Introduction and Methods

This systematic review updates a 2012 report that evaluated the benefits and harms of drug therapies for adults with rheumatoid arthritis (RA).¹ This updated review, however, has a targeted scope focusing solely on patients with early RA. Early RA has no formal consensus definition. Based on guidance from a recent task force of experts,² we define early RA as no more than 1 year of diagnosed disease. Our findings should be considered applicable only to patients with early RA.

The U.S. Food and Drug Administration (FDA) has approved several drug therapy groups for treating patients with RA. Corticosteroids and conventional synthetic disease-modifying antirheumatic drugs (csDMARDs) have been prescribed the longest. Targeted synthetic disease-modifying antirheumatic drugs (tsDMARDs) were approved more recently. Additionally, many trials or observational studies in this review evaluated mainly FDA-approved biologic drugs (both tumor necrosis factor [TNF] and non-TNF drugs). The FDA has approved numerous biosimilars.

We evaluated the benefits and harms of multiple drug monotherapies, combination therapies, and different treatment strategies to determine whether therapeutic approaches differ in their ability to affect important outcomes for patients with early RA. The benefits and harms included (1) reduced disease activity, decreased progression of joint damage, or remission; (2) improved functional capacity or quality of life; (3) harms such as tolerability, serious adverse events, and adverse effects; and (4) benefits and harms among patient subgroups (based on disease activity, prior RA therapy, demographics, or presence of other diseases with or without treatment). Additional details about this systematic review are described in Table A. Two Contextual Questions were also examined: (1) Does treatment of early RA improve disease trajectory and disease outcomes compared with the trajectory or outcomes of treatment of established RA? And (2) What barriers prevent individuals with early RA from obtaining access to indicated drug therapies?

We synthesized the literature qualitatively within and between corticosteroids and classes of disease-modifying antirheumatic drugs (DMARDs), including csDMARDs, tsDMARDs, TNF and non-TNF biologics, and biosimilars. Additionally, combination treatment strategies were examined. We conducted network meta-analysis for five outcomes: American College of Rheumatology 50 percent improvement (ACR50), remission based on Disease Activity Score (DAS), radiographic joint damage, all discontinuations, and discontinuations due to adverse events.

Table A

Summary of characteristics of this systematic review on treatment of patients with early rheumatoid arthritis.

Results and Key Findings

We included 49 studies (reported in 124 published articles) that provided data on at least one of the review’s Key Questions. Of these studies, 41 were randomized controlled trials (RCTs), 4 were observational studies with control groups, and 4 were single-arm observational studies that we included only for evaluating harms of treatment.

We rated a majority of these studies as low or medium risk of bias. We rated 16 studies as high risk of bias for at least some of the eligible outcomes they reported. Studies (n=2) or study outcomes rated high risk of bias were excluded from analyses and used only in sensitivity analyses for the network meta-analysis. We graded strength of evidence (SOE) for numerous outcomes in studies for these drug classes and therapeutic approaches (except that the single-arm observational studies were not included in the SOE assessments).

The range of mean (or median) disease durations across all 49 included studies was 2 weeks to 12 months. Prior treatment use varied widely across drug therapy categories. Among all 49 included studies, five studies did not report any details about prior treatment use,^3–7 leaving 44 studies that did. Of these, 36 enrolled methotrexate (MTX)-naïve patient samples, and the remaining eight studies enrolled patients with at least some prior csDMARD use (including MTX).

In four of these eight studies, prior use of any csDMARDs ranged from 13 to 48 percent. The other four enrolled samples that were entirely csDMARD resistant.^8–11 Among the 15 studies analyzed in our primary or sensitivity network meta-analysis (NWMA), five enrolled patients with some prior csDMARD use other than MTX,^12–16 and three did not report whether patients had used other csDMARDs.^{7, 17, 18}

Five of the eight studies enrolled samples that had previously used MTX specifically: 58¹⁹ and 79²⁰ percent of patients in two studies, and three studies (all trials) enrolling samples that were entirely MTX resistant (i.e., 100% prior use).^8–10

All included studies enrolled patients with moderate to high disease activity at baseline as measured with mean or median Disease Activity Score (DAS) 28 scores (range of 0 to 10); DAS28 scores in these studies ranged from 3.4 to 7.1. A DAS28 score of 3.2 is the threshold for low disease activity; a score exceeding 5.1 translates to high disease activity. Additional detailed information about the DAS28 is available in Appendix F.

More than one-half (ranging from 53% to 83%) of the patient population was women. The mean age range was 46 to 64 years. Study durations ranged from 6 months to 15 years.

We grouped studies based on the primary drug therapy of interest, ordered from oldest (corticosteroids and csDMARDs) to newest (TNF or non-TNF biologics), and then the most complex (combination therapies). We describe the main findings for each group below.

