|Home | About | Journals | Submit | Contact Us | Français|
As in previous years, the consensus group to consider the use of biological agents was constituted by rheumatologists from the universities of Erlangen, Leiden and Vienna in Europe in cooperation with other universities in the USA, Canada and Europe. Pharmaceutical industry support was obtained from a number of companies, but these companies had no part in the decisions regarding the specific programme or about the academic participants at this conference. These sponsors participated in the initial small breakout groups with emphasis on supplying factual information. The companies, on the other hand, had no part in the larger, final consensus group or in the final consensus statement.
The perspective of this consensus is from the treating physician's point of view.
The 160 rheumatologists and bioscientists who attended the consensus conference were chosen from a worldwide group of physicians and other scientists from 21 countries, with expertise in the use of biological agents for the treatment of rheumatic diseases. The number of attendees and participants was limited so that not everyone who might have been interested could be invited.
Based on the new data regarding TNF blocking agents, B cell‐specific agents & IL1rα, an update of the previous consensus statement is appropriate.1 The consensus statement is annotated to document the credibility of the data supporting it as much as possible. This annotation is that of Shekelle et al. and is described in an appendix.2 We have modified the Shekelle annotation by designating all abstracts as “Category D evidence”, whether they describe well‐controlled trials or not, based on a need to be able to describe details of the studies and results. As the number of possible references has become so large, reviews are sometimes used and, if they contain Category A references, will be referred to as Category A evidence. All participants reviewed relevant clinical published articles relating to tumour necrosis factor (TNF) and interleukin (IL) 1 blocking agents, and abatacept and rituximab. They were given a draft consensus statement and were asked to revise the document in small discussion groups; open discussion of the revisions led to a final document, representing this updated consensus statement.
Individual patients differ in the aggressiveness of their disease and its concomitant structural damage, the effect of their disease on their quality of life, and the symptoms and signs engendered by their disease. They also differ in their susceptibility to, and expression of, side effects to drugs. All these factors must be examined when considering biological treatment for a patient, as must the toxicity of previous and/or alternative disease‐modifying antirheumatic drug (DMARD) use.
As increasing evidence has accumulated for treating psoriatic arthritis (PsA) and ankylosing spondylitis (AS) with biological agents, efficacy and clinical use for these diseases will be treated separately from rheumatoid arthritis (RA). Adverse reactions, however, will remain combined for all indications. In general, in RA, when measuring response to therapy or when following patients over time, the American College of Rheumatology (ACR) response criteria (as a combined index) should not be used in a clinical practice setting to monitor individual response, although some validated measure of response (such as those which follow) should be employed (Category B evidence3). Validated quantitative measures such as Disease Activity Score (DAS), Simple Disease Activity Index (SDAI), Clinical Disease Activity Index (CDAI), Health Assessment Questionnaire disability index (HAQ‐DI), visual analogue scales (VAS) or Likert scales of global response or pain by the patient or global response by the physician, other validated measures of pain for individual patient care, joint tenderness and/or swelling counts, and laboratory data all may be used and may be the most appropriate measures for individual patients (Category B evidence3,4). The physician should evaluate a patient's response using the above measures to determine the patient's status and improvement.
For PsA, measures of response such as joint tenderness and swelling, global and pain response measures, functional indices and acute phase reactants have been used and appear responsive (Category A evidence5). For AS, measures such as the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), and the Bath Ankylosing Spondylitis Functional Index (BASFI) have been used in a clinical trial setting but have not been validated for the routine clinical practice setting. Measures such as joint tenderness and swelling, spinal motion, global and pain response measures, functional indices and acute phase reactants have been used and are validated (Category A evidence6,7,8,9,10).
The use of biological agents will require physicians experienced in the diagnosis, treatment and assessment of RA, PsA, AS and other rheumatic diseases. These physicians will need to make long‐term observations for efficacy and toxicity, including cohorts, registries and so on. Because these agents have toxicities, patients or their representatives should be provided with information about potential risks and benefits so that they may give informed consent for treatment.
TNF blocking agents differ in composition, precise mechanism of action, pharmacokinetics, biopharmaceutical properties and so on, but this document emphasises areas of commonality. Data that clearly have differentiated compounds will be discussed if such areas can be identified.
In most patients, TNF blockers are used in conjunction with another DMARD, usually methotrexate (MTX). TNFα blocking agents have also been used successfully with other DMARDs, including sulfasalazine and leflunomide (Category A/B evidence11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37). There is evidence that TNF blockers are effective for the treatment of RA in MTX‐naive patients (Category A evidence,12,13,14,15,16,17,21,22,26,27,31,34,36 Category D evidence19,24,25). TNF blocking agents can be used as the first DMARD in some patients (Category A evidence,12,13,14,15,16,17,18,22,26,27,28,31,33,36 Category D evidence19,26,35,36). Adalimumab and etanercept are both approved as monotherapy for RA. Infliximab is only approved for use with MTX in RA. However, observational data indicate that infliximab, too, is sometimes used as monotherapy (Category C evidence11,23,37). Evidence from several randomised controlled trials suggests that the combination of a TNF blocking agent and MTX yields superior results for RA when compared with monotherapy, particularly with respect to excellent clinical responses (ACR 70, European League Against Rheumatism remission) and radiological outcomes (Category A evidence).11,12,13,14,15,16,18,19,20,22,23,24,25,26,27,28,29,30,31,32,33,34,35,37
The three available anti‐TNF medications, etanercept, infliximab and adalimumab, have been approved globally for the treatment of psoriatic arthritis (PsA) based on demonstration of control of signs and symptoms of joint and skin disease, improvement of function (HAQ), quality of life (SF‐36), and inhibition of structural damage evidenced by x ray (Category A evidence40,41,42,43,44,45,46,47).
