Immune modulation of rheumatoid arthritis




The approval – several years ago – of the first tumour necrosis factor-α (TNF-α) inhibitor for the management of rheumatoid arthritis launched a new era in the therapeutics of rheumatology. Since then an almost cataclysmic discovery of new treatment targets and corresponding biologic agents ensued. Nowadays, the rheumatologist and the rheumatologic patient have the luxury of several immune modulators available to successfully treat the majority of patients with RA or other inflammatory arthritides and conditions.


In this review we focus on a discussion of the approved immune modulators/biologic agents available for the treatment of rheumatoid arthritis. We also present an overview of agents under development. For the immune modulators discussed, we describe their mechanism of action and summarise initial data and recent updates on efficacy and safety.


The era of investigation of immune modulators for the treatment of autoimmune rheumatic diseases began in the early to mid-1990s with the development of antagonists of tumour necrosis factor-α (TNF-α), followed rapidly by inhibitors of other inflammatory cytokines (interleukins (ILs)-1, −6, −12/23), of T-cell co-stimulation, and of B-cell growth and development. In recent years, attention has turned back to oral synthetic inhibitors although still focussing on novel targets such as intracellular kinases. Although many of the agents to be discussed here have been examined and approved in a variety of chronic inflammatory conditions (e.g., rheumatoid arthritis (RA), ankylosing spondylitis, psoriatic arthritis, Crohn’s disease), this review is limited to discussion of these agents for the treatment of RA.


Cytokine inhibitors


Antagonists of TNF-α


The development of antagonists of TNF-α initiated a landmark shift of investigational interest away from agents with poorly understood mechanisms of action towards therapies targeted to a key molecule involved in the pathogenesis of the disease. The hierarchical position of TNF-α in the inflammatory and destructive pathways of RA has been described extensively elsewhere . TNF-α antagonists were the first biologic therapies approved by the US Food and Drug Administration (FDA) and the European Medicines Agency (EMA) for the treatment of RA. The first to be approved were the soluble p75 TNF receptor, etanercept , and the mouse–human chimeric anti-TNF monoclonal antibody, infliximab , in 1998 and 1999, respectively. These were followed by approval of two fully human monoclonal antibodies, adalimumab in 2002 and golimumab in 2009, and later by the pegylated antibody, certolizumab pegol in 2009. All TNF antagonists are presumed to exert their effects by inhibiting the interaction of soluble TNF-α with its endogenous membrane-bound receptors. With limited exceptions, all five agents have been investigated and approved for the treatment of RA as monotherapy and in combination with non-biologic disease-modifying agents (DMARDs), in advanced as well as early RA. An exception is infliximab which is approved in combination with non-biologic DMARDs but not as monotherapy due to its apparent higher propensity to generate anti-chimeric antibodies which are suppressed by co-treatment with methotrexate (MTX).


Although they have never been compared in head-to-head clinical trials, the efficacy and safety of the five TNF antagonists are relatively comparable. For this reason and because data for at least some of these agents have been available now for over 10 years, they will be reviewed only briefly here and as a class rather than individually. The reader is referred to Table 1 for individual information regarding dosing and approved indications.



Table 1

Summary of FDA-approved biologics and current investigational drugs for the treatment of RA.































































































Agent Route Dosing Indication
TNF Inhibitors (TNFi)
Etanercept sq 50 mg weekly Moderate to severe RA
DMARD naïve RA
Infliximab iv 3 mg/kg at 0, 2, 6 weeks, then every 8 weeks thereafter; dose may be increased to 10 mg/kg; interval may be decreased to every 4 weeks. Administer with methotrexate. Moderate to severe RA
DMARD naive RA
Adalimumab sq 40mg every other week; interval can be decreased to every week Moderate to severe RA
DMARD naïve RA
Golimumab sq 50 mg every month Moderate to severe RA
DMARD naïve RA
Certolizumab pegol sq 400 mg at 2,4 weeks then 200 mg every 2 weeks thereafter. 400 mg every 4 weeks is an alternative for maintenance. Moderate to severe RA
DMARD naïve RA
Other approved biologics
Rituximab iv 1000mg at 0, 15 days then based on clinical symptoms (not more frequently than every 16 weeks) Moderate to severe RA, after failure of TNF antagonist
Abatacept iv/sq Iv route: Weight based ( <60Kg:500mg, 60-100Kg:750mg, >100Kg:1000mg) at 0,2,4 weeks and then every 4 weeks thereafter. Sq route: 125 mg weekly following a single iv infusion ( 10mg/kg) Moderate to severe RA
DMARD naïve RA
Tocilizumab iv 4mg/Kg every 4 weeks. May be increased to 8 mg/Kg (not to exceed 800mg per infusion). Moderate to severe RA, after failure of TNF antagonist
Investigational Drugs Mechanism of Action Completed study stage
CP-690,550
(tofacitinib)
po JAK 3 inhibition Phase II
R788 (fostamatinib) po Syk inhibition Phase II
Atacicept iv Anti- BLyS/APRIL Phase II
AIN457
LY2439821
iv IL-17 inhibition Phase I/II
Ocrelizumab
Ofatumumab
iv Anti-CD20 Monoclonal Antibodies Phase III
Phase I/II
Apilimod mesylate (STA-5326) po IL-12/IL-23 Inhibition Phase II trials
Pamapimod
VX-702
po P38 MAPK inhibition Phase II