Corticosteroids: Eight RCTs evaluated corticosteroids, and one single-arm observational study provided additional data on harms. A corticosteroid, when taken with a csDMARD (usually MTX), led to higher remission rates than the csDMARD alone (from 44.8% to 76.7% for combination therapy and 27.8% to 33.3% for MTX monotherapy) (low SOE). Groups did not differ significantly in terms of serious adverse events and discontinuations attributable to adverse events (graded moderate and low SOE, respectively). We could not draw conclusions about disease activity, radiographic changes, or functional capacity because evidence was insufficient.

csDMARDs and tsDMARDs: Twelve RCTs, two observational studies with control groups, and four single-arm observational studies evaluated csDMARDs; only one of these studies compared a tsDMARD (tofacitinib) with a csDMARD (MTX) (Table B). These studies predominantly compared sulfasalazine plus MTX with MTX only. When comparing various csDMARD combination therapies with csDMARD monotherapies, we concluded that groups did not differ in response, remission, functional capacity, serious adverse events, or discontinuations attributable to adverse events (low SOE). Evidence was insufficient to draw conclusions about radiographic changes, csDMARD monotherapies compared with other csDMARD monotherapies, or tofacitinib compared with MTX.

When comparing csDMARD (MTX) plus TNF biologic therapy (adalimumab [ADA]) with ADA only, we concluded that combination treatment led to greater response and remission, less radiographic progression, and greater improvement in functional capacity (moderate SOE) (Table C). The groups also did not differ in serious adverse events or discontinuations attributable to adverse events (moderate SOE).

Treatment with a csDMARD (MTX) plus a non-TNF biologic (TCZ) led to greater remission than TCZ biologic monotherapy, respectively (low SOE), but groups receiving treatment with MTX plus another non-TNF biologic (abatacept [ABA]) did not differ in response or remission from those receiving ABA monotherapy (Table C). Groups receiving MTX plus ABA did not differ from those receiving ABA monotherapy in functional activity (low SOE). The groups also did not differ in serious adverse events or discontinuations attributable to adverse events (low SOE for ABA and moderate SOE for TCZ). Evidence was insufficient to draw conclusions about disease activity for these comparisons or about functional capacity for MTX plus TCZ compared with TCZ.

Biologic DMARDs: Twenty-two RCTs and one single-arm observational study evaluated TNF and non-TNF biologic drugs for treating patients with early RA. Of these, 22 evaluated disease activity, functional capacity, and harms outcomes (Table C). The combination of either a TNF or a non-TNF biologic with MTX, when compared with MTX alone, generally reduced disease activity (mostly moderate and low SOE) and led to higher rates of remission (all moderate and low SOE) and less radiographic progression (mostly moderate and low SOE). Network meta-analyses and head-to-head trials found higher ACR50 response for combination therapy of biologic DMARDs plus MTX than MTX monotherapy (NWMA range of relative risks [RRs], 1.20 [95% confidence interval [CI], 1.04 to 1.38] to 1.57 [95% CI, 1.30 to 1.88]). The groups did not differ with respect to harms (all low SOE).

Table B

Benefits and harms of csDMARDs versus csDMARD or tsDMARD for treatment of patients with early rheumatoid arthritis.

Table C

Benefits and harms of biologic DMARDs for early RA treatment.

The combinations of several TNFs (ADA, CZP, IFX) and non-TNF biologics (rituximab) plus MTX also produced greater functional capacity than MTX monotherapy (all moderate or low SOE). The results for the remainder of the biologics (ETN, ABA, TCZ) were inconclusive. IFX (TNF) plus MTX, when compared with csDMARD combination therapy, resulted in reduced disease activity, but the groups did not differ with respect to other outcomes. Likewise, when IFX was combined with multiple csDMARDs (MTX + SSZ + hydroxychloroquine) and prednisone (PRED) and compared with csDMARD combination therapies plus PRED, outcomes did not differ. No data are available for IFX monotherapy; it is approved and generally given in combination with MTX.

NWMA found higher overall discontinuation rates for MTX monotherapy than combination therapy consisting of biologic DMARDs (ADA, CZP, ETN) plus MTX (range of RR, 1.52 [95% CI, 1.02 to 2.27] to 1.77 [95% CI, 1.32 to 2.36]). However, neither serious adverse events nor discontinuations attributable to adverse events differed between the groups (low SOE). Lack of efficacy is a possible reason that patients may have discontinued the therapy or withdrawn from these studies. Evidence was insufficient for drawing conclusions about several other drug therapy combinations or head-to-head comparisons.

