Rheumatoid arthritis (RA) can be the source of significant pain and functional limitation. The past 20 years have seen a transition in treatment goals away from mere pain management toward disease modification through the suppression of autoimmunity. Disease-modifying anti-rheumatic drugs, such as methotrexate and biologic agents, impair disease progression and joint destruction. However, despite these achievements, a substantial subset of RA patients does not respond to or cannot tolerate current treatments for RA. Scientific insight into the cellular pathways of inflammation has revealed new therapeutic targets for the treatment of autoimmune diseases like RA. Attention has focused on pathways mediated by Janus kinase (JAK), mitogen-activated protein kinase (MAPK), and spleen tyrosine kinase (Syk). This review article summarizes the evidence supporting the use of various kinase inhibitors, including the newly approved JAK inhibitor tofacitinib, in the treatment of RA.
Rheumatoid arthritis (RA) is a chronic and progressive autoimmune disease primarily causing inflammation of the synovial tissues of the joints and tendons. RA affects 0.5–1.0% of the population worldwide. Women are twice as likely to be afflicted with RA as men, with a median age of onset of 40–70 years. RA commonly manifests as symmetric joint pain and swelling stemming from synovial inflammation and destruction of underlying cartilage and bone. Untreated, RA can lead to irreversible joint damage and significantly impaired function. In addition to joint-related morbidity, patients with RA are at increased risk for gastrointestinal, respiratory, cardiovascular, infectious and hematologic diseases.
Current treatment of rheumatoid arthritis
Prior to the 1980s, there were few if any highly effective disease-modifying anti-rheumatic drugs (DMARDs) for the treatment of RA. Thus, management strategies focused largely on symptom control with the use of anti-inflammatory medications such as non-steroidal anti-inflammatory drugs (NSAIDs) and corticosteroids, often with intolerable long-term toxicity. Hydroxychloroquine, sulfasalazine and gold were of marginal value. In the late 1980s, methotrexate (MTX) became widely accepted as a highly effective DMARD and largely superseded these prior therapies. Over the years, MTX has repeatedly been shown to reduce the signs and symptoms of RA, slow structural disease progression and improve functional capacity in patients with RA. MTX remains an important first line DMARD, and often forms the foundation of an RA treatment protocol.[4, 5] In the late 1990s, a new class of DMARDs was introduced: biologicals. These macromolecular proteins are potent immunomodulatory agents that have revolutionized RA disease management, prognosis, and outcomes. Some biologics antagonize inflammatory cytokines like tumor necrosis factor alpha (TNF-α) (adalimumab, certolizumab, etanercept, golimumab and infliximab), interleukin-1 (IL-1) (anakinra) or IL-6 (tocilizumab). In addition, abatacept impairs T cell co-stimulation and rituximab depletes B cell numbers and antagonizes B cell function.
In most instances, traditional synthetic DMARDs, such as MTX, can be used safely and effectively in combination with a biologic agent. Indeed, this combination approach has repeatedly demonstrated reduced RA symptoms and joint damage in patients unresponsive to MTX alone.[6, 7] The current standard of care for RA is to initiate DMARD therapy soon after diagnosis and escalate treatment in an attempt to control inflammatory disease. Ideally, this will achieve disease remission by completely suppressing inflammatory joint disease, preventing progressive joint damage and improving function.
All biologics are either subcutaneously or intravenously administered. The most important adverse effect of biological therapies is immunosuppression, leading to an increased risk of infection. Despite their general safety and effectiveness, wider adoption of biologics has been limited by high drug costs which may affect medication adherence. Furthermore, up to 30% of patients show a primary or secondary non-response to biologic therapies, and an American College of Rheumatology (ACR) criteria response of ACR50 is achieved in approximately 50% or less of participants in most clinical trials of biologic agents.[9-12] Thus, despite all of the advances in disease management, patients with RA continue to experience relapses, unresponsiveness to therapies, unaffordable treatment costs and intolerable medication toxicities. These concerns have paved the way for the development of new, oral, small molecule DMARDs. The most widely studied and developed agents target various kinase pathways. Many kinases play a key role in immune activation and inflammation. Kinases and pharmacologic inhibitors of these pathways will be the topic of this review.
Through protein phosphorylation, kinases regulate multiple essential cellular activities, including signaling, metabolism, transcription and cycle progression. The human genome sequence has allowed for the identification of some 518 protein kinases. Conduction of signaling from the external environment to the cell interior and nucleus is crucial for immune and inflammatory responses and has clear implications in autoimmune disease (Fig. 1).
