INTRODUCTION — The treatment of rheumatoid arthritis (RA) is directed toward the control of synovitis and the prevention of joint injury. In patients with disease resistant to methotrexate (MTX) monotherapy, and whose condition is also resistant to an initial course of a biologic disease-modifying antirheumatic drug (bDMARD), usually a tumor necrosis factor (TNF) inhibitor combined with MTX, further adjustments in the treatment regimen are required to achieve effective disease control and to prevent damage to the joints. The achievement and maintenance of tight control of disease by use of DMARDs is associated with improved outcomes, as these medications and strategies have the potential to control synovitis and to slow or even stop radiographic progression [1-11]. (See "General principles and overview of management of rheumatoid arthritis in adults", section on 'Tight control'.)
The treatment of active RA in adults who are resistant to initial therapy with a bDMARD, usually a TNF inhibitor in the context of ongoing treatment with MTX, will be reviewed here. Details of the general principles of the management of RA, the initial treatment of RA and use of nonpharmacologic and preventive measures, the treatment of patients resistant to initial therapy with nonbiologic conventional synthetic (cs) DMARDs (eg, MTX monotherapy), and the approach to RA patients with severe structural damage are presented separately:
●(See "General principles and overview of management of rheumatoid arthritis in adults".)
●(See "Initial treatment of rheumatoid arthritis in adults".)
●(See "Nonpharmacologic therapies for patients with rheumatoid arthritis".)
●(See "Alternatives to methotrexate for the initial treatment of rheumatoid arthritis in adults".)
●(See "Evaluation and medical management of end-stage rheumatoid arthritis".)
●(See "Total joint replacement for severe rheumatoid arthritis".)
DEFINITION OF RESISTANCE TO DMARDs — Resistance or an inadequate response to initial or subsequent disease-modifying antirheumatic drug (DMARD) therapy is defined as one of the following:
●Failure to achieve remission or low disease activity within three to six months of initiating the nonbiologic (conventional synthetic [cs]), biologic (b), or targeted synthetic (ts) DMARD in maximally tolerated doses within the usual therapeutic range. A therapeutic trial of greater than three months is generally used in patients with partial responses showing progressive improvement, particularly in those with low to moderate levels of disease activity and with limited functional impairment. Treatment should be appropriate for the patient's overall medical status and comorbidities, and should take into consideration the patient's treatment goals and preferences. (See "General principles and overview of management of rheumatoid arthritis in adults", section on 'Assessment and monitoring' and "General principles and overview of management of rheumatoid arthritis in adults", section on 'Tight control' and "Initial treatment of rheumatoid arthritis in adults" and "Assessment of rheumatoid arthritis disease activity and physical function", section on 'Criteria for remission'.)
●A requirement, in addition to DMARDs, for chronic glucocorticoid therapy in a dose of greater than approximately 5 mg/day of prednisone or equivalent to achieve or maintain remission or low disease activity after three to six months of treatment with DMARDs.
●A requirement for multiple courses of treatment with glucocorticoids, in excess of doses used for chronic therapy, for the treatment of recurrent disease flares in patients whose DMARD doses have been increased to the maximally tolerated or acceptable level within the usual therapeutic range.
●Continued progression of erosive disease or structural damage that is not accounted for by prior joint damage.
Resistance or an inadequate response to subsequent DMARD therapy may be defined similarly, depending upon the treatment goals in an individual patient. (See "General principles and overview of management of rheumatoid arthritis in adults", section on 'Tight control'.)
Determination of inflammatory disease activity must account for joint symptoms related to prior joint damage and disability, as well as a concomitant illness, such as fibromyalgia. Symptoms due to these other causes would not benefit from the addition of immunomodulatory therapy with DMARDs.
PRETREATMENT INTERVENTIONS — A number of important precautions should be taken before using disease-modifying antirheumatic drugs (DMARDs), including laboratory assessment (complete blood count, serum creatinine, aminotransferases, and other studies as indicated); evaluation of comorbidities; vaccinations; and screening for hepatitis C, hepatitis B, and latent tuberculosis (TB) infection. Precautions relevant to the use of each new agent being prescribed should be reviewed before initiating such therapy to confirm that all appropriate measures have been performed. Some experts also obtain a baseline chest radiograph prior to initiating treatment with methotrexate (MTX) in patients in whom a recent chest radiograph is not available. These issues are discussed in detail elsewhere. (See "Immunizations in autoimmune inflammatory rheumatic disease in adults" and "Hepatitis B virus reactivation associated with immunosuppressive therapy" and "Tumor necrosis factor-alpha inhibitors and mycobacterial infections" and "Major side effects of low-dose methotrexate", section on 'Pulmonary toxicity' and "General principles and overview of management of rheumatoid arthritis in adults", section on 'Pretreatment evaluation' and "Tuberculosis infection (latent tuberculosis) in adults: Approach to diagnosis (screening)".)
PHARMACOTHERAPY
Approach to therapy — A number of potent biologic agents directed against several different pathways and cells, as well as targeted synthetic medications with similar efficacy and safety to the biologics, are available for the treatment of active rheumatoid arthritis (RA), and experience with the newer or more established of these drugs is increasing; for patients in whom an initial trial of a given biologic agent is inadequate, we typically switch to another of these highly effective drugs that has a different mechanism of action.
Agents from all of these classes appear approximately equivalent therapeutically (see 'Efficacy of switching strategy' below), but they have individual characteristics that make each class or drug more or less appropriate for a given patient (see 'Choice of therapy' below and 'Drug dosing, efficacy, and safety' below). However, the frequency of responses to different tumor necrosis factor (TNF) inhibitors, with which there has been the greatest experience, tends to decrease with successive switching from one TNF inhibitor to another [12-16]; and patients who have had an inadequate primary response to a TNF inhibitor as the first biologic appear to respond better to an agent other than a TNF inhibitor as the next choice [17-19].
Thus, for patients who do not respond adequately to therapy with methotrexate (MTX) plus an initial biologic disease-modifying antirheumatic drug (bDMARD) or targeted synthetic (ts) DMARD (ie, a Janus kinase [JAK] inhibitor), we suggest continuing the MTX and switching from the initial bDMARD or tsDMARD to a bDMARD or tsDMARD with a different mechanism of action or to triple therapy with the conventional synthetic (cs) DMARD combination of MTX plus hydroxychloroquine (HCQ) and sulfasalazine (SSZ). We prefer this approach over switching to another agent in the same class (eg, from one to a second TNF inhibitor), particularly in patients with primary drug failure (ie, an inadequate initial response rather than a non-class-specific adverse reaction or secondary failure associated with anti-drug antibody development).
As an example, a patient with an inadequate response to MTX plus a TNF inhibitor (eg, etanercept, adalimumab, infliximab, certolizumab, golimumab, or biosimilar TNF inhibitor) would continue MTX and receive either a non-TNF inhibitor biologic (eg, abatacept, an interleukin [IL] 6 inhibitor [eg, the IL-6 receptor blockers tocilizumab or sarilumab], or rituximab); a tsDMARD (eg, one of the JAK inhibitors, including tofacitinib, baricitinib, upadacitinib, filgotinib, or peficitinib); or csDMARD triple therapy (if not used previously). Some experts prefer to initially switch to a different TNF inhibitor in this circumstance, given the greater experience with these medications than other biologics or the JAK inhibitors. Similarly and as another example, in a patient whose initial biologic therapy was the IL-6 inhibitor sarilumab, options would include one of the TNF inhibitors, a non-TNF inhibitor biologic other than one of the IL-6 inhibitors, a tsDMARD, or csDMARD triple therapy.
We prefer these agents (noted above) over anakinra, as they all appear more effective than anakinra, based upon their comparability to bDMARDs that have been shown superior to anakinra in indirect comparisons [20]. (See 'Resistant to standard therapies' below.)
We generally continue MTX in such patients, especially with TNF inhibitor therapy, where combination therapy is superior to monotherapy [21], unless contraindicated, to improve the degree of clinical and structural anatomic benefit and to inhibit the development of antibodies directed at the biologic agent that can reduce efficacy. In patients unable to take MTX, a reasonable alternative is leflunomide (LEF), except in women of childbearing age who are not using adequate contraception. These agents may be used without MTX or LEF in patients in whom MTX or LEF are contraindicated due to prior intolerance, pregnancy, or other factors. The JAK inhibitors have been frequently used as monotherapy due to the lack of antidrug antibody formation with small molecule therapies. Several studies have demonstrated good efficacy in this population with JAK inhibitor monotherapy [22,23]. Tocilizumab monotherapy is also frequently utilized with similar clinical efficacy, due to the low frequency of antidrug antibody formation, although there is possibly less radiographic protection compared with combination therapy [24-27].