Early trials in the 1990s demonstrated significant efficacy of etanercept and infliximab compared to placebo in reducing signs and symptoms of RA in patients with advanced disease . However, in head-to-head comparisons in patients with early disease, several of the TNF antagonists were observed to offer only a modest advantage over a conventional non-biologic DMARD, MTX, for suppressing progression of radiographic damage and were equivalent, but not superior, to MTX for reducing signs and symptoms . All of the five agents, when combined with MTX, are superior to MTX monotherapy in both advanced and early disease as measured by suppression of radiographic progression and reduction in signs and symptoms . In advanced disease, the efficacy of combination therapy compared to MTX monotherapy is robust since the patients studied are those who have already had inadequate responses to MTX . By contrast, in early RA patients who are naive to MTX treatment, the advantage of combination therapy of an anti-TNF agent plus MTX compared to MTX alone is considerably more modest . Thus, in clinical practice in patients with newly diagnosed RA, MTX monotherapy or monotherapy with an alternate non-biologic therapy (e.g., sulfasalazine and leflunomide) is frequently used as first line therapy, and a TNF antagonist is added subsequently only if the response to MTX (and/or other non-biologic DMARDs) is inadequate. Initiation of a TNF antagonist as a first-line treatment of RA may occasionally be justified in patients with multiple risk factors for severe disease (e.g., high autoantibody titres, baseline radiographic erosions, and high levels of inflammatory markers) . A distinct advantage of TNF inhibitors over conventional oral DMARDs is their rapidity of action, with clinical improvement evident in many patients within 1 week of initiation.


The most common adverse events (AEs) associated with administration of TNF antagonists are injection site and infusion reactions. Although occurring in up to 30% of patients in clinical trials, these reactions are generally easily managed with topical steroids (for injectable agents) and by slowing the infusion rate and/or by pre-infusion administration of antihistamines or steroids (for infliximab) . The development of antibodies against both chimeric and fully human monoclonal anti-TNF antibodies is reported and this can reduce their long-term efficacy and increase the likelihood of an injection/infusion reaction. This issue is most problematic with infliximab, presumably due to its murine component . Current FDA guidelines on infliximab advise concomitant treatment with MTX to reduce likelihood of development of anti-chimeric antibodies. If anti-TNF antibodies are suspected to be the reason for loss of efficacy, substituting an alternate anti-TNF agent is an option as currently available data do not suggest cross-reactivity between agents . In up to two-thirds of patients, TNF antagonists may also evoke production of antibodies to endogenous molecules, such as antinuclear, anti-double stranded DNA, anti-extractable nuclear and anti-cardiolipin antibodies . Nonetheless, the development of a lupus-like illness is uncommon and usually mild .


A more significant concern with treatment with any TNF antagonist is an increased risk for the development of opportunistic infections, including tuberculosis, histoplasmosis and others . It is imperative that patients undergo screening and treatment for latent and active tuberculosis prior to initiation of a TNF antagonist. This approach has significantly reduced the risk of reactivation of latent tuberculosis, although sporadic cases may still occur. Thus, a high level of suspicion for tuberculosis should be maintained especially in the early months of treatment with a TNF antagonist when most reactivation occurs, particularly in patients from areas with high endemic rates of tuberculosis. Whether TNF antagonists increase the risk for serious infections with common bacterial pathogens is still controversial . However, this uncertainty when coupled with the fact that patients with RA are at higher overall risk for infections due to their underlying disease argues for cautious monitoring. In general, treatment with TNF antagonists should not be initiated or continued during serious infections. TNF antagonists should specifically be avoided in patients with active hepatitis B ; by contrast, in active hepatitis C, limited data suggest that these agents may be safe . Response to prophylactic immunisation does not appear to be impaired by treatment with TNF antagonists . However, live vaccines should be avoided during, or should be administered several weeks prior to initiation of, therapy with a TNF antagonist. Current recommendations advise periodic pneumococcal vaccination and yearly influenza vaccination in patients treated with these agents .


TNF was originally named as such because of its tumouricidal activity against select tumours, and thus concern arose early on that inhibition of TNF could unmask or promote tumour growth. Because RA is itself a risk factor for malignancy, in particular for non-Hodgkin’s lymphoma, it is has been challenging to dissect effects of TNF antagonists on incidence rates of malignancies from the effect of RA itself and from non-biologic DMARDs which are often used concomitantly. Large observational cohort studies and meta-analyses of randomised clinical trials suggest that there may be a small increase in risk for non-Hodgkin’s lymphoma, and melanotic and non-melanotic skin cancers and lung cancer . A cautious approach consisting of avoiding TNF antagonists in patients with the aforementioned malignancies is warranted (with the exception of non-melanotic skin cancers as they are not associated with significant mortality).


TNF inhibitors may also increase the risk of developing new or worsening demyelinating disease and for exacerbating congestive heart failure and should be avoided in patients with these conditions – especially in patients with advanced congestive heart failure .




Interleukin-6 inhibition


IL-6 plays a central role in RA-related inflammation by virtue of its pleiotropy. IL-6 stimulates the differentiation of B cells to plasma cells, activates T cells, stimulates osteoclast activation/differentiation, and is a major inducer of the hepatic acute-phase reactant response and of anaemia of inflammation. In animal models of inflammatory arthritis, excess IL-6 sustains synovial inflammation . While IL-6 gene knock-out mice are protected from development of arthritis .