A comparator trial of methotrexate versus anti‐TNF therapy showed that the latter yielded superior results in joint and skin signs and symptoms, function and quality of life (Category B evidence47,48).
Preliminary data suggest that anti‐TNF therapy can benefit cardiovascular risk in PsA (Category A evidence49). Anti‐TNF therapy has been demonstrated to reduce healthcare resource utilisation (Category B evidence50), and improve employment status, time lost from work and productivity (Category B evidence44).
Adalimumab, etanercept and infliximab have been approved for the treatment of severe, active AS in Europe and North America (Category A evidence,6,7,8,9,10,51,52,53,54,55,56 Category D evidence57). In the clinical trials, efficacy of the TNF blocking agents was demonstrated as monotherapy as well as with concomitant second‐line agents including sulfasalazine or MTX (Category A evidence6,7,8,9,10,51,52,53,54,55,56). TNFα blocking agents maintain efficacy over 1–5 years in open‐label studies (Category B evidence55,56,58).
The approved doses of TNF blockers for treatment of AS are as follows: infliximab, 5 mg/kg intravenously every 6–8 weeks after induction; etanercept, subcutaneously 25 mg twice a week or 50 mg once a week; and adalimumab 40 mg subcutaneously every other week (Category A and B evidence6,7,8,9,10,38,51,52,53,54,55,56,57,58,59). No dose‐ranging studies have been done with any of these drugs. There is no evidence that one TNF blocking agent is more effective than any other for the treatment of musculoskeletal manifestions of AS. TNF blocking agents used for AS may reduce economic resource utilisation (Category D evidence38).
Infliximab has been approved to treat luminal and fistulising Crohn's disease and ulcerative colitis in the USA (Category A evidence;63,64 FDA Summary Basis of Approval). Infliximab is approved for ulcerative colitis and Crohn's disease in Europe (Category A evidence;65,66 European Medicines Agency Summary). Adalimumab is approved for Crohn's disease in the USA.
TNF blocking agents, when given up to the maximum approved dosing regimens for RA and juvenile idiopathic arthritis of the polyarthritic variety, are expected to lead to significant, documentable improvement in symptoms, signs and/or laboratory parameters within 12 weeks (Category A and B evidence6,7,8,9,10,12,13,14,15,16,17,18,19,20,22,24,25,26,27,28,29,31,32,33,34,35,36,37,56,80,81,82,83,84,85,86,87,88,89,90,91,92,93). There is no evidence that any one TNF blocking agent should be used before another one can be tried, just as there is no evidence that any TNF blocker is more effective than any other in RA (Category A and B evidence12,15,24,85,88,89,90,94,95). Patients have been switched from one TNF blocking agent to another, but no well‐controlled switch trials have been published (Category B and D evidence58,72,85,88,89,90,94,95,168,169,170,171). There is evidence that loss of response to a TNF blocking agent can occur and studies suggest that failure to respond to one TNF blocking agent does not preclude response to another (Category B and D evidence85,88,89,90,94,95).
Initial observational data suggest the possibility that primary non‐responding patients are less likely to respond to a second anti‐TNF agent. Patients who have not tolerated one TNF blocker may respond to a second but are also less likely to tolerate a second TNF blocking agent (Category B and D evidence85,90,94,95). The optimal therapy of patients not responding to TNF blockers remains to be determined.
Individually important responses including patient‐oriented measures (eg, HAQ‐DI, patient's global VAS, Medical Outcome Survey Short Form 36 (SF‐36)) or physical measures (for example, joint tenderness)) should be demonstrated within 12 weeks for RA (Category A evidence6,7,8,9,10,12,13,14,15,16,17,18,19,20,25,26,27,28,29,31,32,33,34,35,36,37,56,80,81,82,83,84). If such improvement occurs, treatment should be continued. If patients show no response to these agents, their continued use should be re‐evaluated. Observations with infliximab suggest that increasing the dose or reducing the dosing intervals may provide additional benefit in RA, as may the addition or substitution of other DMARDs (Category A evidence15,18,31,37). Raising the dose of etanercept does not seem to have an added benefit at 12 or 24 weeks in RA (Category D evidence89,90,91). However, because regression to the mean may occur, caution is needed when interpreting apparent improvements following dose escalation in practice (Category C evidence92).
There are data showing that TNF blocking agents slow and/or inhibit radiographic progression in RA, even in some patients without a clinical response (Category A evidence12,15,19,21,22,27,34,40,82,84,93,96). Although some RA patients without measurable clinical response have slowing of radiographic progression (Category A evidence34,93,97), the long‐term clinical implications of these changes still remain unknown.