Importantly, we do not combine any of the bDMARDs with each other or combine a bDMARD with a tsDMARD (eg, a JAK inhibitor). These regimens are known to or likely to be associated with an increased frequency of severe adverse effects, particularly serious infections, compared with combinations of a nonbiologic (ie, conventional synthetic) DMARD with a biologic or targeted synthetic DMARD. (See "General principles and overview of management of rheumatoid arthritis in adults".)
Efficacy of switching strategy — Several types of evidence support our approach to patients who have had an inadequate response to a first biologic. Only a few randomized trials, several large registry-based studies, and case series have demonstrated the benefit of switching from one bDMARD to another or to a tsDMARD (ie, JAK inhibitor) if inefficacy or toxicity limits the use of a given agent [12-18,28-34].
Most trials in patients with an inadequate response to an initial biologic, usually a TNF inhibitor, have demonstrated the effectiveness of non-TNF inhibitor agents compared with placebo controls [28-32], or they have evaluated responses to successive TNF inhibitors, in which decreasing response rates are seen with each switch [12-16]. Only a limited number of randomized trials have been designed to directly compare individual biologic agents with tsDMARDs (ie, JAK inhibitors) in patients who have had an inadequate response to a TNF inhibitor. A randomized trial comparing upadacitinib with abatacept in patients with RA with an inadequate response to bDMARDs demonstrated superior clinical efficacy at week 12 with upadacitinib, with less of a difference reported by 24 weeks. Adverse events and serious adverse events were more common with upadacitinib [35]. (See 'JAK inhibitors' below.)
The benefits of switching from a first TNF inhibitor in a patient with an insufficient response to a non-TNF-targeted biologic have been illustrated by an open-label pragmatic randomized trial, which compared the use of a second TNF inhibitor with the use of any one of three other non-TNF-targeted biologic agents [17]. In this trial, involving 300 patients with active RA resistant to an initial TNF inhibitor, patients were randomly assigned to receive either a non-TNF-targeted agent (non-TNF; abatacept, rituximab, or tocilizumab) or a different TNF inhibitor (etanercept, adalimumab, certolizumab, or infliximab). The choice of agent within the group of drugs to which the patient was randomized was made by the individual clinician caring for each patient. At week 24, a good or moderate European Alliance of Associations for Rheumatology (EULAR; formerly known as European League Against Rheumatism) response was significantly more frequent in the patients who received a non-TNF biologic (69 versus 52 percent, odds ratio [OR] 2.06, 95% CI 1.27-3.37). These data provide some of the best evidence supporting this approach, until sufficiently large and well-designed randomized pair-wise comparisons in this population are available, despite some limitations of this pragmatic trial strategy, such as lack of blinding, controls for co-interventions, and common or specified follow-up protocols; and a potential risk of bias in the outcome assessments [36].
Several network meta-analyses of randomized trials have provided largely indirect comparisons of a number of these agents, primarily based upon placebo-controlled trials, in patients who have had an inadequate response to a biologic agent [19,33,34,37]; these have had mixed results, with varying absolute and relative benefit with different agents, but in the aggregate have found evidence that switching to any of the bDMARDs or tsDMARDs with a different mechanism of action than the original biologic resulted in clinically significant levels of benefit.
Several observational studies have also suggested that rituximab has greater benefit than a second TNF inhibitor after a first TNF inhibitor is inadequate [18,38].
Choice of therapy
Major factors affecting drug choice — Treatment options for patients with an inadequate response to initial therapy with a bDMARD or equivalent agent should be individualized based upon patient-specific medical characteristics and preferences, as the bDMARDs, tsDMARDs, and csDMARD triple therapy that are available for this purpose have similar efficacy in this setting but other distinguishing characteristics that may favor or disfavor a particular choice in a given patient. (See 'Efficacy of switching strategy' above and 'Drug dosing, efficacy, and safety' below.)
The major factors upon which individualization of drug choices are based include:
●Mechanism of action of the initial biologic or tsDMARD – A drug with a different mechanism of action should be chosen. As an example, a patient with an inadequate response to a TNF inhibitor would be treated by switching to a non-TNF inhibitor therapy, such as abatacept, an IL-6 inhibitor, or a JAK inhibitor before trying a second TNF inhibitor. (See 'Efficacy of switching strategy' above.)
●Patient comorbidities – Patients with certain comorbidities should avoid medications that may be more likely to result in an adverse effect when such avoidance is feasible. As examples, patients with chronic obstructive pulmonary disease (COPD) were found in clinical trials to be at greater risk for respiratory adverse events with abatacept [39,40], although more recent real-world data have not demonstrated an increased risk; older adults and patients with a history of cardiovascular risk factors may be at greater risk of major adverse cardiovascular events (MACE), malignancies, and venous thromboembolism (VTE) with use of a JAK inhibitor; IL-6 inhibitors and JAK inhibitors may not be preferred in patients with lower white blood cell counts or diverticulitis, which may increase the risk of colonic perforation [41]. Rituximab might be preferred over other agents in patients with a history of lymphoproliferative malignancy.
●Pregnancy status and planning – Patients planning pregnancy should not use MTX, LEF, or JAK inhibitors, and there is a paucity of data regarding drug safety for some of the DMARDs, as discussed separately. (See "Rheumatoid arthritis and pregnancy".)
●Preference for a particular route of administration – Some patients may strongly prefer an oral agent, while others might find subcutaneous self-administration or an intravenous infusion at a more or less frequent interval acceptable.
●Regulatory or insurance limitations and patient cost – Government agencies or insurers may have requirements that limit the availability of certain agents or require use of certain medications before others are permitted. Out-of-pocket costs of some drugs may limit their availability for some patients.
Patient populations affecting drug choice — Characteristics of the individual agents, including drug administration, adverse effects, precautions regarding comorbidities, and other features, are described in detail for each agent and class (see 'Drug dosing, efficacy, and safety' below). Briefly, the following groups of patients have comorbidities, preferences, or regulatory/cost issues that affect drug choice:
●Older adults and patients with significant comorbidities – In older patients with significant risk for infection, we prefer abatacept or etanercept, and in patients with a history of malignancy, we prefer rituximab. However, other bDMARDs (eg, the IL-6 inhibitors tocilizumab and sarilumab) are appropriate alternative therapies. Based upon clinical trial data and European Medicines Agency (EMA)/US Food and Drug Administration (FDA) warnings, the available tsDMARDs (ie, JAK inhibitors) would only be used after failure of other bDMARDs due to a numerical increase in risk of MACE, malignancies, and overall mortality seen in patients 65 years or older [42]. Tocilizumab and sarilumab should be avoided in patients with absolute neutrophil counts <2000/mm3, platelet counts <100,000/mm3, or if alanine aminotransferase or aspartate aminotransferase levels are >1.5 times the upper limit of normal; and patients with cytopenias or abnormal liver testing who do receive these drugs should be closely monitored (see 'IL-6 inhibitor therapies' below). Additionally, patients with a history of venous thromboembolic disease should avoid use of a JAK inhibitor. (See 'JAK inhibitors' below.)
There is limited direct evidence to support the preference of abatacept over bDMARDs or tsDMARDs. A comparison of abatacept with upadacitinib in bDMARD incomplete responders demonstrated superior efficacy at 12 weeks for upadacitinib but more serious adverse events; treatment discontinuations and liver function abnormalities were reported with upadacitinib [35]. Indirect comparisons also suggest that abatacept may have fewer adverse effects than the other biologic agents used in RA [43]. In addition, in a cohort study using Medicare data and involving 31,801 new treatment episodes for RA with a biologic agent, patients with prior exposure to a biologic had a significantly higher risk for hospitalized infections upon exposure to etanercept, infliximab, or rituximab, compared with abatacept (hazard ratio [HR] 1.24, 95% CI 1.07-1.45; HR 1.39, 95% CI 1.21-1.53; and HR 1.36, 95% CI 1.21-1.53, respectively) [44]. Numerical trends suggesting increased rates with the other agents were not statistically significant. Crude incidence rates ranged from 13.1 (abatacept) to 18.7 (rituximab) per 100 patient-years. (See 'Abatacept' below and 'IL-6 inhibitor therapies' below and 'Rituximab' below.)