IL-6 exerts its effect by binding to its cognate receptor which, in turn, triggers interaction with gp130, followed by the transduction of intracellular signals that culminate in gene activation and the proinflammatory effects described above . IL-6 is abundantly present in the serum and synovium of subjects with RA and its levels in the joint and serum parallel the degree of systemic inflammation . For all these reasons, IL-6 became an attractive therapeutic target, which subsequently led to the development of several anti-IL-6 receptor antibodies, only one of which (tocilizumab) is currently approved by the FDA.


Tocilizumab (TCZ) is a humanised anti-IL-6 receptor antibody that contains complementarity determining regions of murine anti- human-IL-6 receptor antibody bound to human IgG1 which by blocking membrane-bound and soluble IL-6 receptors antagonises the action of endogenous IL-6 .


Tocilizumab (Actemra) is approved for the treatment of adults with moderate to severe RA who have had inadequate response to at least one anti-TNF agent; it is also approved for systemic onset juvenile arthritis (SOJA). It can be administered either as monotherapy or in combination with non-biologic DMARDs. The recommended starting dose in adults is 4 mg kg −1 intravenously monthly with the option to increase to 8 mg kg −1 as needed for clinical response; dosing for children with SOJA is 8 or 12 mg kg −1 , depending on age, intravenously every 2 weeks. Its approval was preceded by a series of randomised clinical trials where it was studied as monotherapy, or in combination with MTX (CHARISMA , OPTION , RADIATE , LITHE , and AMBITION ), or other non-biologic DMARDs (TOWARD ). In TOWARD only anti-TNF agent naive patients were enrolled, but in all other studies patients could have been previously exposed to TNF antagonists. However, failure to respond to anti-TNF therapy was a requisite inclusion criterion in only one study (RADIATE ).


The primary efficacy outcome in most of these trials was an ACR20 response. Overall tocilizumab in both doses administered as monotherapy or in combination with a DMARD had superior ACR20 responses compared to MTX or other DMARDs in all of the above trials. The dose of 8 mg kg −1 was not consistently statistically superior to 4 mg kg −1 in achieving efficacy outcomes. In the CHARISMA study , only the 8 mg kg −1 dose in combination with MTX was statistically more efficacious than placebo. Statistically significant differences against placebo were noted in measures of physical functionality (HAQ) and quality of life (SF36). Two studies demonstrated the ability of tocilizumab to slow progression of joint damage more effectively than MTX or other DMARDs .


Clinical response to tocilizumab was noted as early as 4 weeks and decreases in C-Reactive protein (CRP) as early as 2 weeks, following initiation of treatment. In most of the trials, American College of Rheumatology (ACR) response rates for tocilizumab continued to improve at the time of completion of the trial (usually 20–24 weeks). In addition, an improvement in haemoglobin levels was noted in the majority of the patients who met ACR response criteria.


Rates of serious infections in patients participating in tocilizumab clinical trials were similar to those reported in RCTs with TNF antagonists. In tocilizumab RCTs, the rate of serious infections in controls (14.4/100 patient-years (PY)) was similar to subjects on tocilizumab 4 mg kg −1 (13.6/100PY) or tocilizumab 8 mg kg −1 (14.5/100PY) . Pooled data from open-label extension studies did not show a further increase in the rates of serious infection with prolonged exposure to tocilizumab. The most common types of infection were cellulitis and pneumonia. Patients were screened and treated for latent or active TB prior to enrolment, and only seven tocilizumab-treated patients developed TB among a total of 9414 patient-years of observation.


Perforations in the lower gastrointestinal tract occurred at a rate of 1.9/1000 patient-years in patients treated with tocilizumab and were more common in patients with a history of diverticular disease. Whether there is a pathophysiologic link between diverticular perforation and IL-6 inhibition is not understood. However, the rates observed in tocilizumab-treated patients are higher than those in patients treated with TNF inhibitors or non-biologic DMARDs , suggesting a drug-specific mechanism.


Rare anaphylactic reactions to tocilizumab infusion were noted during the clinical trials usually early in treatment and most commonly with the lower dose of tocilizumab. Overall in pooled analyses, eight anaphylactic reactions were reported in 9414 patient-years of observation. Milder allergic reactions may be explained at least partially by the development of tocilizumab antibodies which have been reported in up to 25% of recipients, especially those without co-administration of MTX or DMARDs.


Transaminitis – manifested mainly as Alanine aminotransferase (ALT), and to a lesser degree as Aspartate aminotransferase (AST), elevations – was noted in Randomized controlled trial (RCT) and in pooled data analyses. Increases to more than 3 times the upper normal limits for ALT were noted in 9.5% of patients receiving Tocilizumab (TCZ). Similar increases were noted only in 3.7% of patients on MTX and in 1.1% of patients treated with other non-biologic DMARDs. The transaminase increases were dose related and parallelled the timing of the infusions in a saw-tooth pattern but were reversible upon dose reduction or discontinuation of tocilizumab. This phenomenon was accentuated in patients receiving MTX. Only in a minority of patients, liver function tests were elevated more than 5 times the upper limit or normal. Gradual dose-related increases were noted in bilirubin levels that were not related to concomitant intake of MTX. Frequent hepatic testing is recommended and, if transaminase elevation is observed, discontinuation or interruption of treatment until normalisation is recommended .