As noted above, the three available anti‐TNF medications, etanercept, infliximab and adalimumab, have been approved globally for the treatment of psoriatic arthritis (PsA).40,41,42,43,44,45,46,47,48,49,50,51,98,99,100,101 In addition, statistically significant improvement of enthesitis and dactylitis has been demonstrated with infliximab.42,100,102,103 Also, improvement of fatigue has been noted with adalimumab.40 Improvement of signs and symptoms, function and quality of life occurs within 12 weeks.40,41,42,43,44,45,46,47,48,49,50,51 A systematic evidence‐based review of the various domains of PsA (joints, enthesitis, spondyloarthropathy, skin) and efficacy of their treatment has recently been published.99,100,101,103,104,105,106,107,108
Durability of clinical and radiographic data at 2 years in PsA has been demonstrated with etanercept and adalimumab.43,47 Similar durability at 1 year has been demonstrated with infliximab in PsA.44,46
Preliminary data suggest that one can achieve benefit for joint and skin signs and symptoms by switching to a different anti‐TNF medication, even if efficacy from a previous anti‐TNF agent was never achieved, was lost or if toxicity occurred.48,49,50,109,110
Anti‐TNF therapy is superior to methotrexate for joint and skin signs and symptoms, function and quality of life.48
An exploratory trial with anakinra in PsA failed to demonstrate significant clinical or synovial improvement as determined histologically and by MRI.111 An exploratory trial with efalizumab, approved for the treatment of psoriasis in the USA and Europe, failed to show statistically significant improvement in PsA.112 An exploratory trial with alefacept, approved for treatment of psoriasis in the USA, did show statistically significant improvement in joint and skin signs and symptoms in PsA.113
Safety and tolerability of anti‐TNF therapy in PsA is no different than an RA clinical trial. Psoriasis, particularly pustular psoriasis, can rarely occur when using anti‐TNF therapy.109
Generally, a reduction in signs and symptoms, and improvement in quality of life will be observed by 12–16 weeks in response to treatment with a TNF blocking agent.114 In clinical trials, improvement in signs and symptoms was assessed by patient‐reported outcomes (BASDAI, BASFI, patient global VAS, SF‐36), physical measures and laboratory parameters. The Assessment in Ankylosing Spondylitis working group has published recommendations for the use of TNF blocking agents in AS (Category A evidence54,115). If a 50% improvement or an absolute improvement of 2 points (on a 0–10 scale) of the BASDAI is not reached within 12 weeks, their continued use should be re‐evaluated (Category A and B evidence6,7,9,10,53,54,57,81,115).
Several studies have demonstrated that active inflammation of the sacroiliac joints or spine, as shown by MRI, is significantly reduced by all three TNF blocking agents (Category C evidence).41,42,43,57 Inhibition of radiographic progression, as measured by plain radiography, has not been evaluated in prospective randomised controlled trials in AS. In addition, changes in MRI findings have not yet been correlated with changes in plain radiographs in the spine or sacroiliac joints in patients with AS (Category C evidence).57,116,117
Clinical trial data with conventional DMARDs are few but suggest efficacy of sulfasalazine for the peripheral, but not spinal, manifestations of AS.54,115 In those who fail to respond to sulfasalazine, a TNF blocking agent should be considered next (Category D evidence115).
The IL‐RA antagonist anakinra has been investigated in two open‐label studies in AS with no clear efficacy (see section on IL1 receptor blockade). No data are available at the moment regarding efficacy of other biologicals such as rituximab or abatacept in AS.
An increased susceptibility to tuberculosis (TB) or reactivation of latent TB should be considered a characteristic of TNF blocking agents. The risk of TB is increased by the use of DMARD per se and is increased further by TNF blocking agents.121,122,123
The clinical picture of active TB may be atypical in these patients (for example, miliary or extrapulmonary presentations) as has been seen with other immunocompromised patients (Category C evidence118,119,123,124,125,126). There have been more reported cases of TB as a proportion of the total number of individuals treated in patients using infliximab and adalimumab than using etanercept (Category C evidence118,120,122,123,124,125,126,127,128,129). This may be due in part to differences in mechanism of action, biology or kinetics as compared to the soluble receptor (Category D evidence)123,126,128 but may also be, in part, due to the fact that populations treated with the various TNF blocking agents differ and the data come from registries and voluntary reporting systems. No head‐to‐head comparisons among TNF blocking agents have been carried out and thus no definitive data on comparisons between these agents are available regarding the incidence of reactivation of latent TB.
Screening of patients about to start TNF blocking agents has reduced the risk of activating TB for patients treated with these agents (Category B evidence118,119,123,129,130,131). Every patient should be evaluated for the possibility of latent TB, including a history that includes evaluation for the risk of latent TB (Category B evidence58,119,120,126,127,129,132). This history should include seeking a history of prior exposure, prior drug addiction or active drug addiction, HIV infection, birth or extended living in a region of high TB prevalence and a history of working or living in TB high‐risk settings such as jails, homeless shelters and drug rehabilitation centres (Category D evidence132). In addition, physical examination and screening tests such as skin tests and chest radiographs should be carried out before anti TNF‐therapy is initiated, according to local recommendations (Category B, C and D evidence118,119,122,123,129,130,131). Continued vigilance is required to prevent activation of latent tuberculosis or acquisition of new cases. The role of repeated or serial tuberculin skin testing during anti‐TNF therapy is unclear. In the USA, repeat or annual testing has not been recommended in this or other settings of immune suppression unless patients are at risk for ongoing TB exposure.132
In treating latent TB, the optimal timeframe between starting preventive therapy for latent TB infection and starting TNF blockade is unknown. Given the low numbers of bacilli present in latent TB infection, it is likely that waiting long time periods between initiating preventive TB therapy and TNF blockade are unnecessary. While there are no prospective trials assessing this question, observational data from Spain suggest that initiating isoniazid therapy 1 month prior to TNF blockade substantially decreases the risk of latent TB reactivation (Category C evidence122,129,130,131).