However, observational studies comparing DMARDs should be interpreted with particular caution, given evidence that differences in age, medical history, and disease activity can substantially confound nonrandomized comparative studies of available bDMARDs' safety and effectiveness [45].
●Preference for very reduced parenteral dosing frequency – Rituximab may be preferred in patients in whom monthly drug administration may be problematic, as a treatment course of two intravenous infusions is administered no more frequently than every six months. Patients with rheumatoid factor or with anti-citrullinated peptide antibodies may have a greater response to the drug than seronegative patients [46,47]. Otherwise, the efficacy and safety of rituximab in patients with RA appear similar to those of other biologic agents [43,48]. (See 'Rituximab' below.)
●Preference for an oral agent – The choice between orally administered treatment options depends upon prior drug usage, age, and cardiovascular risk. In patients who prefer an oral route of administration, we had previously preferred a JAK inhibitor or triple therapy (MTX plus SSZ plus HCQ) for patients who have not tried this combination prior to biologic usage; this was consistent with data from patients with RA enrolled in the randomized trials of the JAK inhibitors that suggested that they are associated with similar rates of infection and overall mortality to those reported in RA for the bDMARDs [49]. However, a trial of tofacitinib compared with TNF inhibitors in RA patients 50 years or older with cardiovascular risk factors demonstrated that tofacitinib 5 and 10 mg twice daily was associated with a numerical increase in risk of MACE, malignancies, and pulmonary embolism/VTE. Based upon this trial, the FDA recommended that all JAK inhibitors only be used after TNF inhibitors in csDMARD incomplete responders [50,51]. In view of this FDA warning, we would prefer triple therapy over a JAK inhibitor in this higher-risk population. (See 'JAK inhibitors' below.)
●History of serious adverse event or primary failure with initial TNF inhibitor – Some experts advocate switching to another class of bDMARD or tsDMARD (JAK inhibitor), particularly in patients who have experienced a serious adverse event (as defined by the FDA) with an initial TNF inhibitor [52] or in those who have had little or no initial response (primary failures). The FDA definition of a serious adverse event includes an adverse event associated with the use of a medical product that resulted in a life-threatening event, hospitalization, prolongation of a hospitalization, disability or permanent damage, a congenital anomaly or other disability since birth in a child of a parent using the medication, or an adverse event requiring intervention to prevent permanent impairment or damage.
●Lack of prior use of conventional nonbiologic DMARD triple therapy – In patients in whom triple therapy was not employed prior to the initiation of MTX plus anti-TNF therapy, an effective alternative to switching to another biologic is a switch to (or addition of) triple therapy, which has been shown to result in significant improvement in this setting [53]. (See "Treatment of rheumatoid arthritis in adults resistant to initial conventional synthetic (nonbiologic) DMARD therapy", section on 'Efficacy of triple DMARD therapy versus methotrexate/TNF inhibitor'.)
●Seropositive versus seronegative patients – In seropositive patients for rheumatoid factor or anti-citrullinated peptide antibody, rituximab and abatacept, as well as tofacitinib [54], have been found more effective than in seronegative patients (see 'Abatacept' below and 'Rituximab' below and 'Tofacitinib' below). The IL-6 receptor antibodies and most analyses of tsDMARD trials have demonstrated similar benefit in seropositive and seronegative patients. (See "Progressive multifocal leukoencephalopathy (PML): Epidemiology, clinical manifestations, and diagnosis" and "Rituximab: Principles of use and adverse effects in rheumatoid arthritis", section on 'Opportunistic infections and viral reactivation' and 'Abatacept' below and 'Tocilizumab' below and 'Rituximab' below.)
Drug dosing, efficacy, and safety
Abatacept — Abatacept is a soluble fusion protein that consists of cytotoxic T-lymphocyte-associated antigen 4 (CTLA4) and the Fc portion of immunoglobulin G1 (IgG1); thus, it is also termed CTLA4-Ig. Through its high-affinity binding to CD80 (B7-1) and CD86 (B7-2), it inhibits transmission to CD28 by these CD28 ligands of the second signal required for T-cell activation [55,56].
●Dosing and precautions – In patients for whom MTX plus abatacept is the choice of therapy following an inadequate response to an initial or subsequent biologic agent, such as one or more of the TNF inhibitors, abatacept can be administered either by intravenous infusion or subcutaneous self-administration [57-60]:
•For intravenous use, it is administered every four weeks (750 mg per dose for patients 60 to 100 kg, adjusted for lower or higher weight to 500 or 1000 mg, respectively) after the three initial doses given at two-week intervals.
•For subcutaneous use, if initiating with an intravenous loading dose, the intravenous infusion (using the weight-based dosing) should be administered initially, followed by administration of 125 mg subcutaneously within 24 hours of the infusion, then followed by 125 mg subcutaneously once weekly thereafter.
•Patients whose disease is well controlled on long-term intravenous abatacept can switch to receiving the medication by subcutaneous administration while maintaining clinical efficacy and without increased safety issues [59]. For transitioning from intravenous to subcutaneous therapy, the first subcutaneous dose should be administered instead of the next scheduled intravenous dose.
The decision regarding route of administration can be based upon patient preference; the efficacy and safety of the subcutaneous and intravenous preparations of abatacept were comparable in a large randomized trial involving patients with a previous inadequate response to MTX [58], but have not been compared in patients who have previously received another biologic agent. We generally use abatacept in combination with MTX, after appropriate pretreatment measures have been performed, but it may also be administered as monotherapy or in combination with other nonbiologic DMARDs [61]. It should not be used in combination with other bDMARDs, such as TNF inhibitors, because of an increased risk of serious adverse events, including infections [62].
Potential adverse effects of abatacept include infusion reactions, which may occur within an hour after beginning the intravenous administration of the drug and which may be characterized by headache, dizziness, and hypertension; anaphylactoid reactions are rare. Abatacept also appears to increase the risk of serious infections, including pneumonia, pyelonephritis, cellulitis, and diverticulitis. In patients with COPD, if abatacept is used, it should be with particular caution because of the higher rates of COPD exacerbations and respiratory tract infections reported in such patients in the randomized trials of this agent. A definite association with TB has not been shown, although screening for latent TB prior to treatment should be performed. Treatment of RA patients with abatacept has not been associated with an increased frequency of malignancy in the RA clinical trials participants [63]. Live vaccines should not be given concurrently or within three months of stopping abatacept.
The overall safety of abatacept appears comparable to or possibly slightly better than that of the TNF inhibitors, although there are few direct comparisons. An indirect comparison of biologic agents in data from a 2011 meta-analysis of randomized trials and extension studies showed a statistically nonsignificant trend for abatacept compared with the other agents toward fewer serious adverse events (OR 0.65, 95% CI 0.42-1.01) and serious infections (OR 0.57, 95% CI 0.30-1.08), while other biologics generally showed similar risks compared with each other [43]. A direct comparison of abatacept with upadacitinib in RA patients with active disease despite bDMARDs demonstrated more frequent adverse events and serious adverse events with upadacitinib [35].
●Efficacy of abatacept – A randomized trial and a larger open-label study have described the benefits of abatacept compared with placebo in patients who have had an inadequate response to a TNF inhibitor [28,64]. Most trials and other studies documenting the efficacy of abatacept in patients with RA have been in patients responding inadequately to MTX, but who are naïve to biologic agents. (See "Treatment of rheumatoid arthritis in adults resistant to initial conventional synthetic (nonbiologic) DMARD therapy", section on 'Methotrexate plus abatacept'.)
In the randomized trial, which involved 391 patients on a nonbiologic DMARD, usually MTX, with an inadequate response to at least three months of treatment with a TNF inhibitor, patients receiving abatacept were significantly more likely compared with placebo-treated patients to achieve at least American College of Rheumatology response criteria for 20 percent improvement in the level of disease activity after 24 weeks of therapy (ACR20 of 50 versus 20 percent) [28]. The rate of serious infections was the same in both groups (2.3 percent).
In a large observational study, continued use of intravenous abatacept after two years was more likely in biologic-naïve patients compared with patients with a prior inadequate response to bDMARDs (55 versus 45 percent) [65]; the study also showed greater retention rates in patients who were both rheumatoid factor and anti-citrullinated peptide antibody positive compared with those who were not.