Tocilizumab treatment was also associated with elevations in lipids including total, HDL and LDL cholesterol levels. Increases in triglyceride levels were less frequent and were not associated with AEs such as pancreatitis. These changes occurred as early as 6 weeks after initiation of treatment, stabilised upon continuation of treatment and responded to therapy with statins . As cardiovascular events are known to be increased in frequency in RA patients compared to matched controls, this effect of TCZ could theoretically further increase CV risk in this population. It reassures however that there was no significant change in mean atherogenic indices (e.g., total cholesterol/HDL cholesterol and apolipoprotein A/apolipoprotein B) in TCZ-treated patients . Furthermore, insofar as atherosclerosis is recognised to be a low-grade inflammatory state and IL-6 as well as other proinflammatory cytokines are represented in atherosclerotic plaque , it has been argued that inhibition of these cytokines may reduce CV risk . Finally, RA patients in the pre-clinical stage have lower lipid levels than controls, presumably as a result of inflammation . Thus, increases in lipid levels following treatment with an anti-inflammatory agent may reflect a return of the patient to his/her pre-disease lipid status . A long-term clinical trial is now in progress to ascertain the risk of CV events in RA patients treated with TCZ .


Dose-dependent neutropenia was also observed in RCTs and open-label extensions of TCZ in RA. A decline in absolute neutrophil levels occurs early after initiation of treatment – as early as in 2 weeks – and afterwards remains stable. Only infrequently (4.1% of patients in pooled analysis) are neutrophils suppressed below 1000 μl −1 and more rarely (0.6%) below 500 μl −1 . Neutropenia was almost invariably reversible upon discontinuation of drug and frequently spontaneously resolved even when tocilizumab was continued. Fortunately, there was no association of neutropenia with the occurrence of infection. Persistent or recurrent neutropenia may warrant discontinuation of the agent .


In summary, tocilizumab is a potent and efficacious recent addition to the therapeutic armamentarium for patients who have failed treatment with a TNF inhibitor. Several additional IL-6 inhibitors are in development.


Novel cytokine targets


IL-12/IL-23 inhibition


IL-12 and IL-23 are proinflammatory molecules belonging to the IL-12 cytokine family along with IL-27 and IL-35. IL-12 is a heterodimer comprised of two subunits, p35 and p40. IL-23, also a heterodimer, is composed of the p40 subunit shared with IL-12 and of a second subunit, p19 .


IL-12 induces development of Th1 cells, which produce interferon-γ (IFN- γ). IL-23 plays an important role in the development and maintenance of Th17 cells , which, in turn, produce several inflammatory cytokines (IL-17A, IL-17F, IL-6, IL-1, and TNF-α). IL-23 also activates macrophages to produce IL-1, TNF-α, and IL-23 itself . In addition, IL-23 promotes osteoclastogenesis via induction of RANKL expression and increased levels of IL-23 have been associated with increased disease activity and radiographic changes in early RA .


An oral IL-12/23 inhibitor, apilimod mesylate (STA-5326), is currently being studied in RA treatment. It downregulates both IL-12 p35 and IL-12/IL-23 p40 at the transcriptional level by blocking the nuclear translocation of a transcription factor, c-Rel, and inhibiting the production of IL-12 and IL-23. A preliminary report of a phase IIa RCT in 29 patients with active RA despite baseline MTX, randomised to STA-5326 versus placebo, suggested lack of efficacy as the ACR20 response was no different in the drug-treated group than in the placebo arm. The most common AEs in the STA-5326 arm were gastrointestinal symptoms, dizziness and headache (ACR 2010 Abstract No.1095).


IL-17 inhibition


The IL-17 family of cytokines is composed of six members, IL-17A-F; of these, IL-17A has been the most studied in RA. IL-17 levels are significantly higher both in the synovial tissue and in fluid of RA patients when compared to osteoarthritis patients . IL-17A exerts inflammatory actions directly and synergistically with IL-1β and TNF-α, (1) including stimulating macrophages and synoviocytes to produce IL-6, IL-8, IL-1β, TNF-α and PGE2; (2) inducing MMP expression and (3) stimulating osteoclastogenesis .


Two phase I/II clinical trials using IL-17 inhibitors have been reported. The first one tested an IL-17A antibody ( AIN457 ) in RA patients refractory to MTX but the ACR20 response at week 16 failed to meet statistical significance when compared to placebo . The second study evaluated different doses of an IL-17 mAb ( LY2439821 ) versus placebo in RA patients refractory to DMARDs. At week 10, ACR20 responses were 74%, 70%, 90% and 78% in the 0.2 mgkg −1 , 0.6 mg kg −1 , 2 mg kg −1 , and all drug-combined groups, respectively compared to 55.6% in the placebo arm ( P < 0.05 vs. placebo for the 2 mg kg −1 group) .


In the AIN457 RA trial, overall AEs occurred in 81% in the AIN457 group versus 65% in the placebo arm. One severe AE (SAE), a laryngeal abscess, occurred in an AIN457-treated patient after study termination . In the LY2439821 study, the most common AEs in LY2439821-treated patients were headache, diarrhoea, leukopaenia and vertigo, with one SAE (skin ulcer requiring hospitalisation) occurring in the 0.6 mg kg −1 group .


Data are too preliminary for these IL-12/L-23 and IL-17 inhibitors but early studies do not suggest significant efficacy.




Interleukin-6 inhibition


IL-6 plays a central role in RA-related inflammation by virtue of its pleiotropy. IL-6 stimulates the differentiation of B cells to plasma cells, activates T cells, stimulates osteoclast activation/differentiation, and is a major inducer of the hepatic acute-phase reactant response and of anaemia of inflammation. In animal models of inflammatory arthritis, excess IL-6 sustains synovial inflammation . While IL-6 gene knock-out mice are protected from development of arthritis .