Treatment of TB in immunocompromised hosts is as effective as in non‐immunocompromised patients, suggesting that long periods of preventive treatment prior to starting TNF blocking agents may not be as necessary as previously thought. Prior to initiating preventive anti‐TB therapy in accordance with local guidelines, consulting with a TB expert should be considered.
Opportunistic infections have occurred in the setting of TNF blocking agent use (Category C evidence12,13,15,16,17,19,20,21,22,24,25,64,84,118,119,120,132,133,134,135,136,137,138,139,140,141). Particular vigilance is needed when considering those infections whose containment is macrophage/granuloma‐dependent, such as listeriosis, coccidiomycosis or histoplasmosis, including reactivation of latent histoplasmosis, (Category C and D evidence118,119,120,126,127,134,137,139,140,141,142) but the incidence of opportunistic infections is extremely low (Category C and D evidence126,127,129).
Serious bacterial infections have been observed in patients receiving TNF blocking agents at rates between 0.07 and 0.09/patient year compared with 0.01–0.06/patient year in controls using other DMARDs but not TNF blocking agents (Category C evidence118,119,120,143,144). Risk ratios of 1 to up to 3 were documented (Category B evidence144,145). High doses of TNF blocking agents may further increase the risk of serious infections (Category B evidence146). Other studies indicate that serious infections in certain sites are more common when using TNF blocking agents, such as the skin, soft tissues and joints (Category B evidence144,147).The possible contribution of corticosteroids to increasing the risk of infection should always be considered (Category C evidence148). The incidence of other infections (not designated as serious) may be increased when using TNF blocking agents (relative risk (RR) 2.3–3.0, 95% confidence interval (CI) 1.4 to 5.1) (Category C evidence143). The incidence of serious infections is higher when IL1rα and etanercept or abatacept with any of several biologicals are used in combination (3.9% in combination vs 1.0–1.6% in controls) (Category A evidence118,119,120,149,150). Therefore, a combination of biologicals is not recommended. TNF blocking agents should not be started or their discontinuation should be considered when serious infections and/or opportunistic infections occur, including septic arthritis, infected prostheses, acute abscess, osteomyelitis, sepsis, systemic fungal infections and listeriosis (Category C evidence; FDA12,13,15,16,17,19,20,21,22,24,25,33,64,84,119,120,126,134,135,137,139,140,141,142,145). Treatment with TNF blockers in such patients may be resumed if the infections have been treated adequately (Category D evidence; FDA 120,123,124,125,126,134,137,139,140,141,142,145).
TNF blocking agents do not significantly influence the development of protective antibodies after vaccination, although there is a small decrease in the prevalence of adequate protection and titre of response, especially in combination with methotrexate (Category B evidence151,152). Vaccination with live attenuated vaccines (eg, BCG, yellow fever) is not recommended (Category D evidence).
In placebo‐controlled trials, injection site reactions, most of which were mild to moderate but some of which resulted in drug discontinuation, were more common with subcutaneously administered TNF blocking agents than with placebo (Category B evidence10,12,13,14,15,16,17,18,24,25,31,32,33,61,62,82,83,96,119). One study indicates that human antichimeric antibodies (HACA) against infliximab was associated with decreased response and increased infusion reactions (Category C evidence153). Infusion reactions after infliximab and adalimumab are uncommon and are usually mild to moderate, but may, rarely, be serious (Category A evidence,12,15,21,24,25,31,62,82,83,96,119 Category B and C evidence18,22,63,84).
The incidence of lymphoma is increased in chronic inflammatory diseases such as RA. This increase is associated with high disease activity (Category C evidence155,156). The risk for lymphoma is increased two to fivefold in patients with RA as compared to the general population (especially non‐Hodgkin's lymphoma). A similar risk is seen in RA patients who have received anti‐TNF therapy (Category C evidence120,155,157). There are conflicting data about whether there is an increased risk for lymphoma and solid malignancies with anti‐TNF therapies for RA. Several large observational databases and a case control study did not demonstrate an increased incidence of solid tumours in patients receiving TNF blocking agents compared with matched controls, while two meta‐analyses of anti‐TNF therapies (on infliximab and adalimumab) reported a higher rate of solid malignancies including skin cancers (Category A and C evidence156,157,158,159,160,161). One population‐based study showed that the incidence of both melanoma and non‐melanoma skin cancer were slightly increased when TNF blocking agents were used (RR: 1.4–2.0; Category C evidence160). In patients at risk for malignancies (for example, smokers) or in patients with chronic obstructive pulmonary disease (COPD), there may be an increased risk of lung cancers. In a trial of patients with COPD assigned to infliximab versus placebo, nine developed lung cancers during the trial and another four lung cancers were found during open‐label follow‐up (Category A evidence162). In a study of Wegener's granulomatosis, the use of etanercept with cyclophosphamide was associated with six solid malignancies versus none in the cyclophosphamide–placebo group (Category A evidence163). The concomitant use of azathioprine with infliximab in adolescents has been associated with the occurrence of rare hepatosplenic lymphomas (FDA). There is limited information about the risk of developing a future malignancy in patients receiving anti‐TNF therapy who had a previous malignancy (Category D evidence164). Vigilance with respect to the occurrence of lymphomas and other malignancies, including recurrence of solid tumours, remains warranted in patients using these medications.