IL-6 inhibitor therapies — There are two interleukin (IL) 6 receptor inhibitors, tocilizumab and sarilumab, available for use together with MTX or as monotherapy, which have shown efficacy in patients resistant to other bDMARDs, including TNF inhibitors. Both agents are anti-IL-6 receptor antibodies. (See 'Tocilizumab' below and 'Sarilumab' below.)
A number of randomized trials have demonstrated the benefits of the IL-6 inhibitors, using a variety of different outcomes measures. However, caution should be exercised when interpreting trial results for the IL-6 inhibitors because these agents have pronounced effects on levels of acute phase reactants, which can influence the calculation of some composite outcome measures. As an example, the Disease Activity Score in 28 joints (DAS28), which gives relatively greater weight to levels of acute phase reactants, can overestimate the clinical response compared with the use of measures that rely less or not at all on acute phase reactants, such as the Simplified Disease Activity Index (SDAI) and the Clinical Disease Activity Index (CDAI) [66]. (See "Assessment of rheumatoid arthritis disease activity and physical function".)
Tocilizumab — Tocilizumab, a humanized IgG1 anti-human IL-6 receptor antibody, competes for both the membrane-bound and the soluble forms of human IL-6 receptor, thereby inhibiting the binding of the native cytokine to its receptor and interfering with the cytokine's effects.
●Dosing, administration, and adverse effects – It can be administered either by intravenous infusion or by subcutaneous injection:
•Intravenous administration – When administered intravenously, it is given every four weeks (at an initial dose of 4 mg/kg per infusion, which may be increased to 8 mg/kg per infusion, based upon the clinical response, to a maximum of 800 mg/infusion).
•Subcutaneous administration – When taken subcutaneously in patients <100 kg, the dose is 162 mg every other week, which may be increased to every week based upon the clinical response; in patients ≥100 kg, the dose is 162 mg every week. The subcutaneous route of administration (162 mg) has comparable efficacy and safety when compared with the intravenous route (8 mg/kg) [67].
Tocilizumab is available for use in RA patients with active disease despite treatment with MTX, but we use it primarily in patients with RA who have not responded adequately to TNF inhibitors. It is usually used together with MTX but can be given as monotherapy in patients in whom MTX is contraindicated. An analysis of data from multiple European patient registries, involving 2057 patients treated with tocilizumab, found that cotherapy of tocilizumab with a nonbiologic (traditional) DMARD or DMARD combination had similar efficacy to tocilizumab monotherapy, although patients receiving tocilizumab alone discontinued the therapy earlier [24]. Over 80 percent of the study patients on a nonbiologic DMARD were receiving MTX.
The drug may cause dyslipidemia but is generally well tolerated. In a systematic review, patients receiving tocilizumab were more likely to experience an elevation in the ratios of low-density lipoprotein (LDL) to high-density lipoprotein (HDL) cholesterol (20 versus 12 percent, relative risk [RR] 1.7, 95% CI 1.2-2.2) and of total to HDL cholesterol (12 versus 7 percent, RR 1.7, 95% CI 1.2-2.6), but they were less likely, for any reason, to withdraw from treatment (8.1 versus 14.9 percent, RR 0.6, 95% CI 0.5-0.8) [68]. Dyslipidemia should be managed according to available guidelines. Despite this impact on lipids, risk of cardiovascular events with tocilizumab was similar to etanercept in a phase 4 clinical trial in RA patients with cardiovascular risk factors [69]. (See "Low-density lipoprotein cholesterol-lowering therapy in the primary prevention of cardiovascular disease" and "Management of low density lipoprotein cholesterol (LDL-C) in the secondary prevention of cardiovascular disease" and "Coronary artery disease in rheumatoid arthritis: Implications for prevention and management", section on 'Lipid lowering with statins'.)
A dose adjustment or drug discontinuation may be required in patients with significant liver enzyme (aminotransferase) elevations, neutropenia, or thrombocytopenia. Other adverse effects include serious infections, including mycobacterial and other opportunistic infections. The risk of adverse effects is greater in patients on concomitant immunosuppressive therapy. Intestinal perforations have been reported, especially in older patients and in those with a history of diverticulitis, which is a contraindication to the use of tocilizumab.
Another meta-analysis of randomized trials found that adverse events were increased in patients receiving tocilizumab (8 mg/kg) plus MTX compared with placebo plus MTX (OR 1.5, 95% CI 1.3 to 1.9), as was the rate of infection (OR 1.3, 95% CI 1.1 to 1.6) [70]. However, there was no significant increase in the rates of malignancy, TB reactivation, or hepatitis.
Tocilizumab was the initial IL-6 inhibitor approved for use in RA and systemic juvenile idiopathic arthritis (JIA) in the United States, and is available in Europe and Japan and other countries [71,72]. The use of tocilizumab in systemic JIA is discussed elsewhere. (See "Systemic juvenile idiopathic arthritis: Treatment", section on 'Interleukin 6 inhibitors'.)
●Efficacy of tocilizumab – The efficacy and safety of tocilizumab as monotherapy and together with MTX have been characterized in patients with RA in a number of randomized trials [30,73-78] and in systematic reviews and meta-analyses [68,70], usually in comparison with placebo.
The efficacy of tocilizumab was demonstrated in a systematic review of randomized trials involving 3334 patients, in which patients receiving tocilizumab (8 mg/kg administered intravenously every four weeks) were significantly more likely than those receiving placebo to achieve a clinical response consistent with ACR50 while also receiving MTX (38.8 versus 9.6 percent, RR 3.2, 95% CI 2.7-3.7) [68]. However, only one of these trials included patients resistant to a bDMARD [30]; in that randomized trial, involving 499 patients with an inadequate response to TNF inhibitor therapy, the use of tocilizumab (8 or 4 mg/kg administered intravenously every four weeks), together with continued treatment with MTX, significantly improved treatment outcomes at six months compared with placebo plus MTX (ACR20 of 50 and 30 versus 10 percent). Substantial improvement was noted within four weeks and may be maintained for at least several years in long-term follow-up [32]. The rates of serious adverse events, serious infections, and adverse events leading to drug discontinuation were similar in all three groups; adverse events that were numerically more common in patients on tocilizumab included infections (49 and 47 versus 41 percent), gastrointestinal symptoms (37 and 33 versus 19 percent), and rash (22 and 31 versus 14 percent).
In a large registry-based observational study, tocilizumab was comparably effective in patients regardless of whether or not they were positive for anti-citrullinated peptide antibodies [79].
The efficacy of tocilizumab in patients resistant to nonbiologic DMARDs is described in detail separately. (See "Treatment of rheumatoid arthritis in adults resistant to initial conventional synthetic (nonbiologic) DMARD therapy", section on 'Methotrexate plus IL-6 inhibitor/IL-6 inhibitor monotherapy'.)
Sarilumab — Another human monoclonal antibody, sarilumab, like tocilizumab, is directed against the membrane-bound and soluble forms of the IL-6 receptor and has also been effective in several clinical trials and is available for the treatment of RA [80-82]. In addition to its efficacy in patients resistant to or intolerant of conventional DMARDs, sarilumab has also been effective in patients with RA who have had an inadequate response or intolerance to TNF inhibitors.
In a randomized trial, involving 546 such patients who continued background conventional DMARDs, sarilumab (150 or 200 mg subcutaneously every two weeks) was more likely than placebo to result in an ACR20 response at week 24 (56 and 61 versus 34 percent) [82]. Improvement in physical function at week 12 was also found, compared with placebo.
The adverse effect profile and relative safety of sarilumab are similar to those reported with tocilizumab, as the mechanism of action is similar. Neutropenia, liver function test abnormalities, dyslipidemia, serious infectious episodes, and intestinal perforations have been reported with increased frequency compared with csDMARDs [83,84].
JAK inhibitors — Several orally administered small molecule Janus kinase (JAK) inhibitors, which are considered tsDMARDs, have been developed that decrease signaling by a number of cytokine and growth factor receptors. These include tofacitinib, baricitinib, upadacitinib, filgotinib, and peficitinib. There is a lack of data in patients resistant or intolerant to biologic agents for peficitinib, which is only available in Asia. (See 'Tofacitinib' below and 'Baricitinib' below and 'Upadacitinib' below and "Treatment of rheumatoid arthritis in adults resistant to initial conventional synthetic (nonbiologic) DMARD therapy".)