IL-6 exerts its effect by binding to its cognate receptor which, in turn, triggers interaction with gp130, followed by the transduction of intracellular signals that culminate in gene activation and the proinflammatory effects described above . IL-6 is abundantly present in the serum and synovium of subjects with RA and its levels in the joint and serum parallel the degree of systemic inflammation . For all these reasons, IL-6 became an attractive therapeutic target, which subsequently led to the development of several anti-IL-6 receptor antibodies, only one of which (tocilizumab) is currently approved by the FDA.


Tocilizumab (TCZ) is a humanised anti-IL-6 receptor antibody that contains complementarity determining regions of murine anti- human-IL-6 receptor antibody bound to human IgG1 which by blocking membrane-bound and soluble IL-6 receptors antagonises the action of endogenous IL-6 .


Tocilizumab (Actemra) is approved for the treatment of adults with moderate to severe RA who have had inadequate response to at least one anti-TNF agent; it is also approved for systemic onset juvenile arthritis (SOJA). It can be administered either as monotherapy or in combination with non-biologic DMARDs. The recommended starting dose in adults is 4 mg kg −1 intravenously monthly with the option to increase to 8 mg kg −1 as needed for clinical response; dosing for children with SOJA is 8 or 12 mg kg −1 , depending on age, intravenously every 2 weeks. Its approval was preceded by a series of randomised clinical trials where it was studied as monotherapy, or in combination with MTX (CHARISMA , OPTION , RADIATE , LITHE , and AMBITION ), or other non-biologic DMARDs (TOWARD ). In TOWARD only anti-TNF agent naive patients were enrolled, but in all other studies patients could have been previously exposed to TNF antagonists. However, failure to respond to anti-TNF therapy was a requisite inclusion criterion in only one study (RADIATE ).


The primary efficacy outcome in most of these trials was an ACR20 response. Overall tocilizumab in both doses administered as monotherapy or in combination with a DMARD had superior ACR20 responses compared to MTX or other DMARDs in all of the above trials. The dose of 8 mg kg −1 was not consistently statistically superior to 4 mg kg −1 in achieving efficacy outcomes. In the CHARISMA study , only the 8 mg kg −1 dose in combination with MTX was statistically more efficacious than placebo. Statistically significant differences against placebo were noted in measures of physical functionality (HAQ) and quality of life (SF36). Two studies demonstrated the ability of tocilizumab to slow progression of joint damage more effectively than MTX or other DMARDs .


Clinical response to tocilizumab was noted as early as 4 weeks and decreases in C-Reactive protein (CRP) as early as 2 weeks, following initiation of treatment. In most of the trials, American College of Rheumatology (ACR) response rates for tocilizumab continued to improve at the time of completion of the trial (usually 20–24 weeks). In addition, an improvement in haemoglobin levels was noted in the majority of the patients who met ACR response criteria.


Rates of serious infections in patients participating in tocilizumab clinical trials were similar to those reported in RCTs with TNF antagonists. In tocilizumab RCTs, the rate of serious infections in controls (14.4/100 patient-years (PY)) was similar to subjects on tocilizumab 4 mg kg −1 (13.6/100PY) or tocilizumab 8 mg kg −1 (14.5/100PY) . Pooled data from open-label extension studies did not show a further increase in the rates of serious infection with prolonged exposure to tocilizumab. The most common types of infection were cellulitis and pneumonia. Patients were screened and treated for latent or active TB prior to enrolment, and only seven tocilizumab-treated patients developed TB among a total of 9414 patient-years of observation.


Perforations in the lower gastrointestinal tract occurred at a rate of 1.9/1000 patient-years in patients treated with tocilizumab and were more common in patients with a history of diverticular disease. Whether there is a pathophysiologic link between diverticular perforation and IL-6 inhibition is not understood. However, the rates observed in tocilizumab-treated patients are higher than those in patients treated with TNF inhibitors or non-biologic DMARDs , suggesting a drug-specific mechanism.


Rare anaphylactic reactions to tocilizumab infusion were noted during the clinical trials usually early in treatment and most commonly with the lower dose of tocilizumab. Overall in pooled analyses, eight anaphylactic reactions were reported in 9414 patient-years of observation. Milder allergic reactions may be explained at least partially by the development of tocilizumab antibodies which have been reported in up to 25% of recipients, especially those without co-administration of MTX or DMARDs.


Transaminitis – manifested mainly as Alanine aminotransferase (ALT), and to a lesser degree as Aspartate aminotransferase (AST), elevations – was noted in Randomized controlled trial (RCT) and in pooled data analyses. Increases to more than 3 times the upper normal limits for ALT were noted in 9.5% of patients receiving Tocilizumab (TCZ). Similar increases were noted only in 3.7% of patients on MTX and in 1.1% of patients treated with other non-biologic DMARDs. The transaminase increases were dose related and parallelled the timing of the infusions in a saw-tooth pattern but were reversible upon dose reduction or discontinuation of tocilizumab. This phenomenon was accentuated in patients receiving MTX. Only in a minority of patients, liver function tests were elevated more than 5 times the upper limit or normal. Gradual dose-related increases were noted in bilirubin levels that were not related to concomitant intake of MTX. Frequent hepatic testing is recommended and, if transaminase elevation is observed, discontinuation or interruption of treatment until normalisation is recommended .