Combination Therapies and Treatment Strategies: Four RCTs evaluated different combination therapies and treatment strategies for early RA with moderate to high disease activity; in addition, two observational studies contributed data on harms. Patients receiving combination therapy containing MTX, SSZ plus tapered high-dose PRED (60 mg/day tapered to 7.5 mg/day), or MTX plus IFX (TNF biologic) had lower disease activity (moderate SOE) and greater functional capacity (low SOE) at 1 year (DAS<2.4: 71 to 74% vs. 53 to 64%) and less radiographic progression (moderate SOE) at 4 and 5 years (modified Total Sharp/van der Heijde score [mTSS]: 2.5 to 3.0 vs. 5.0 to 5.5) than patients receiving sequential csDMARD or step-up combination therapies starting with MTX. Groups did not differ with respect to remission (moderate SOE), serious adverse events (low SOE), or other outcomes over the longer term. We could not draw any conclusions about immediate or step-up combination therapies containing MTX and either additional csDMARD(s) or ETN (TNF biologic).

Results Among Subgroups of Patients: Only four RCTs compared drug therapies among different subpopulations defined by demographics, disease activity, or coexisting conditions (Table A). We could not draw any conclusions about response rates between older and younger patients or about response rate and radiographic changes between people with different levels of disease activity who were taking MTX with or without a TNF biologic (ADA or IFX). Evidence was also insufficient to draw any conclusions about serious adverse events as defined by FDA between older and younger patients who were taking MTX or the TNF biologic ETN. No data were available for the other agents.

Discussion and Findings in Context

We conducted a systematic review and NWMA to update the 2012 review of the comparative effectiveness of drug therapies for rheumatoid arthritis (RA);¹ in this report we focused solely on early RA in adults (within 1 year of diagnosis). Although level of disease severity was not used as a criterion to determine study eligibility, all of our early RA studies included patients with moderate to high disease activity. In a clinical setting, patients with early RA may present with varying levels of severity.

Current clinical practice guideline recommendations for therapy for patients with early RA and moderate-to-high disease severity are consistent with our findings but ours go further and also support additional therapies in patients with moderate to high disease activity.^{42, 43} When disease activity remains moderate or high (DAS28 ≥ 3.2) despite initial treatment, the ACR RA guidelines recommend double or triple csDMARD therapy or a TNF or non-TNF biologic (with or without MTX). We found that when biologics were used in combination with MTX therapy, patients achieved better disease control, higher functional capacity, and higher remission rates than with biologics or MTX monotherapy

This report assessed the comparative effectiveness based on current evidence. While not directly comparable, the ACR clinical guidelines move beyond evidence to make recommendations when evidence is limited. Clinical practice guidelines use systematic reviews as evidence and if evidence is not enough they may consider other resources. The recommendations were based on a consideration of the balance of relative benefits and harms of the treatment options under consideration and the quality of the evidence. Additionally, the ACR recommendations included consideration of patients’ values and preferences.

Although the literature in this review supports the effectiveness of MTX plus biologics in early RA for patients with medium and high disease activity, it is not currently the standard of care for a number of potential reasons. Some data indicate that certain patients with early disease may respond well to MTX monotherapy, although no information is available about how to predict which patients will do so. Second, many insurers require inadequate response to MTX as a prerequisite to adding a biologic (this policy is probably based on findings of the effectiveness of MTX). Third, patients may be wary, for a variety of reasons, of a combination therapy approach in early disease (e.g., cost, side effects, injections). Additionally, patients may find it difficult to balance the burden of multiple drugs and potentially higher risks.

Current European League Against Rheumatism early RA guidelines recommend adding a TNF or non-TNF biologic to a csDMARD but only when patients have poor prognostic factors (e.g., high disease activity, early joint damage, autoantibody positivity).⁴³ The available evidence in this review (from 10 studies comparing combinations of biologics and MTX with either biologic or MTX monotherapy) supports this recommendation. Specifically, these studies indicate that patients receiving combination therapies may achieve higher remission rates than those receiving monotherapy.^{12–15, 17, 32–34, 37, 41}

However, our data were limited because we did not find available studies that specifically examined therapies in patients with early RA and less severe disease activity compared with patients with early RA plus poor prognostic factors.

Contextual Questions: In one review comparing early versus delayed treatment trials, RA patients treated immediately at presentation with csDMARDs had improved patient function and reduced radiographic progression than patients whose DMARD treatment had been delayed 6 to 12 months.⁴⁴ Some of the barriers preventing early RA patients from accessing indicated drug therapies included access to primary health care services, difficulties in diagnosing RA in the primary care setting, obtaining of insurer approval of biologic DMARDs, high out-of-pocket expenses, and limited access to specialty care, especially in rural areas.⁴⁵ Other challenges included contraindications for some drug therapies, especially among patients with coexisting conditions and older patients, and patients’ reluctance to begin therapies.^{46, 47}

Key Limitations and Research Gaps