Tyrosine and seronine/threonine-specific kinases represent the largest families of kinases. Cytokines such as interleukins and interferons rely on the activation of receptor-associated tyrosine kinases such as the Janus kinases (JAKs). JAK molecules direct rapid downstream signaling and gene transcription via many mechanisms, including phosphorylation of signal transducer and activator of transcription (STAT) molecules. This pathway is discussed in greater detail later.
Src is a cytoplasmic kinase that is integral to T and B cell antigen receptors. Activation of Src leads to phosphorylation of associated immunoreceptor tyrosine-based activation motifs (ITAMs). Phosphorylated ITAMs serve as docking points for spleen tyrosine kinase (Syk), which allows for further downstream signaling and mediation of lymphocyte function. Syk is also a necessary component to integrin signaling, promoting cell–cell and cell–extracellular matrix interactions.
Mitogen-activated protein kinase (MAPK) pathways consist of a unit of three protein kinases functioning as a signaling cascade. There are at least six mammalian MAPK pathways, including the seronine/threonine p38 MAPK path, which is essential for signal conduction secondary to inflammation and environmental stressors. The MAP kinase signaling cascade impacts cytokine gene expression through downstream phosphorylation of additional kinases and transcription factors. Investigation into treatment options for rheumatoid arthritis has included inhibition of MAPK, JAK and Syk.
Mitogen-activated protein kinase inhibition
Mitogen-activated protein kinases (MAPK) were one of the first kinases targeted for the treatment of RA. Specifically, the p38 MAPK is an important intracellular signaling pathway for the production of TNF-α, IL-1β and IL-6, all of which have implications in RA.[15-17] Pamapimod and VX-702 were both developed to inhibit the alpha isoform of p38 MAPK, and each has shown favorable outcomes in animal models of RA.[15, 18] However, clinical trials have not consistently demonstrated statistically significant improvement in ACR response criteria when compared to placebo.[15, 16, 18] Interestingly both drugs showed a rapid and marked suppression in C-reactive protein (CRP) levels, but this was not sustained over time. This transient effect on CRP levels led to concerns that inhibition of p38 could trigger up-regulation of alternate inflammatory pathways.[16, 18] Most recently, a phase 2 clinical trial of a third p38 MAPK inhibitor, SCIO-469, again failed to demonstrate clinical response over placebo, but also showed a transient decrease in CRP levels. Further development of p38 MAPK inhibitors has been largely shelved by industry because of poor performance in RA clinical trials as compared to inhibitors of JAK and Syk. These agents block more proximally in the signaling cascade, which may explain their clinical success. In contrast, p38 MAPK may be too distal in the signaling pathway to be a relevant target.
The JAKs were initially discovered in the 1990s. The JAK family of tyrosine kinases consists of four members, JAK1, JAK2, JAK3 and tyrosine kinase-2 (TYK2). Although JAKs were initially coined ‘just another kinase’ due to their uncertain function, these molecules are now known to play a central role in cytokine signaling when coupled with STAT molecules. The JAK/STAT pathway is responsible for signal transduction of the type I and type II cytokine receptor family, which act as receptors of interferons, interleukins and colony-stimulating factors. Erythropoietin, thrombopoeitin, growth hormone, prolactin and leptin also associate with these receptors and rely on JAK signaling. Upon receptor ligation, a single JAK or combination of JAKs selectively associate with the receptor's cytoplasmic domain, leading to phosphorylation and activation of STATs. STATs are DNA binding proteins that, once phosphorylated, dimerize and translocate into the nucleus where they regulate transcription of STAT-dependent genes. JAK1 and JAK3 are mostly aligned with inflammation activation, whereas JAK2 plays a large role in hematopoiesis (Table 1). TYK2 is associated with immune response and may play a role in allergic inflammation. Interestingly, JAK3 associates with the common gamma chain-containing receptor that shares IL-2, IL-4, IL-7, IL-9, IL-15 and IL-21 as ligands. In the mid-1990s it was shown that mutations in JAK3 lead to severe combined immune deficiency (SCID) due to failure of signaling of the aforementioned cytokines and the subsequent failure of development of functional B, T and natural killer (NK) cells. This discovery provided great insight into the potential role of JAKs as immunomodulators (Table 2). As shown through recent drug development and clinical trials, JAK inhibition is now poised to expand the treatment options for RA.