Tocilizumab treatment was also associated with elevations in lipids including total, HDL and LDL cholesterol levels. Increases in triglyceride levels were less frequent and were not associated with AEs such as pancreatitis. These changes occurred as early as 6 weeks after initiation of treatment, stabilised upon continuation of treatment and responded to therapy with statins . As cardiovascular events are known to be increased in frequency in RA patients compared to matched controls, this effect of TCZ could theoretically further increase CV risk in this population. It reassures however that there was no significant change in mean atherogenic indices (e.g., total cholesterol/HDL cholesterol and apolipoprotein A/apolipoprotein B) in TCZ-treated patients . Furthermore, insofar as atherosclerosis is recognised to be a low-grade inflammatory state and IL-6 as well as other proinflammatory cytokines are represented in atherosclerotic plaque , it has been argued that inhibition of these cytokines may reduce CV risk . Finally, RA patients in the pre-clinical stage have lower lipid levels than controls, presumably as a result of inflammation . Thus, increases in lipid levels following treatment with an anti-inflammatory agent may reflect a return of the patient to his/her pre-disease lipid status . A long-term clinical trial is now in progress to ascertain the risk of CV events in RA patients treated with TCZ .


Dose-dependent neutropenia was also observed in RCTs and open-label extensions of TCZ in RA. A decline in absolute neutrophil levels occurs early after initiation of treatment – as early as in 2 weeks – and afterwards remains stable. Only infrequently (4.1% of patients in pooled analysis) are neutrophils suppressed below 1000 μl −1 and more rarely (0.6%) below 500 μl −1 . Neutropenia was almost invariably reversible upon discontinuation of drug and frequently spontaneously resolved even when tocilizumab was continued. Fortunately, there was no association of neutropenia with the occurrence of infection. Persistent or recurrent neutropenia may warrant discontinuation of the agent .


In summary, tocilizumab is a potent and efficacious recent addition to the therapeutic armamentarium for patients who have failed treatment with a TNF inhibitor. Several additional IL-6 inhibitors are in development.


Novel cytokine targets


IL-12/IL-23 inhibition


IL-12 and IL-23 are proinflammatory molecules belonging to the IL-12 cytokine family along with IL-27 and IL-35. IL-12 is a heterodimer comprised of two subunits, p35 and p40. IL-23, also a heterodimer, is composed of the p40 subunit shared with IL-12 and of a second subunit, p19 .


IL-12 induces development of Th1 cells, which produce interferon-γ (IFN- γ). IL-23 plays an important role in the development and maintenance of Th17 cells , which, in turn, produce several inflammatory cytokines (IL-17A, IL-17F, IL-6, IL-1, and TNF-α). IL-23 also activates macrophages to produce IL-1, TNF-α, and IL-23 itself . In addition, IL-23 promotes osteoclastogenesis via induction of RANKL expression and increased levels of IL-23 have been associated with increased disease activity and radiographic changes in early RA .


An oral IL-12/23 inhibitor, apilimod mesylate (STA-5326), is currently being studied in RA treatment. It downregulates both IL-12 p35 and IL-12/IL-23 p40 at the transcriptional level by blocking the nuclear translocation of a transcription factor, c-Rel, and inhibiting the production of IL-12 and IL-23. A preliminary report of a phase IIa RCT in 29 patients with active RA despite baseline MTX, randomised to STA-5326 versus placebo, suggested lack of efficacy as the ACR20 response was no different in the drug-treated group than in the placebo arm. The most common AEs in the STA-5326 arm were gastrointestinal symptoms, dizziness and headache (ACR 2010 Abstract No.1095).


IL-17 inhibition


The IL-17 family of cytokines is composed of six members, IL-17A-F; of these, IL-17A has been the most studied in RA. IL-17 levels are significantly higher both in the synovial tissue and in fluid of RA patients when compared to osteoarthritis patients . IL-17A exerts inflammatory actions directly and synergistically with IL-1β and TNF-α, (1) including stimulating macrophages and synoviocytes to produce IL-6, IL-8, IL-1β, TNF-α and PGE2; (2) inducing MMP expression and (3) stimulating osteoclastogenesis .


Two phase I/II clinical trials using IL-17 inhibitors have been reported. The first one tested an IL-17A antibody ( AIN457 ) in RA patients refractory to MTX but the ACR20 response at week 16 failed to meet statistical significance when compared to placebo . The second study evaluated different doses of an IL-17 mAb ( LY2439821 ) versus placebo in RA patients refractory to DMARDs. At week 10, ACR20 responses were 74%, 70%, 90% and 78% in the 0.2 mgkg −1 , 0.6 mg kg −1 , 2 mg kg −1 , and all drug-combined groups, respectively compared to 55.6% in the placebo arm ( P < 0.05 vs. placebo for the 2 mg kg −1 group) .


In the AIN457 RA trial, overall AEs occurred in 81% in the AIN457 group versus 65% in the placebo arm. One severe AE (SAE), a laryngeal abscess, occurred in an AIN457-treated patient after study termination . In the LY2439821 study, the most common AEs in LY2439821-treated patients were headache, diarrhoea, leukopaenia and vertigo, with one SAE (skin ulcer requiring hospitalisation) occurring in the 0.6 mg kg −1 group .


Data are too preliminary for these IL-12/L-23 and IL-17 inhibitors but early studies do not suggest significant efficacy.




