the primary outcome of DAS28-CRP remission at 2 years in the high-risk treatment groups. There were also no differences in the results at 2 years in the low-risk patients between the two treatment arms; except COBRA-Slim had twice as many patients in clinical disease activity index remission than the strategy of MTX monotherapy and stepping up over time. There may be explanations for negative results, including being underpowered (there were 322 patients analysed among the five treatment arms). However, many of the analyses had P-values that were nowhere near statistical significance. The primary outcome if changed could have shown differences: if clinical disease activity index remission had been used as the primary outcome at 2 years, there were statistical differences in treatment strategies in both the high- and low-risk patients. Also, in general, when using a treat-to-target strategy in year 2, this could minimize differences because drugs are adjusted to target remission. Using a tight control strategy yields better results (more remission) than treating without a target. This has been demonstrated repeatedly [3]. Other important questions to ask are: (i) Was the dose and route of administration of MTX optimal? and (ii) Should there have been tapering to discontinuation of one of the combination DMARDs within strategies that included either SSZ or LEF (with MTX remaining) and discontinuation of CSs before 1 year? Higher dosing of MTX, such as 25 mg/week, is more likely to demonstrate a superior response [4, 5], but the CareRA trial used only 15 mg/week. Subcutaneous MTX has a better effect than oral MTX in ERA, as was demonstrated in response and retention on treatment in a large multi-site Canadian incident cohort [4]. The EULAR ERA guidelines suggest a rapid titration of MTX to 20–30 mg/week and switching to parenteral MTX if there is an inadequate response or intolerability [6]. One of the research areas they highlighted was to determine whether combination conventional synthetic DMARDs (csDMARDs) had a greater benefit than MTX monotherapy [6]. The use of DMARD combinations in the CareRA trial did not include MTX with both HCQ and SSZ, also known as triple therapy. Triple therapy was demonstrated to be superior to monotherapy in ERA [7]. Consistently two-thirds of patients fail to reach a target in active ERA when treated with MTX monotherapy [7]. In the 2017 EULAR RA guidelines, there were 23 votes for MTX as part of the first treatment strategy and 22 for csDMARD monotherapy or combination therapy as initial treatment, irrespective of CS use [8]. This likely reflects the ambiguity of the literature on the optimal initial treatment for RA. The side effects and drop-outs of MTX monotherapy compared with triple therapy are not too different, and the cost incrementally is not much more compared with advancement to biologics or targeted synthetic DMARDs, so the side effect profile should not be a limiting step for using triple therapy [5–7]. There are less randomized controlled trial (RCT) data of combination csDMARDs vs MTX alone, with or without background steroids. In general, tapering of traditional DMARDs only occurs when patients are in sustained remission to reduce the chance of flaring [9]. So, tapering in the CareRA protocol may have increased the flare rate and minimized between-groups differences. With respect to CS use, EULAR RA guidelines have recommended bridging CSs in ERA using low-dose prednisone, but it appears that many patients with RA remain on CSs over time. For instance, RCTs in active established RA demonstrate that ∼4–6 of 10 patients are on background steroids [5]. Ideally, the use of CSs will increase the likelihood and speed of remission in ERA and will reduce radiographic progression and then be discontinued [8]. It seems that a significant minority of patients who start prednisone for RA are unable to taper and discontinue this treatment. Certainly, even tapering low doses of prednisone (1–4 mg/d) in patients with well-controlled RA is difficult and less effective than maintaining steroids [10]. Indeed, steroids are also analgesic and help to alleviate pain from swollen joints. There is a down-side to chronic prednisone use that includes osteoporosis, glucose and lipid abnormalities, weight gain, hypertension, and cataracts. Many people have myalgias and more pain when they taper their steroids, even if there is not obvious adrenal suppression. Chronic steroid use is associated with more cardiovascular events [11]. Steroid initiation in RA has to be balanced considering the benefits and risks. Although there are data gaps, the use of parenteral CSs such as intramuscular and/or intra-articular could result in less steroid dependence. This may be partially due to a lower cumulative dose and the lack of ability of patients to self-adjust their treatment if it is given parenterally. Several patients seem ‘addicted’ to their prednisone. In the high-risk group of the CareRA trial there were still 16–22% of patients requiring chronic oral CSs, and in the low-risk patients the level was 11–12%. Even in the low-risk group within the step-up allocation, more glucocorticoids were required, both by injection and orally (as measured by cumulative dose during year 2), but the proportion receiving chronic steroids was not different. The expected harms of chronic prednisone use could outweigh the expected side effects of combination csDMARDs when patients are not exposed to chronic prednisone as an initial adjunctive treatment strategy. Chronic steroids increase cardiovascular events [11]. There are certainly people with ERA who use triple therapy and require chronic steroids; however, the chronic steroid rate could be lower if prescribing was only ‘as required’ instead of part of a standard initial regimen. Following patients who added low-dose chronic CSs to their MTX treatment from a 2-year trial showed more remission if patients were treated with initial steroids, but more cerebrovascular events (but not myocardial events) at 10 years [12, 13]. It is difficult to ascertain which strategy is optimal for radiographic erosions, because only 4% had radiographic progression above the smallest detectable difference in the CareRA trial [1]. In general, in ERA the strongest predictor of radiographic progression is baseline erosions [6]. In summary, the Stouten article analysing the CareRA data [1] is an excellent example of what we can learn from a pragmatic trial, such as: when using the treat-to-target principles, many strategies have similar outcomes. However, questions still remain with respect to optimal dosing of MTX in ERA patients who have a high risk of disease progression and damage. The route of administration of MTX needs further study. The use of steroids, although recommended in RA guidelines [6, 8], may need further study whereby the route of administration and ability to fully taper can be understood more thoroughly. There remain questions about the short-term gains of combination csDMARDs in ERA and how and when to effectively taper DMARDs. Funding: No specific funding was received from any funding bodies in the public, commercial or not-for-profit sectors to carry out the work described in this manuscript. Disclosure statement: The author has declared no conflicts of interest.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.