Rare instances of pancytopaenia and aplastic anaemia have been reported (Category A and C evidence18,31,33,82,119,165). If haematological adverse events occur, TNF blockers should be stopped and patients evaluated for evidence of other underlying disease or other causative medications (Category D evidence).
High‐dose infliximab (10 mg/kg) appears to be associated with an increased relative risk of worsening congestive heart failure (CHF) and mortality, particularly in RA patients with New York Heart Association class III–IV CHF (Category B and D evidence118,120,166,167). There is presently no substantive evidence that infliximab 5 mg/kg, or etanercept at 25 mg twice a week increases the incidence of CHF or CHF‐related mortality in patients with functional class I CHF (Category B and D evidence118,166,167). However, it should be noted that well‐controlled RA studies have excluded patients with complicating illnesses, including CHF, and RA per se appears to be associated with increased atherosclerotic cardiovascular disease (ASCVD) and ASCVD‐related mortality (Category C evidence118). One RA study using infliximab demonstrated an increase in total, HDL and LDL lipids (Category D evidence168). One cohort observational study in RA patients without overt CHF showed a possible decrease in myocardial infarction‐related mortality when using TNF blocking agents (Category D evidence3,169). Each patient's risk versus benefit should be carefully considered before TNF blocking agents are begun or continued (Category D evidence; FDA).
The long‐term safety of TNF blockers in patients with chronic viral hepatitis is not known. In Hepatitis C, observational studies and one controlled investigation revealed that etanercept did not have an effect on viral load and there was no increased incidence of adverse effects; in addition, there was evidence, in this study, that symptoms and liver function tests improved (Category C and D evidence167,172,173,174,175,176,177,178).
With respect to hepatitis B, patients treated with all three agents have experienced increased symptoms and worsening of viral load (Category C evidence,173 Category D evidence.177,179). Elevations have been observed in liver function tests with therapeutic strategies employing the use of adalimumab, infliximab and etanercept with ALT‐AST elevated in 3.5–4.9% and elevations of these liver enzymes >2X ULN in about 0.1–0.2%, although concomitant medications and other clinical circumstances confound the interpretation of this information (FDA; Category C evidence18,173,175,176,177,180,181). As a result, specific warnings regarding hepatitis B reactivation have been added to the US label by the FDA (FDA). Therefore TNF blockers should not be used in patients with known hepatitis B infection; in the event that hepatitis B infection is discovered during use of TNF blockers, prophylactic antiviral therapy can be employed (Category C evidence;58,173,177,179,182 Canadian Regulatory Authorities).
The follow‐up and monitoring for liver function test elevations should be governed by the patient's concomitant medications, conditions and patient‐related risk factors.
There is insufficient information at present to safely counsel the continuation of TNF blockers during pregnancy. Some women have become pregnant while being treated with TNF blocking therapy. A small pharmacovigilance study and a survey study comprising about 200 pregnancies in women being treated with TNF blockers187 revealed that the rates of normal live births, miscarriages and therapeutic terminations in patients taking TNF blockers did not differ from published rates for the normal population (Category D evidence183,184,185,186,188,189). Pre‐pregnancy planning is strongly recommended in patients who are either on TNF blockers or considering their use; the need for disease control with alternate methods should be the subject of these conversations. There are no data available on the subsequent safety of a pregnancy if the male partner is taking TNF blockers.
It is advised that women and physicians discuss the issue of TNF blocking therapy when pregnancy planning takes place or if pregnancy is discovered during ongoing TNF blocking therapy, and that this discussion is documented.
Using peak concentrations as a measure, only about 0.0001% of the peak plasma concentration of etanercept is found in the breast milk (Category D evidence190).
Antiphospholipid syndrome, syndromes resembling drug‐induced lupus and vasculitis have occurred in patients receiving TNF blocking agents, and treatment should be stopped if there is clinical evidence of a drug‐induced lupus‐like syndrome. These symptoms are highly likely to resolve on discontinuation of the TNF blocking agent (Category C and D evidence31,33,63,82,118,120,191,192,193,194,195,196,197,198,199). There is an increased incidence of several autoantibodies (for example, antinuclear antibody (ANA), anti‐double‐stranded DNA) after infliximab and it is probably not a class effect (category C evidence191,192,193,194,195,198,199). However, there is no evidence that patients with RA who had, or develop, a positive ANA, anticardiolipin antibodies and/or dsDNA are at significantly increased risk for the development of drug‐induced lupus (FDA; Category C and D evidence18,31,82,120,191,192,196,197,198,199).
The incidence of demyelinating‐like syndromes, optic neuritis, transverse myelitis, multiple sclerosis and Parkinson's disease is no greater than expected in the general population (Category C evidence31,82,120,165,200,201,202). However, rare cases of these syndromes have been reported, more often with etanercept than with infliximab, most but not all improving or remitting after the TNF blocker was withdrawn (Category C evidence82,120,165,200,201,202,203). There is insufficient evidence to suggest that TNFα blockers unmask latent disease, although this remains a possibility. These agents should be stopped if a demyelinating‐like disorder or optic neuritis occurs. Patients with a history of definite demyelinating disease or optic neuritis should not receive TNF blocking agents (Category D evidence).
Some cases of new‐onset psoriatic skin lesions or exacerbations of pre‐existing psoriasis have been reported in patients with RA who used TNF blocking agents. (Category C evidence109).