B cell Inhibition and depletion


B lymphocytes are extensively represented in the rheumatoid synovium and have a wide repertoire of functions that argue for an important role in the pathogenesis of RA. Not only are B cells effective antigen-presenting cells, they also provide co-stimulatory signalling critical for the function of CD4+ cells , and themselves secrete cytokines . As plasma cells they produce autoantibodies, rheumatoid factor (RF) and anti-cyclic citrullinated peptide (anti-CCP) antibodies, that are thought, with their cognate antigens, to form immune complexes and further promote inflammation and joint destruction .


Anti-CD20 antibodies


Rituximab (RTX) is a chimeric monoclonal anti-CD20 antibody that targets the cell-surface molecule, CD20, that is expressed during B-cell lineage development from pre-B cells to activated B cells and then disappears with maturation to plasma cells. Initially, FDA approved for the treatment of CD20 + B-cell non-Hodgkin’s lymphoma, RTX, induces the rapid depletion of circulating CD20 + B cells.


Rituximab’s efficacy in the treatment of RA has been demonstrated in a number of studies published since 2004. In the first large study of RTX in RA, using a dosage regimen of 1000 mg on days 1 and 15, the combination of RTX with MTX resulted in significantly better ACR 20, 50, and 70 responses compared to MTX alone . Subsequent studies were conducted in patients who had failed treatment with 1 or more TNF antagonists (REFLEX trial ), and in patients who had failed DMARDs including TNF antagonists (DANCER Study group) . Efficacy of Rituximab in combination with MTX was demonstrated against placebo + MTX after a single course and persisted with multiple courses of Rituximab in open-label extensions . Response was generally observed within the first month following treatment and persisted in many patients for more than 6 months . Efficacy was also demonstrated in terms of patient reported outcomes , including improvement in measures of physical function and quality of life . Similar findings comparing Rituximab plus MTX with MTX alone were reported in patients with active early disease who were naive to MTX. Slowing of progression of radiographic joint damage has also been demonstrated with Rituximab .


These studies led in 2006 to the approval by the FDA of Rituximab, in combination with MTX, for the treatment of RA patients who had inadequate response to TNF antagonists; peri-infusional corticosteroid administration in conjunction with Rituximab was also recommended. The suggested dose of RTX is 1000 mg intravenously twice separated by 2 weeks although accumulating evidence shows that different dosage schemas (such as 500 mg infusions 2 weeks apart) may be as efficacious (MIRROR study ). MTX co-administration seems to increase the efficacy of Rituximab .


Evidence suggesting safety for earlier introduction of Rituximab – e.g., in patients who have failed MTX or even in MTX naive patients – is accumulating (SERENE study – TNF naïve , MIRROR – TNF naïve , IMAGE study – MTX naive) .


Response to Rituximab should be evaluated using validated response criteria after the first 16 weeks and repeat dosing should not be attempted earlier than 24 weeks after the initial infusion according to published consensus and guidelines . The decision to re-treat is usually based on increasing clinical activity, as can be measured, for example, by worsening Disease activity score (DAS) scores. An approach suggesting that regular intervals of infusions every 24 weeks may be beneficial compared to infusions at disease relapse intervals has been associated with concerns about ‘overtreatment’. Thus, overall the right timing for retreatment is still under debate .


While it was initially thought that re-treatment should only be considered for patients who responded to the initial course of treatment , recent data show that a second course of treatment in initial non-responders can be efficacious . This may be explained by studies employing high-sensitivity flow cytometry showing that lack of clinical response is usually associated with incomplete B-cell depletion in the periphery and in the synovium . This observation challenges the initial hypothesis, that is, that residual or persistent disease activity after treatment with Rituximab is due to B-cell-independent mechanisms.


The most common safety issues related to Rituximab treatment in RCTs were infusion-related reactions and infections. Infusion reactions may be encountered in 30–45% of the patients but decrease in intensity after repeated infusions. Premedication including 100 mg of intravenous methylprednisolone may prevent the occurrence or alleviate the intensity of infusion reaction.


Progressive lowering of immunoglobulin (Ig) levels with repeated courses of treatment with Rituximab are a concern because of the potential for infection . In the largest longitudinal safety report of Rituximab to date, pooled AE data from 2587 Rituximab-treated patients (5013 patient-years of follow-up) were evaluated. The majority of the patients were followed for more than one year, almost half for more than 3 years and a small number of patients for more than 5 years. Most of the patients had received more than 2 courses of Rituximab and patients who had received up to 9 or 10 courses were included in data analysis, although statistical evaluation was limited to patients having received up to 5 courses of treatment due to power/sample size limitations. This study demonstrated good tolerability of repeated treatments with Rituximab, with no significant increase in rate of incident serious AEs or infections with repeated courses and prolonged time of follow-up. Repeated treatment with Rituximab was associated with a decrease in serum IgM levels in 23% of patients but this decrease was not associated with an increased risk of infection. However, in the 1% of the patients with sustained decreases in IgG – the immunoglobulin most highly associated with susceptibility to infection – a non-statistically significant trend towards an increased risk of infection was noted . Moreover, a study following 1303 RTX-treated patients totalling 1629 patient-years of observation reported that low IgG levels prior to initiation of treatment were a risk factor for severe infection .


Overall, even though not established by robust evidence, opportunistic infections – including tuberculosis and disseminated fungal infections – do not seem to be as significant a concern as in patients treated with TNF inhibitors, although it should not be overlooked that Rituximab is co-administered with corticosteroids, thus adding additional immunosuppression . Another reason to exercise caution in interpreting these data is that RA participants in Rituximab clinical trials underwent a “wash-out of TNF inhibitor” prior to enrolment, whereas in clinical practice this precaution may not be followed routinely.