Table 1. JAK family kinases and effects on cytokine signaling[21, 22, 49]
Tofacitinib is a small-molecule selective inhibitor of JAK1, JAK3 and to a lesser extent JAK2. Tofacitinib is the first kinase inhibitor to be approved for use in the United States for the treatment of moderately to severely active RA. However, in July 2013, the European Medicines Agency voted not to approve tofacitinib for use in RA. This decision stemmed largely from concerns that there was not a consistent enough reduction in disease activity and structural damage to outweigh the risks of serious infection, malignancy and laboratory abnormalities. Table 3 summarizes the phase 2 and phase 3 clinical trials of tofacitinib. A 6-week phase 2a trial studied patients with an inadequate response or intolerance of MTX, etanercept, infliximab or adalimumab. The study compared twice daily tofacitinib 5, 15 or 30 mg versus placebo. By week 6, tofacitinib at all three doses demonstrated statistically significantly improved ACR20, ACR50 and ACR70 response rates in comparison to placebo. A 24-week phase 2b trial then looked at five doses of tofacitinib (1, 3, 5, 10 and 15 mg) or adalimumab monotherapy versus placebo in patients with an inadequate response to DMARDs. At week 12, patients receiving adalimumab were switched to tofacitinib 5 mg twice daily for the remaining 12 weeks of the study. The trial demonstrated significantly improved ACR20, ACR50, ACR70, Health Assessment Questionnaire Disease Index (HAQ-DI), Disease Activity Score of 28 joints erythrocyte sedimentation rate (DAS28-ESR) and DAS28-CRP responses for tofacitinib in doses greater than or equal to 3 mg twice daily in comparison to placebo. Adalimumab was included as an active comparator and also to discern the safety of transitioning from adalimumab to tofacitinib. Patients who switched from adalimumab to tofacitinib had similar ACR20 response rates at week 24 to those treated with 5 mg twice a day at week 12. Furthermore, there appeared to be no complications of transitioning from a TNF-α inhibitor and tofacitinib. A subsequent 24-week phase 2b trial compared six doses of tofacitinib (20 mg once daily, 1 mg twice daily, 3 mg twice daily, 5 mg twice daily, 10 mg twice daily or 15 mg twice daily) versus placebo in patients taking background MTX with inadequate response. Tofacitinib doses greater than or equal to 3 mg twice daily again demonstrated statistically significant ACR20, ACR50, ACR70, HAQ-DI and DAS28-CPR response rates in comparison with placebo. Tofacitinib in combination with MTX was well tolerated, with an acceptable safety profile.
Table 3. Tofacitinib phase 2 and phase 3 clinical trials
ACR, American College of Rheumatology (response criteria); ADA, adalimumab; DAS28, disease activity score (28 joint count); DMARDs, disease modifying anti-rheumatic drugs; ETA, etanercept; HAQ-DI, health assessment questionnaire disability index; IFX, infliximab; MTX, methotrexate; MTXir, methotrexate inadequate response; PBO, placebo; RA, rheumatoid arthritis; TNFir, TNFα-antagonist inadequate response; TOFA, tofacitinib.
TOFA (3 doses) vs. PBO in RA patients with inadequate response to MTX, ETA, IFX, or ADA
TOFA 5 mg, 15 mg, and 30 mg BID superior to PBO by ACR20, ACR50 and ACR70
Several landmark phase 3 studies have recently demonstrated the efficacy of tofacitinib in the treatment of RA.[22, 28-31] In a 12-month study by van Vollenhoven et al., tofacitinib (5 and 10 mg twice daily) and adalimumab were compared to placebo in patients taking background MTX. Both tofacitinib and adalimumab in combination with MTX demonstrated statistically significant reduction in ACR20, ACR50, ACR70, HAQ-DI and DAS28-ESR responses in comparison to MTX alone. Importantly, although not a formal non-inferiority comparison study, tofacitinib appeared at least as efficacious as adalimumab in achieving response in patients failing MTX. A second 6-month study by Fleischmann et al. compared tofacitinib monotherapy to placebo in patients with an inadequate response to non-biologic or biologic DMARDs. Tofacitinib monotherapy achieved significant improvement in ACR20, ACR50, ACR70 and HAQ-DI results over placebo.