Safety and tolerability data with anti‐TNF medications in PsA have not demonstrated any adverse events that were significantly different from RA trials. However, because liver biopsy studies suggest that patients with psoriasis and PsA demonstrate a greater proclivity for hepatotoxicity with MTX therapy than RA patients (Category B evidence204), it is not known if the safety profile from RA trials is completely comparable with PsA.
The safety of TNF blockade is unknown or has not been established in the following situations:
Other areas where knowledge is lacking are highlighted in the consensus group's recommendations for areas most urgently requiring further research.
Among a number of potential areas requiring action and/or further research, the consensus group felt the following projects or directions were most important in each of three areas: registries, efficacy and toxicity.
TNF blocking agents have proved to be effective DMARDs and are a major advance in the treatment of RA, PsA, AS and juvenile idiopathic arthritis. Their use is expanding to other rheumatic diseases. Studies in selected areas of efficacy, toxicity and general use of TNF blocking agents are needed to help further define the most appropriate use of these agents. Further considerations when using TNF blocking agents in these diseases should balance efficacy, toxicity and cost issues, and recognise that data in subpopulations are still being acquired. It is hoped that this statement, which is based on the best evidence available at this time and is modified by expert opinion, will facilitate the optimal use of these agents for patients with RA.
Only one IL1 blocking agent (anakinra) has been approved and the discussion below refers to this agent.
Anakinra may be used for treatment of active RA, alone or with MTX, at a dose of 100 mg per day subcutaneously (Category A evidence205,206,207,208,209,210). In Europe, the anakinra label requires its use with methotrexate. Anakinra is recommended for the treatment of active RA after an adequate trial of another conventional DMARD, of which MTX is a common example (Category D evidence). The safety of anakinra has also been studied with other DMARDs (Category A evidence207,208).
The use of anakinra as the first DMARD for the treatment of RA should, at present, be limited because no trials in early RA have been published.
Anakinra has been studied for a variety of off‐label indications (see Appendix 2). It appears to be highly active in the auto‐inflammatory syndromes (also called “periodic fever syndromes”), such as Muckle–Wells syndrome, neonatal‐onset multisystem inflammatory disease (NOMID) and TNF receptor‐associated periodic syndrome (TRAPS). It also appears to be active in systemic‐onset juvenile arthritis and adult‐onset Still's disease. It has been used to treat osteoarthritis (Category D evidence213,214) and SLE (Category D evidence215,216).
Anakinra can lead to significant improvement in symptoms, signs and/or laboratory parameters within 16 weeks, and can slow the rate of radiographic progression (Category A evidence205,206,209). If improvement is not observed by 16 weeks, the continued use of anakinra should be re‐evaluated.
Trials of patients failing TNF blocking agents demonstrate mixed responses (Category C evidence210). Anakinra did not inhibit anti‐tetanus antibody response in a controlled trial (Category A evidence217). A dose‐related incidence of injection site reactions, affecting up to 70% of patients, has occurred with the use of anakinra. These reactions often do not require treatment and seem to moderate with continued use in some patients (Category A evidence205,206,207). There are no data to advise either termination or continuation of anakinra if a woman becomes pregnant.
Serious infections are increased in patients receiving anakinra, and its incidence is higher than in patients with RA using other DMARDs. This increased incidence was magnified by corticosteroid use (Category A evidence208). Anakinra should not be started or should be discontinued when serious infections occur (Category A evidence,12,36,61,96 Category D evidence34). Treatment with anakinra in such patients should only be resumed if the infection(s) has been adequately treated (Category D evidence). To date, there is no indication that anakinra is associated with an increased incidence of TB (Category D evidence).
In combination with etanercept, there was an increased rate of serious infections compared to either monotherapy, and no increase in efficacy. Therefore, the combination should not be used (Category A evidence150).
The safety of anakinra is unknown or has not been established in the following situations:
Other areas where knowledge is lacking are highlighted in the consensus group's recommendations for areas most urgently requiring further research.
Among a number of potential areas requiring action and/or further research, the consensus group felt the following projects or directions were most important in each of three areas: registries, efficacy and toxicity.
Anakinra is effective for the treatment of rheumatoid arthritis but its specific place (for example, before or after TNF blocking agents) in the rheumatological armamentarium is not yet defined. Publication of studies in selected areas of efficacy, toxicity and general use of anakinra is needed to help further define the most appropriate use of these agents. Further considerations when using anakinra in this disease must include cost issues and the recognition that data in subpopulations are still being acquired. It is hoped that this statement, which is based on the best evidence available at the time of its creation and is modified by expert opinion, will facilitate the optimal use of anakinra for patients with RA.
One agent which modulates T cell activation (abatacept) has been approved, and references are therefore to this product.
Abatacept is approved in North America for use alone or with background DMARDs for treatment of active RA. It is administered as intravenous infusions of up to 10 mg/kg (500 mg for weights less than 60 kg; 750 mg for weights of 60–100 kg and 1000 mg for weights over 100 kg) at 0, 2, 4 weeks then monthly (FDA product label218,219). However, it is not recommended for use with other biologicals (although there are no data regarding the use of rituximab in combination with abatacept). Abatacept is recommended for treatment of active RA after an adequate trial of MTX, another effective DMARD or TNFα blocking agent (Category A evidence218,219,220,221,222).
Abatacept may be substituted directly for the next dose of a TNF blocking agent, when switching is undertaken (Category C evidence223). Abatacept has been used with MTX and other DMARDs (Category A evidence149). (Pharmacoeconomics studies are being carried out (Category D evidence).)