The JAK are cytoplasmic protein tyrosine kinases that are critical for signal transduction to the nucleus from the plasma membrane receptors for type I and II cytokines, including IL-2, IL-4, IL-6, IL-7, IL-9, IL-12, IL-15, IL-21, IL-23; type I interferons; gamma interferon; granulocyte-macrophage colony-stimulating factor (GM-CSF); and thrombopoietin and erythropoietin. For RA, these drugs are typically taken orally on a daily basis, but in clinical trials have generally similar efficacy and a range and degree of adverse effects similar to the bDMARDs. The biology, principles of use, and adverse effects of the JAK inhibitors are described in more detail separately. (See "Janus kinase inhibitors for rheumatologic and other inflammatory disorders: Biology, principles of use, and adverse effects".)
Tofacitinib — Tofacitinib is taken in an oral dose of 11 mg extended release once daily or 5 mg twice daily for RA. Tofacitinib is effective and has been approved since 2012 either as monotherapy or combined with MTX (our usual approach) or other nonbiologic DMARDs in patients with moderately to severely active RA who have had an inadequate response to MTX or to another nonbiologic DMARD [85-91]. Other nonbiologic DMARDs may be used with tofacitinib as an alternative to MTX in patients intolerant of MTX therapy. It should not be taken in combination with biologic agents or with certain other potent immunosuppressants, including azathioprine and cyclosporine. Tofacitinib preferentially inhibits JAK-1 and JAK-3, while inhibiting JAK-2 to a lesser degree in preclinical studies.
The efficacy and safety of tofacitinib in RA have been evaluated in a series of randomized trials in patients with an inadequate response to MTX or another nonbiologic DMARD or bDMARD [85-90,92].
Tofacitinib has also shown benefit as cotherapy with MTX in patients who have not had an adequate response to MTX alone; and had comparable effective to a TNF inhibitor in this setting. The drug has also been compared as monotherapy with MTX in patients who are naïve to these agents [91]. The evidence supporting the use of tofacitinib in patients with RA resistant to nonbiologic DMARD therapy and in patients naïve to DMARDs is reviewed separately. (See "Treatment of rheumatoid arthritis in adults resistant to initial conventional synthetic (nonbiologic) DMARD therapy", section on 'Tofacitinib' and "Alternatives to methotrexate for the initial treatment of rheumatoid arthritis in adults", section on 'Tofacitinib'.)
Several trials document the efficacy of tofacitinib in patients with an inadequate response to a bDMARD regimen:
●Tofacitinib has shown efficacy in patients who have not responded adequately to TNF inhibitor therapy. In a randomized trial involving 399 patients with an inadequate response to MTX plus a TNF inhibitor, the addition of tofacitinib (5 or 10 mg twice daily) to MTX, compared with placebo plus MTX, resulted at three months in a significantly greater response rate (ACR20 of 42 and 48 versus 24 percent) and in significantly greater reductions in the degree of disability (Health Assessment Questionnaire Disability Index [HAQ-DI] -0.43 and -0.46 versus -0.18) [88].
●Similar results were seen in a post-hoc subgroup analysis of patients who had an inadequate response to a biologic agent, most often a TNF inhibitor, as part of a larger randomized trial of tofacitinib monotherapy in patients resistant to nonbiologic or biologic DMARDs. Tofacitinib monotherapy (5 mg twice daily) resulted more often in reductions in signs and symptoms of active RA after three months of treatment, compared with placebo (ACR20 of 43 versus 18 percent) [85].
An analysis of combined data from five trials of tofacitinib found higher response rates among patients who were both rheumatoid factor and anti-cyclic citrullinated peptide antibody positive than those who were seronegative for both antibodies [54].
The relative safety of tofacitinib has generally appeared similar to that of bDMARDs, including increased risk of infections and liver function test abnormalities; additional concerns that require attention in clinical use include neutropenia, lymphopenia, hyperlipidemia, and, possibly, increased serum creatinine [85,86,93-95]. Gastrointestinal perforations have also been reported. An increased risk of herpes zoster has been reported compared with placebo. In the clinical trial program of tofacitinib, overall cardiovascular adverse events, malignancies, and VTEs were similar to those reported with bDMARDs. However, in a phase 3b/4 trial in patients older than age 50 with at least 1 cardiovascular risk factor comparing tofacitinib with TNF inhibitors, a numerical increase in MACE, malignancy, and pulmonary embolism/VTE was reported, and tofacitinib was not noninferior to the comparator TNF inhibitors for MACE/malignancy [96]. Patients at highest risk were males, smokers, and those over age 65. Approximately 80 percent of the MACE and malignancy events occurred in patients >65 years old or who had ever smoked.
The relative safety of tofacitinib was evaluated in an analysis of the data from patients in the randomized trials and long-term extension studies of tofacitinib, including 7061 patients with 23,194 patient-years of exposure to the drug, with safety data followed for up to 9.5 years [97,98]. Adverse event frequencies, expressed as events/100 patient-years, were serious infections, 2.5, which was stable over time; herpes zoster, 3.6; all-cause mortality, 0.3; and adverse events leading to discontinuation, 7.1. These rates were comparable to those previously reported in patients with RA receiving bDMARDs. Factors independently associated with an increased risk of serious infection with tofacitinib use were age, glucocorticoid dose, diabetes, tofacitinib dose, and lymphocyte counts of less than 0.5 x 103/mm3. Patients should discontinue tofacitinib if lymphocyte counts drop below this level.
Herpes zoster was reported in 5 percent of patients (239 cases) in the trials and extension studies; the risk of herpes zoster was significantly increased in patients receiving tofacitinib compared with those receiving placebo (incidence of 4.4 per 100 patient-years, 95% CI 3.8-4.9) [99]. Only one case was multidermatomal, and none involved visceral dissemination or death. The only factors independently associated with increased risk of herpes zoster with tofacitinib use were older age and participation in the trials in Asia. If possible, vaccination against zoster is advised prior to initiation of tofacitinib. (See "Immunizations in autoimmune inflammatory rheumatic disease in adults", section on 'Zoster vaccines'.)
Baricitinib — Baricitinib is a small molecule, orally administered, JAK-1 and JAK-2 inhibitor. It is commercially available for use in both Europe (at a dose of either 2 or 4 mg daily) for csDMARD incomplete responders and the United States for TNF inhibitor incomplete responders at the 2 mg dose. It has been effective in patients with RA compared with both placebo and with other active DMARDs and in studies with patients who are naïve to DMARD therapy as well as patients with inadequate responses to prior therapies [100-105]. Baricitinib preferentially inhibits JAK-1 and JAK-2. The efficacy of baricitinib in patients naïve to DMARDs, including in comparison to MTX; and in patients with an inadequate response or intolerance to conventional nonbiologic DMARDs, including in comparison with adalimumab are described in detail separately. (See "Treatment of rheumatoid arthritis in adults resistant to initial conventional synthetic (nonbiologic) DMARD therapy", section on 'Baricitinib' and "Alternatives to methotrexate for the initial treatment of rheumatoid arthritis in adults", section on 'Baricitinib'.)
The efficacy of baricitinib in patients resistant to a bDMARD was demonstrated in a randomized trial involving 527 patients with active RA who had an inadequate response or intolerance to prior TNF inhibitor therapy [105]. Some had also previously used other biologics. Patients were more likely to respond to baricitinib (4 mg orally once daily) than placebo at 12 weeks (ACR20 of 55 versus 27 percent). Improvements in physical function and other composite clinical response measures were also seen. In patients treated with baricitinib (4 or 2 mg daily) and placebo, respectively, serious adverse events occurred over 24 weeks of treatment in 10, 4, and 7 percent; any adverse effect in 77, 71, and 64 percent; and infections in 40, 44, and 31 percent. Nonmelanoma skin cancer and MACE were each seen in two patients (1 percent) among those receiving the 4 mg dose of baricitinib.
The relative safety of baricitinib and other JAK inhibitors in the clinical trials and observational studies appears similar. Like the other JAK inhibitors, baricitinib has a boxed warning for thromboembolic events at the 4 mg dose as in the placebo controlled RA clinical trials, 6 VTEs (1.4 per 100 patient-years) occurred on 4 mg baricitinib with none reported on 2 mg or placebo. Similar to tofacitinib, it is associated with a higher risk of herpes zoster. Studies demonstrated minor increases in serum creatinine and LDL cholesterol, and small reductions in blood neutrophil counts. As examples, from week 0 to 52 in one trial, the serum creatinine increased by 0.086 (±0.005) mg/dL; the LDL cholesterol increased by 18 (±1) mg/dL; and the neutrophil count decreased by 1230 (±90) per mm3 [104].