The effect of Rituximab treatment on response to vaccinations has also been examined. As might be expected, Rituximab had no effect on responses to immunisations with T-cell-mediated antigens such as tetanus toxoid but did suppress responses to B-cell-mediated antigens such as pneumococcal polysaccharide. Responses to neoantigens were also significantly attenuated . These studies indicate the importance of completion of immunisations, if possible, prior to the initiation of Rituximab. While live vaccines should be avoided during treatment, immunisations with killed virus should still be administered even though response may be attenuated .


A new infectious concern arose from a report of progressive multifocal leucoencephalopathy (PML) in an RA patient receiving Rituximab monotherapy without prior TNF inhibitor treatment and without any known risk factors . Up to this point, the only two cases of PML in Rituximab-treated RA patients, both fatal, were in individuals with known risk factors (concomitant malignancy who had received chemotherapy). The FDA has issued a warning advising vigilance in patients treated with Rituximab who present with neurologic symptoms .


In summary, RTX is the first B-cell-directed therapy for the treatment of RA and an important drug in our armamentarium for RA and other autoimmune diseases. Questions remain as to the ideal timing of re-administration and long-term safety. As with all biologic agents, risk evaluation and mitigation are important and potential benefits that must be weighed against potential risks.


Anti-CD20 monoclonal antibodies in development


Ocrelizumab is a novel humanised anti-CD20 mAb reported to have enhanced antibody-dependent cell-mediated cytotoxicity (ADCC) and reduced complement-dependent cytotoxicity when compared to Rituximab . Two phase III RCTs in RA have been completed. The first one tested ocrelizumab versus placebo in MTX-refractory RA patients. ACR20 response rates were 35.7% (placebo), 56.9% (ocrelizumab 200 mg) and 54.5% (ocrelizumab 500 mg) at 24 weeks, and 27.6%, 58.3% and 62.1%, respectively, at 48 weeks ( P < 0.0001 vs. placebo for both doses at both timepoints) . The second RCT by Tak et al. (currently in press) compared the efficacy of ocrelizumab versus placebo in patients who had failed at least one TNF-α-inhibitor. ACR20 responses were 22.0% (placebo), 42.2% (ocrelizumab 200 mg) and 47.9% (ocrelizumab 500 mg) at 24 weeks, and 19.5%, 48.7% and 50.7% at 48 weeks, respectively ( P < 0.0001 vs. placebo for both doses at both timepoints). In both trials, serious infections were more common in the ocrelizumab-treated arms. Although efficacious, further development for this agent in RA has apparently been halted.


Ofatumumab , an anti-CD20 IgG1κ human mAb FDA-approved for the treatment of chronic lymphocytic leukaemia, is currently being studied in RA. It binds a novel membrane-proximal epitope and dissociates from its target at a slower rate than Rituximab, thus raising the potential for increased potency . In a phase I/II RCT in 225 patients with active RA despite DMARDs, ACR20 at 24 weeks was achieved in 40%, 49% and 44% for the 300 mg, 700 mg and 1000 mg doses, respectively, compared to 11% in the placebo group ( P < 0.001) . The most common AEs in the ofatumumab-treated groups were hypotension, hypertension, rash, urticaria, pyrexia, nausea, headache, pruritus, dyspnoea and upper respiratory tract infections. SAEs in the ofatumumab-treated groups included anaphylactoid-like reaction with urticaria and loss of consciousness, urticaria with periorobital oedema, cardiac ischaemia, bronchopneumonia, bronchospasm and breast CA. Infections occurred in 29% and 22–25% of the patients in the placebo and ofatumumab groups, respectively . Currently, phase III trials with ofatumumab are in process.


Inhibitors of B-cell growth factors


B-lymphocyte stimulator (BLyS) and A proliferation-inducing ligand (APRIL) are key regulators of B-cell maturation, proliferation and survival . Two TNF receptor-related receptors, transmembrane activator and CAML interactor (TACI) and B-cell maturation molecule (BCMA), bind both BLyS and APRIL . In murine models of inflammatory arthritis, TACI-Fc treatment substantially inhibited inflammation, bone and cartilage destruction and disease development . In addition, the presence of elevated levels of BLyS and APRIL in serum and synovial fluid of RA patients has suggested that they could be therapeutic targets, although the clinical benefit of decreasing the production of RA-related autoantibodies remains unclear.


Atacicept is a recombinant fusion protein containing the BLyS/APRIL-binding extracellular portion of the TACI molecule fused to the Fc portion of human IgG1; it neutralises BLyS and APRIL by competing for binding to their cognate receptor . In two phase II trials in RA , one in MTX- and one in anti-TNF-α -inadequate responders, the drug was tolerated well overall, but despite significant treatment-related decreases in Igs, RF and anti-CCP antibody levels, ACR20 responses were not significantly different from controls in either study .


In summary, while anti-CD20 B-cell depleting agents appear to be uniformly efficacious in RA (although only Rituximab is FDA approved), inhibition of B-cell growth factors has not proved to be efficacious despite similar depletion of B cells. A clear explanation for this apparent discordance has not been forthcoming. Current efforts to identify biomarkers that will predict responders to B-cell depleting therapy are in progress.

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Nov 11, 2017 | Posted by in RHEUMATOLOGY | Comments Off on Immune modulation of rheumatoid arthritis

Full access? Get Clinical Tree

Get Clinical Tree app for offline access