Abatacept is indicated to decrease signs and symptoms (including major clinical response), slow and/or inhibit progression of structural damage and improve physical function in adult patients with moderate to severely active RA who have had an inadequate response to one or more DMARDs such as MTX or TNF blocking agents (Category A evidence,218,219,223 FDA label). Improvement occurs within 16 weeks, although additional improvement can occur for up to 2 years (Category D evidence).224,225 If clinical improvement occurs, treatment should be continued (Category D evidence);where no meaningful improvement occurs within 12–16 weeks the continued use of abatacept should be re‐evaluated.
There is evidence for efficacy in juvenile idiopathic arthritis (Category C evidence227). Studies in early RA, undifferentiated arthritis and systemic lupus erythematosus are ongoing (Category D evidence).
An increased incidence of serious infection has been observed when comparing abatacept to placebo (3.0% with abatacept vs 1.9% with placebo). In combination with other biological agents, the rate of serious infections is 4.4% (vs 1.5% in controls).228 The use of abatacept with other biologicals is not recommended (Category A evidence221,228). There are no data regarding the combination of abatacept and rituximab. Patients with COPD treated with abatacept had more adverse events than patients treated with placebo; therefore use in RA patients with COPD should be undertaken with caution (Category D evidence149).
Based on theoretical concerns, live vaccines should not be given with abatacept or within 3 months of using abatacept (Category D evidence).
A numerical imbalance was reported in the number of lung cancers observed in the placebo‐controlled trials (4 abatacept, 0 controls). An additional nine cases were reported during open‐label extension studies (cumulatively 13 cases/4134 patient years). The overall incidence of lung cancer in abatacept‐treated patients during these studies was not higher than expected, based on large RA cohorts, although it was higher than in the general population.
There has been one case of lymphoma occurring in double‐blind period with abatacept versus 0 in the placebo group and four additional cases in the open‐label extension (cumulatively 5/4134 patient years) (Category B evidence229). While this number is consistent with that expected from large RA cohorts, ongoing surveillance is necessary.
All patients in abatacept phase 3 trials were screened for TB and excluded if the screen was positive. The risk for activation of latent TB or for developing new TB when using abatacept is unknown. Until the risk is known, it is prudent to screen patients considered for abatacept therapy for TB according to local practice (Category D evidence).
The safety of abatacept is unknown or has not been established in the following situations:
Other areas where knowledge is lacking are highlighted in the consensus group's recommendations for areas most urgently requiring further research.
Among a number of potential areas requiring action and/or further research, the following projects or directions are important in each of three areas: registries, efficacy and toxicity:
Rituximab is a chimeric anti‐CD20 monoclonal antibody, which was approved in 1997 for treatment of indolent CD20, B cell non‐Hodgkin's lymphoma. More than 1000000 patient exposures (usually four infusions per patient) have been documented over 9 years in post‐marketing surveillance of these non‐rheumatic diseases. Recently, a consensus statement on the use of rituximab in patients with RA (Category D evidence248) has been published.
Rituximab has been approved by the FDA in the USA for the treatment of moderate‐to‐severe RA in patients who have had an inadequate response to TNF inhibitors (Category A and D evidence;230,231,232 FDA and European Medicines Agency label; Category C and D evidence233,234,235,236,237,238,239). Patients should have at least moderate disease activity despite MTX therapy. Rituximab is administered intravenously as two 1 g rituximab infusions (given with 100 mg methylprednisolone or equivalent) separated by an interval of 2 weeks; two 500 mg doses can also be used with little decrease in efficacy (Category A evidence233,234,235,236,237,238,239,240). It may also be used when TNF inhibitors are not suitable (Category D evidence232,241,242). In RA, it may be used alone or in combination with MTX (Category A and D evidence230,231,232,241,242,243,244). Appropriate supportive equipment should be available when rituximab is used in case of rare anaphylactoid reactions.
Current evidence on the efficacy of rituximab relates to rheumatoid factor (RF) positive patients (Category D evidence233,234). Divergent ACR responses were seen with rituximab in RF‐negative patients, in TNF non‐responders and in DMARD non‐responders (Category D evidence233,234,235). Patients who were both RF and anti‐CCP negative had a reduced response in TNF non‐responders (Category D evidence242,244). Since small numbers of RF‐negative patients were evaluated, the role of RF and anti‐CCP antibodies is not clear.
In clinical trials, rituximab results in significant improvement in signs and symptoms and/or laboratory measures by 8–16 weeks (Category D evidence230,231,232,233,234,235,236,237,238,241,242,243,244) in patients with an inadequate response to MTX who have failed conventional DMARDS or have one or more TNF inhibitors (Category A evidence240,245). Improvement has also been demonstrated in patient‐related outcomes such as HAQ‐DI, patient global VAS, fatigue, disability and quality of life. Evidence from randomised controlled trials show that the combination of rituximab with MTX yields superior clinical efficacy for RA when compared with monotherapy (Category A evidence233,234,235,240,245). The optimal treatment schedule is currently under investigation (Category D evidence230,233,234,235,242,244). Preliminary data have shown that repeat treatment courses are effective in previously responsive RA patients (Category D evidence;236 FDA). Most of the patients who received subsequent courses did so after 24 weeks after the previous course and none received repeated courses earlier than 16 weeks from the previous course (Category D evidence;236 FDA). No data are available on repeated treatment in patients who failed to respond to the initial course.