Upadacitinib — Upadacitinib is a small, orally active drug that preferentially inhibits JAK-1 over JAK-2, JAK-3, and non-receptor tyrosine-protein kinase, although it is not entirely specific for JAK-1. Upadacitinib (15 mg once daily) is commercially available for use in the United States in patients with moderately to severely active RA resistant or intolerant to MTX. Safety issues with upadacitinib are similar to those with the other JAK inhibitors, and in the United States, the drug carries a boxed warning regarding serious infections, malignancies, and thrombosis, as do the other JAK inhibitors. The use and efficacy of upadacitinib in patients resistant to or intolerant of csDMARDs, including MTX, are described separately. (See "Treatment of rheumatoid arthritis in adults resistant to initial conventional synthetic (nonbiologic) DMARD therapy", section on 'Upadacitinib'.)
The efficacy and safety of upadacitinib in patients with an inadequate response or intolerance to prior bDMARD(s) have been evaluated in comparison with placebo and in another trial with abatacept:
●Comparison with placebo – Efficacy and safety of upadacitinib were evaluated in a randomized trial involving 499 patients with active RA and an inadequate response or intolerance to at least one (and up to three) bDMARD(s). The addition of upadacitinib (15 or 30 mg daily) to background nonbiologic DMARD therapy was more likely, compared with placebo, to result in an ACR20 response at week 12 (65 and 56 versus 28 percent) [106]. Low disease activity (DAS28-C-reactive protein [CRP] ≤3.2) was achieved more frequently with upadacitinib at 12 weeks as well (43 and 42 versus 14 percent). Stable low-dose glucocorticoids (prednisone ≤10 mg daily or equivalent) were allowed and were taken by approximately half of the patients during the trial. Adverse events were more common with the higher upadacitinib dose, and serious adverse events occurred in 5 and 7 percent of the upadacitinib recipients but not with placebo; more patients receiving the higher upadacitinib dose had serious infections, herpes zoster, or an adverse event leading to discontinuation of therapy than the other groups. One patient receiving upadacitinib died during the study.
●Comparison with abatacept – Upadacitinib showed modestly greater efficacy but numerically more adverse events compared with abatacept in a trial involving 613 patients with moderate to severely active RA resistant to or intolerant of at least one bDMARD [35]. Patients were randomly assigned to receive either upadacitinib (15 mg daily) or abatacept (by intravenous infusion of 500, 750, or 1000 mg, based upon weight), while continuing background csDMARDs and glucocorticoids. Abatacept was administered on day 1 and after 2 and 4 weeks, then every 4 weeks. At week 12, the patients receiving upadacitinib had a greater reduction in their mean DAS28-CRP score compared with those receiving abatacept (-2.52 versus -2.00 [difference, -0.52, 95% CI -0.69 to -0.35]); and patients who received upadacitinib were more likely to achieve DAS28-CRP remission (30.0 versus 13.3 percent [difference, 16.8, 95% CI 10.4-23.2]). Serious adverse events were numerically more common in the upadacitinib group, including one death, one nonfatal stroke, and two venous thromboembolic events. Four cases of herpes zoster were seen in each group. Elevations of aminotransferase levels >3 times the upper limit of normal were also more frequent among upadacitinib patients.
Measures that would not be affected by the influence of JAK inhibition on the CRP, independent of effects upon disease activity, such as the CDAI, trended in the same direction as the results with the DAS28-CRP, but differences between groups in these analyses were less striking and sometimes not statistically significant, often showing overlapping confidence intervals. Consistent with this finding, reductions in swollen joint counts were similar in the two groups.
JAK inhibitor cycling — There are limited data about switching Janus kinase (JAK) inhibitors in RA patients with lack of efficacy or intolerance. A small study from Spain reported efficacy after switching, and a more recent preliminary report from the JAK-pot registries demonstrated similar efficacy with switching to a second JAK or switching to a biologic DMARD [107,108].
Given the benefit seen in these trials in active and difficult-to-treat RA, JAK inhibitors were comparable or superior in efficacy to bDMARDs. Overall safety in the clinical trial programs and observational registries have demonstrated comparable safety profiles to biologic DMARDs. However, the safety issues seen with tofacitinib in older patients with cardiovascular risk factors resulted in a class warning from the FDA, who recommended use only in TNF failures. For patients failing an initial bDMARD who have cardiovascular/VTE or malignancy risk factors such as smoking, we would not suggest a JAK inhibitor at this time if alternatives exist. For younger patients without risk factors, we consider JAK inhibitors appropriate options (see "Treatment of rheumatoid arthritis in adults resistant to initial conventional synthetic (nonbiologic) DMARD therapy", section on 'Upadacitinib'). Further trials, including additional comparisons with other bDMARDs and comparisons with other JAK inhibitors, will be useful in better determining its place in therapy.
Rituximab — Rituximab is a monoclonal anti-CD20 antibody that depletes B cells. It is available for use in RA, systemic vasculitis, and other rheumatic diseases. It is an important therapeutic agent for the treatment of B cell malignancies and is used in a number of other disorders.
●Use, dosing, and adverse effects – We generally administer rituximab as an intravenous infusion of 1000 mg, repeated once two weeks later, together with ongoing weekly MTX. Courses of rituximab can be administered as frequently as every four to six months depending upon disease activity, but most patients are treated less frequently due to adequate disease control. Studies have demonstrated that a treat-to-target approach based on an increase in disease activity is more effective than physician preference to treat [109]. Preliminary reports suggest lower doses of rituximab may be as effective as the approved 1000 mg dose [110]. Rituximab biosimilars are available in the clinic with comparable efficacy and safety, and data have demonstrated no significant issues with switch to biosimilar rituximab as far as efficacy, safety, or immunogenicity [111,112].
We generally wait six months after administration of rituximab before switching to an alternative medication for lack of efficacy or for adverse effects because of the duration of its biologic effects. Rituximab is given in combination with ongoing MTX treatment unless MTX is contraindicated. Use either with LEF or as monotherapy are alternatives in patients unable to take MTX. A small open-label trial [113] and observational studies with much larger numbers of patients have shown comparable efficacy and safety of rituximab in combination with either MTX or LEF [114-116]. With rituximab monotherapy, good responses to treatment are often seen [115], but discontinuation is more frequent than when it is used in combination with either MTX or LEF [114].
The use, pretreatment testing, administration, monitoring, and adverse effects of rituximab in RA are described in detail separately. (See "Rituximab: Principles of use and adverse effects in rheumatoid arthritis".)
●Efficacy – Rituximab can provide benefit following an inadequate response to a biologic agent or agents, including TNF inhibitors, and may be more effective than a second or third TNF inhibitor in many patients. However, direct comparisons of rituximab with other biologics in randomized trials are lacking.
The efficacy of rituximab plus continued therapy with MTX in patients with an inadequate response to a TNF inhibitor was shown in comparison with placebo plus continued MTX in a randomized trial involving 520 patients [29]. At six months, significantly more patients receiving rituximab showed at least 20 percent improvement in clinical activity compared with placebo-treated patients (ACR20 of 51 versus 18 percent). Additional evidence suggests benefit with retreatment six months after an initial course [117].
A registry-based study suggested that, when the reason for stopping a TNF inhibitor was lack of efficacy rather than an adverse event, rituximab resulted in a greater decrease in disease activity compared with treatment with another TNF inhibitor (DAS28 decrease of -1.34 versus -0.93) [18]. Similarly, greater benefit with switching to rituximab, compared with a second or third TNF inhibitor, has also been described in several other observational studies [38,118-120].
As an example, a registry-based study compared efficacy as measured by DAS28 at three and six months in patients who failed one or two anti-TNF agents and were switched to either rituximab or another anti-TNF agent [121]. In patients who failed one anti-TNF, rituximab was as effective as and, in some cases (when compared with infliximab), better than a second anti-TNF agent. In patients who had previously failed both etanercept and a monoclonal anti-TNF, rituximab was superior to the third anti-TNF. Similar results were observed in a multicenter, prospective, observational study in Spain in which patients switched to rituximab showed a better response than patients switched to adalimumab or infliximab [119]. In one study, the presence of a specific B lymphocyte stimulator (BLyS) promoter haplotype was associated with responsiveness to rituximab in patients with RA who had failed anti-TNF therapy and who were seropositive (for rheumatoid factor and/or anti-cyclic citrullinated peptide antibodies) [122].
Several reports indicate that positive testing for rheumatoid factor or anti-cyclic citrullinated peptide antibodies predicts a greater likelihood of responsiveness to rituximab [38,47,123-125].
Data from a large registry suggest that the efficacy of rituximab is reduced in patients over age 75 compared with younger patients at 12, 18, and 24 months, although no differences were seen at six months [126]. It is not mentioned whether there were differences between frequency of retreatment with rituximab or number and/or doses of concomitant DMARDs in the different groups [126].
TNF inhibitors — The available TNF inhibitors (etanercept, infliximab, adalimumab, certolizumab, and golimumab) are similarly effective, and the specific agents, dosing, and safety are presented in detail separately (see "Treatment of rheumatoid arthritis in adults resistant to initial conventional synthetic (nonbiologic) DMARD therapy", section on 'Methotrexate plus TNF inhibitor'). Biosimilars for etanercept, infliximab, and adalimumab have similar efficacy, safety, and immunogenicity to the originators. Immediately switching to a second TNF inhibitor without a washout period after the first appears safe and effective in such patients [127].
An inadequate response to one anti-TNF agent does not predict resistance to other agents in this class, although inefficacy and discontinuation rates increase with successive switches, particularly for ineffectiveness (ie, primary inadequate responsiveness). This is less of a concern with adverse effects that are not class-related or with the development of anti-drug antibodies that reduce drug effectiveness (secondary inadequate responsiveness). (See "Treatment of rheumatoid arthritis in adults resistant to initial conventional synthetic (nonbiologic) DMARD therapy", section on 'Definition of resistance to initial therapy with conventional synthetic DMARDs' and 'Efficacy of switching strategy' above.)
The best data illustrating that switching to a second TNF inhibitor is effective after discontinuation of a first are from a randomized trial of golimumab as the second agent, although there is less clinical experience with golimumab than with several of the other TNF inhibitors [13]. In this trial, 461 patients with active RA despite prior use of at least one TNF inhibitor were randomly assigned to golimumab or placebo while continuing stable doses of baseline nonbiologic DMARDs (MTX, SSZ, and/or HCQ), glucocorticoids, and nonsteroidal antiinflammatory drugs (NSAIDs). The prior TNF inhibitor could have been discontinued due to either ineffectiveness (58 percent) or other factors unrelated to effectiveness, including intolerance or inaccessibility (53 percent); some discontinued the first TNF inhibitor for multiple reasons. Among those who had discontinued the first TNF inhibitor due to lack of effectiveness, patients in the combined golimumab group (who received either 50 or 100 mg every four weeks) were significantly more likely to achieve an ACR20 or higher response at week 14 (39 versus 18 percent). Patients who discontinued the first drug for reasons other than ineffectiveness were also more likely to achieve at least an ACR20 response with golimumab (34 versus 20 percent). There was no increase in serious adverse events in patients on golimumab compared with those on placebo. The results of this trial are consistent with findings from observational studies of other TNF inhibitors.
The relative benefits of switching to a second or a third TNF inhibitor (etanercept, infliximab, or adalimumab) were examined in a prospective observational study of 373 patients in the South Swedish Arthritis Treatment Group registry [16]. After three months of treatment, patients receiving their first compared with their second anti-TNF agent and patients receiving their second compared with their third anti-TNF agent were more likely to achieve an ACR20 (61 versus 51 versus 35 percent, respectively) or ACR50 response (37 versus 27 versus 18 percent, respectively). EULAR overall response rates in the three groups were 76 versus 71 versus 58 percent, respectively.
Thus, because the likelihood of a response decreases with second switch of TNF inhibitors, we suggest a change to a non-TNF inhibitor bDMARD or tsDMARD (JAK inhibitor) after the failure of two such agents.
In patients who discontinue the second TNF inhibitor, analyses of data from a large national patient registry in Great Britain suggested that discontinuation was primarily due to the same reason for which the first drug was stopped (inefficacy or toxicity) [15].
Resistant to standard therapies — There are several options for patients who are unable to take or who have an inadequate response to TNF inhibitors, abatacept, IL-6 inhibitors, rituximab, and JAK inhibitors. The choice between these further options depends upon patient preferences regarding route of administration, patient comorbidities, and clinician preference based upon degree of experience with a given agent. There have been no head-to-head comparisons of these therapies.
●Nonbiologic (traditional) DMARDs – Treatment options in this setting should include the use of nonbiologic DMARDs, such as LEF, or DMARD combinations, such as triple therapy or MTX plus LEF, if they were not used in the original sequence of drug regimens in a given patient. These agents may be tried before less effective biologic agents, such as anakinra, or before the other nonbiologic DMARDs used infrequently in practice, such as those mentioned below. (See "Treatment of rheumatoid arthritis in adults resistant to initial conventional synthetic (nonbiologic) DMARD therapy", section on 'Leflunomide'.)
●Anakinra – Anakinra, a recombinant human IL-1 receptor antagonist, differs from the native human protein by having an added N-terminal methionine, and it is not glycosylated. Anakinra, administered daily via subcutaneous injection, operates, like the naturally occurring inhibitor, through competitive binding to the IL-1 receptor. This agent can be used as monotherapy and in combination with MTX or other conventional nonbiologic DMARDs [128,129]; it has modest efficacy compared with placebo, but it appears less effective than other bDMARDs in RA [20]. Its effectiveness in patients who have failed to respond to MTX plus a TNF inhibitor has not been formally evaluated.
●Other nonbiologic traditional DMARDs – Other nonbiologic DMARDs that have less efficacy, greater toxicity, or both, compared with other available medications are listed below.
•Azathioprine – The role of azathioprine in RA is poorly defined, although it was approved in 1981 by the FDA for use in RA. Clinical trial findings suggest that azathioprine is superior to placebo [130-134], comparably effective to parenteral gold [135], and probably less effective than MTX [136].
Azathioprine may be utilized as monotherapy but also in combination regimens. There is a limited published experience reporting on therapy combining azathioprine with a bDMARD [137,138], but use of azathioprine in such combinations can help to prevent the development of anti-drug antibodies against monoclonal antibody therapies [139]. (See "Pharmacology and side effects of azathioprine when used in rheumatic diseases".)
•Gold – Parenteral gold, which is no longer generally available, had largely similar efficacy but greater toxicity compared with other traditional nonbiologic DMARDs used in RA, and oral gold has less efficacy than most of these other agents. The addition of parenteral gold injections to MTX had shown benefit in some patients with inadequate responses to MTX in whom other options were lacking or insufficient [140,141].
•Cyclosporine – The combination of cyclosporine and MTX has generally been more effective than either agent used individually [142-146]; cyclosporine has not been formally evaluated in patients resistant to bDMARD therapy. Tacrolimus, which like cyclosporine is a calcineurin inhibitor, has also been a subject of very limited study [147].
REEVALUATION AND MONITORING — Disease activity and the response to therapy should be regularly reassessed, along with monitoring for drug toxicities, every 4 to 12 weeks following a change in the treatment regimen until the patient is stable and until disease is under control [8,148-150]. Subsequently, assessments should not be less frequent than every three to six months. The approach to reevaluation and monitoring of these patients is described in more detail separately. (See "Treatment of rheumatoid arthritis in adults resistant to initial conventional synthetic (nonbiologic) DMARD therapy", section on 'Reevaluation and monitoring'.)
SYMPTOMATIC DRUG THERAPY — Antiinflammatory drugs are used as a supplement to disease-modifying antirheumatic drugs (DMARDs) under several conditions, including as bridging therapies until newly instituted DMARD regimens become effective, as adjuncts to DMARDs on a chronic basis, and for the management of disease exacerbations (flares) (see 'Antiinflammatory therapy' below and 'Drug therapy for flares' below). Analgesic effects of nonsteroidal antiinflammatory drugs (NSAIDs) or acetaminophen may also give additional relief. (See 'Analgesics' below.)