There are data indicating that rituximab can slow radiographic progression in patients who have had an inadequate response to one or more TNF inhibitors (Category D evidence234).
Rituximab has been used in primary Sjögren's syndrome, systemic lupus erythematosus, Wegener's granulomatosis, hepatitis C‐associated cryoglobulinaemia, antineutrophilic cytoplasmic antibodies (ANCA), associated vasculitis disorders other than Wegener's such as polyarteritis nodosa, dermatomyositis/polymyositis (Category C evidence), autoimmune haemolytic anaemia, idiopathic thrombocytopenia purpura, antiphospholipid syndrome and scleroderma (see Appendix 3).
The most frequent adverse events are infusion reactions, which are most common with the first infusion of each course (approximately 35%) and are reduced with the second infusion (approximately 10%). Intravenous corticosteroids were shown to reduce the incidence and severity of infusion reactions by about 30% without changing efficacy (Category A, C and D evidence230,231,233,234,235,240,241,242,244).
Recently, two cases of progressive multifocal leukoencephalopathy (PML) have been communicated in cases of patients with severe SLE who had received an extensive additional immunosuppressive therapy. Twenty‐three cases of PML in SLE cases not using rituximab have also been documented. Twenty‐seven cases of PML are known in the NHL literature. It is unclear whether this incidence reflects the overall PML occurrence in these heavily immunocompromised individuals. As of today, no case of PML has been reported in RA (FDA reports).
Serious infections, including bacterial infections, have been observed in patients receiving rituximab. Similar to the anti‐TNF agents and the other biologicals, a small increase in serious infections (not intracellular infections) in patients receiving rituximab 2×1000 mg compared to placebo has occurred: 4.7/100 versus 3.2/100 patient years in the DANCER study and 5.2/100 versus 3.7/100 patient years in the REFLEX trial (Category D evidence230,233,234,235,239,241).
Rituximab should not be given in the presence of serious or opportunistic infections (Category D evidence).
In general, patients who did not respond to TNF inhibitors will also have been pre‐screened for the presence of active or latent TB. In the RA clinical trials of rituximab in TNF inhibitor non‐responders, patients with active TB were excluded. Others were screened by chest x ray, but were not screened for latent TB by PPD testing. Moreover, there is no evidence of an increased frequency of TB in patients with lymphoma treated with rituximab (Category D evidence). Based on expert opinion, it is felt that at this time there is no evidence indicating the necessity to screen for TB prior to initiating rituximab in RA (Category D evidence).
Since decreased levels of immunoglobulin (Ig) M, A and G have been observed with rituximab, it may be useful to determine baseline Ig levels (Category D evidence,231,237,241,244). In clinical trials no increase in serious infections have been reported in the very few patients with reduced levels of IgM after rituximab therapy compared to their normal IgM levels previously (Category D evidence237,244). Further analyses are required to address this issue. B cell levels have been measured in clinical trials but their utilisation in routine practice has not been proven. Depletion of peripheral levels of the CD20+ B cell subpopulation was not found to be predictive of achieving or maintaining a clinical response in RA patients (Category D evidence238). This suggests that the timing of retreatment should be based on disease activity rather than repletion of peripheral B cell levels (Category D evidence).
The safety of TNF inhibitor use in patients (n=78) who did not respond to rituximab and who had B cell levels below normal demonstrated a numerical increase in serious infections (Category D evidence239). Further data are needed to address this important issue.
Since rituximab causes B cell depletion, it is recommended that any vaccinations required by the patient, such as those to prevent pneumonia and influenza, should be given before commencing this agent. While on therapy, appropriate vaccination (such as against influenza) should be given when indicated, although the success may be suboptimal.
Until further data are available, the use of live attenuated vaccines should only be given prior to the use of rituximab (Category D evidence).
Patients should be screened for hepatitis B and C before starting rituximab as hepatitis B reactivation has been reported in oncological practice (Category D evidence246).
Until further data are available, rituximab should not be used during pregnancy.
There is no evidence that rituximab is associated with an increased incidence of solid tumours in RA. Nevertheless, vigilance regarding the occurrence of solid malignancies remains warranted during treatment with rituximab (Category D evidence).
The consensus group felt the following projects or directions were most important in each of three areas: registry, efficacy and toxicity.
abs - abstract
ACR - American College of Rheumatology
ANA - antinuclear antibody, AS, ankylosing spondylitis
ASCVD - atherosclerotic cardiovascular disease
BASDAI - Bath Ankylosing Spondylitis Disease Activity Index
BASFI - Bath Ankylosing Spondylitis Functional Index
CDAI - Clinical Disease Activity Index
CHF - congestive heart failure
COPD - chronic obstructive pulmonary disease
DAS - Disease Activity Score
DMARD - disease‐modifying antirheumatic drug
FDA - Food and Drug Administration
HAQ‐DI - Health Assessment Questionnaire disability index
MTX - methotrexate
PPD - purified protein derivative
PsA - psoriatic arthritis
RA - rheumatoid arthritis
SDAI - Simple Disease Activity Index
TB - tuberculosis
VAS - visual analogue scale
Note: abstracts have not been considered in the above evidence scheme, as they are not complete and may change by the time the data are published, or they may not be published as full papers at all. Evidence from abstracts alone, therefore, is considered as Category D evidence and noted as “abs” until those data are published as a complete, peer‐reviewed paper.