Antiinflammatory therapy — We use NSAIDs or systemic and/or intraarticular glucocorticoids when needed for ongoing control of inflammation while awaiting the response to modifications in DMARD therapy. Glucocorticoids can help to rapidly control inflammation and to improve symptoms. However, they should be used in the lowest dose required once such control is achieved, and they should be tapered and discontinued as soon as feasible. Some patients require ongoing therapy with low doses of glucocorticoids (eg, prednisone 5 to 7.5 mg daily) to maintain remission or a low level of disease activity. (See "Initial treatment of rheumatoid arthritis in adults", section on 'NSAIDs' and "Initial treatment of rheumatoid arthritis in adults", section on 'Glucocorticoids' and "Use of glucocorticoids in the treatment of rheumatoid arthritis".)
Drug therapy for flares — Rheumatoid arthritis (RA) has natural exacerbations (also known as flares) and reductions of continuing disease activity. It is important to distinguish a disease flare, characterized by symptoms and by physical and laboratory findings of increased inflammatory synovitis, from noninflammatory causes of local or generalized increased pain. The severity of the flare and background drug therapy influence the choice of therapies. The treatment of such flares is described in detail elsewhere. (See "Clinical manifestations of rheumatoid arthritis" and "Use of glucocorticoids in the treatment of rheumatoid arthritis" and "General principles and overview of management of rheumatoid arthritis in adults", section on 'Assessment and monitoring' and "General principles and overview of management of rheumatoid arthritis in adults", section on 'Drug therapy for flares'.)
Analgesics — In addition to the medications noted above, including NSAIDs, which also have analgesic effects, we use other analgesic medications, such as acetaminophen and/or the opioid tramadol, for additional pain relief, if required. However, we generally avoid the use of opioids, whenever possible, because pain can be controlled in most patients with RA by effective use of antiinflammatories and DMARDs that control the disease process. Patients without evidence of very significant joint injury who appear to require opioids for adequate pain relief should be evaluated for chronic widespread centralized pain and fibromyalgia or other comorbid causes of pain (eg, fracture, tumor, spinal disorders, neuropathy, or others). (See "Overview of chronic widespread (centralized) pain in the rheumatic diseases", section on 'Rheumatoid arthritis' and "Clinical manifestations and diagnosis of fibromyalgia in adults".)
DURATION OF THERAPY — Most patients with rheumatoid arthritis (RA) require sustained therapy and adjustments in their treatment regimen over months to years to achieve treatment goals. In the minority of patients who achieve a sustained clinical remission of greater than one year, we cautiously try to reduce nonbiologic and biologic disease-modifying antirheumatic drug (DMARD) doses or drug frequency, while maintaining close monitoring to facilitate recognition of any recurrence of disease activity. We generally avoid discontinuing all DMARD treatment.
Continued close monitoring is required in patients who discontinue or reduce any of their medications. The risk of disease recurrence in such patients is high, and flares of disease may occur even several years after stopping therapy [151-155]. Moreover, clinical remission can be difficult to achieve upon resumption of DMARD therapy. Our approach to tapering of DMARD therapy in patients in remission is discussed in more detail separately. (See "Treatment of rheumatoid arthritis in adults resistant to initial conventional synthetic (nonbiologic) DMARD therapy", section on 'Duration of therapy'.)
SPECIAL POPULATIONS
Pregnancy — Considerations relevant to the management of rheumatoid arthritis (RA) during pregnancy are reviewed separately. (See "Rheumatoid arthritis and pregnancy" and "Safety of rheumatic disease medication use during pregnancy and lactation".)
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Rheumatoid arthritis".)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topics (see "Patient education: Rheumatoid arthritis (The Basics)" and "Patient education: Disease-modifying antirheumatic drugs (DMARDs) (The Basics)")
●Beyond the Basics topics (see "Patient education: Rheumatoid arthritis treatment (Beyond the Basics)" and "Patient education: Rheumatoid arthritis symptoms and diagnosis (Beyond the Basics)")
SUMMARY AND RECOMMENDATIONS
●A number of important precautions should be taken before using disease-modifying antirheumatic drugs (DMARDs), including laboratory assessment (complete blood count, serum creatinine, aminotransferases, and other studies as indicated); evaluation of comorbidities; vaccinations; and screening for hepatitis C, hepatitis B, and latent tuberculosis (TB) infection. Precautions relevant to the use of each new agent being prescribed should be reviewed before initiating such therapy to confirm that all appropriate measures have been performed. (See 'Pretreatment interventions' above.)
●In patients with rheumatoid arthritis (RA), resistance to therapy with DMARDs is defined as failure to achieve remission or low disease activity within three to six months despite the use of maximally tolerated doses within the usual therapeutic range; a need for excessively high doses of glucocorticoids in addition to DMARDs; or the occurrence of recurrent flares of disease requiring multiple courses of glucocorticoids with doses in excess of those acceptable for chronic therapy, despite maximally tolerated or acceptable doses of the DMARDs being used. (See 'Definition of resistance to DMARDs' above.)
●For patients who do not respond adequately to therapy with methotrexate (MTX) plus an initial biologic (b) DMARD or targeted synthetic (ts) DMARD (ie, a Janus kinase [JAK] inhibitor) we suggest continuing the MTX and switching from the initial bDMARD or tsDMARD to a bDMARD or tsDMARD with a different mechanism of action or to triple therapy with the conventional synthetic (cs) DMARD combination of MTX plus hydroxychloroquine (HCQ) and sulfasalazine (SSZ) (Grade 2C). We prefer this approach over switching to another agent in the same class (eg, from one to a second tumor necrosis factor [TNF] inhibitor), particularly in patients with primary drug failure (ie, an inadequate initial response). Choice between these agents depends primarily upon patient preference; regulatory, insurance, and cost limitations; comorbidities; and clinician experience. (See 'Approach to therapy' above and 'Efficacy of switching strategy' above and 'Choice of therapy' above.)
●For patients whose treatment regimen has changed, reevaluation may be required up to every 4 to 12 weeks for the effectiveness of therapy and possibly more frequently for the monitoring of possible drug toxicity. We assess and monitor disease activity using a quantitative composite measure at each assessment. Further management is dependent upon disease response. (See 'Reevaluation and monitoring' above and "Treatment of rheumatoid arthritis in adults resistant to initial conventional synthetic (nonbiologic) DMARD therapy", section on 'Reevaluation and monitoring'.)
●For patients who do not respond adequately to the therapies above, including trials of MTX and trials of a TNF inhibitor, we suggest the use of abatacept, tocilizumab, a tsDMARD (ie, a JAK inhibitor), or rituximab, rather than the administration of a second TNF inhibitor (Grade 2B). Choice between these agents depends primarily upon patient preference; regulatory, insurance, and cost limitations; comorbidities; and clinician experience. (See 'TNF inhibitors' above and 'Efficacy of switching strategy' above and 'Choice of therapy' above.)
●The US Food and Drug Administration (FDA) has issued a warning for the class of JAK inhibitors in RA patients at increased risk for cardiovascular disease, venous thromboembolism (VTE), or malignancy, and if alternatives exist, JAK inhibitors would not be initial choice after bDMARD failures in this population. (See 'Choice of therapy' above and 'JAK inhibitors' above.)
●We continue nonbiologic and biologic DMARD therapy at reduced doses or dose frequency, if possible, for patients in remission, rather than discontinuing treatment with DMARDs. (See "Treatment of rheumatoid arthritis in adults resistant to initial conventional synthetic (nonbiologic) DMARD therapy", section on 'Duration of therapy'.)
●For patients experiencing a disease flare that is not controlled with nonsteroidal antiinflammatory drugs (NSAIDs), we suggest treatment with intraarticular or oral glucocorticoids rather than switching or continuing NSAIDs as the only additional agent (Grade 2A). We use antiinflammatory drug therapy, including NSAIDs or glucocorticoids, on a temporary basis to quickly achieve control of signs and symptoms of disease, and we then taper and withdraw these medications once a new DMARD regimen has taken effect. Some patients benefit from chronic low-dose glucocorticoid therapy (eg, prednisone 5 mg daily). (See 'Symptomatic drug therapy' above and "Treatment of rheumatoid arthritis in adults resistant to initial conventional synthetic (nonbiologic) DMARD therapy", section on 'Drug therapy for flares' and "General principles and overview of management of rheumatoid arthritis in adults", section on 'Drug therapy for flares'.)
ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Peter Schur, MD, who contributed to an earlier version of this topic review.