A third phase 3 trial was conducted to study tofacitinib versus placebo in patients on background MTX who had failed TNF-α inhibitor therapy. A third of these patients had failed two or more TNF-α inhibitors, yet tofacitinib still demonstrated significantly improved ACR20, ACR50, ACR70, DAS28 and HAQ-DI responses at 6 months, as compared to placebo. Another phase 3 trial was conducted by van der Heijde et al. to study the 24-month clinical and radiographic efficacy of tofacitinib versus placebo in patients on background MTX. At 12 months, this trial reported improved ACR20, ACR50 and ACR70 clinical responses in both the 5 and 10 mg doses, as well as improved HAQ-DI and DAS28-ESR in the 10 mg dose. Radiographic inhibition of structural change was only statistically improved in the tofacitinib 10 mg twice daily group, but not the group receiving tofacitinib 5 mg twice daily. However, a post hoc analysis of patients with poor prognostic factors and greater risk for joint destruction showed reduced structural damage for both tofacitinib 5 mg and 10 mg in comparison to placebo. Collectively, these studies demonstrate that tofacitinib provides clinical responses at 5 mg and 10 mg twice daily. Furthermore, results suggest that tofacitinib is effective as monotherapy or in combination with MTX, and it can be an option for patients having failed anti-TNF-α biologics. Tofacitinib also likely confers protection against progressive structural damage.
JAK/STAT signaling has pleiotropic effects in multiple pathways of cell growth, development and function. Accordingly, concerns have been raised about the safety of kinase inhibitors since their inception (Table 4). Across phase 2 and 3 trials, infectious illnesses were reported more frequently for tofacitinib than for placebo. Given the role of JAKs in immune function, this is not an entirely unexpected consequence of JAK inhibition. The most commonly reported infections included nasopharyngitis, upper respiratory infections and urinary tract infections. More severe infectious complications noted in the tofacitinib groups included pulmonary tuberculosis, tuberculous pleural effusion, lymph node tuberculosis, herpes zoster, pneumonias, Pneumocystis jiroveci pneumonia, esophageal candidiasis and cytomegalovirus infection. While one cannot draw too much of a conclusion based on limited head-to-head data, the infection rate of tofacitinib was comparable to that of biologic agents.
Table 4. Select adverse effects of kinase inhibitors
Significant adverse events
Data compiled from multiple studies referenced in main body of manuscript.
CA, cancer; HDL, high density lipoprotein; LDL, low density lipoprotein; NP, nasopharyngitis; PNA, pneumonia; TB, tuberculosis; URI, upper respiratory infection; UTI, urinary tract infection; VZV, varicella zoster virus.
Increased LDL and HDL
Elevated hepatic transaminases
Infections: NP, PNA, UTI, TB, VZV
Malignancies: lung CA, breast CA
No change in hemoglobin
No change in neutrophil count
No change in lymphocyte counts
Reduced platelet count
Increased LDL and HDL
Elevated hepatic transaminases
Infections: PNA, TB
Reduced neutrophil count
Reduced platelet count
Elevated hepatic transaminases
Infection: URI, PNA
Lymphoma, cervical CA
Malignancy risk has also been raised as another potential concern of the broad-sweeping inhibitory effects of tofacitinib. A long-term extension trial reported a case of lymphoma in a patient treated with tofacitinib, but the rate of lymphoproliferative disease was consistent with the rate seen in all patients with RA, including those treated with biologics. Similarly, occurrences of basal cell cancer, non-Hodgkin's lymphoma, stomach adenocarcinoma, breast mucinous adenocarcinoma and bone squamous cell carcinoma were reported in phase 3 trials. Further investigation has pooled phase 2 and 3 data to reflect 5651 patient-years of tofacitinib treatment. The most common malignancies reported were lung and breast cancer. Three cases of lymphoma were identified. The incidence for all malignancies (excluding non-melanoma skin cancer) is consistent with that of RA patients taking traditional small-molecule DMARDs and biologic agents.
Laboratory abnormalities were observed with tofacitinib treatment. Neutrophil levels decreased and studies showed suppressed hemoglobin levels (contrary to the rise in hemoglobin typically seen with biologic therapy). Since JAK2 is integral in the signaling of erythropoietin and colony stimulating factors, these cytopenias are felt to be a consequence of JAK2 inhibition. Notably, low density lipoprotein (LDL) and high DL (HDL) levels increased in tofacitinib study groups. While analyses of phase 3 trials and long-term open label extension studies have not demonstrated an increased risk of cardiovascular events compared to control RA patients, it may be too soon to conclude that these changes in lipid levels are inconsequential. Small, but statistically significant elevations in serum creatinine and infrequent increases in serum transaminase levels were also demonstrated. While long-term trials of tofacitinib are still ongoing, the available data regarding the safety profile of tofacitinib is encouraging and in keeping with the safety profile seen in biologic therapy.
Additional JAK inhibitors under investigation
Additional JAK inhibitors are under clinical investigation in RA (Table 5). Baricitinib (INCBO28050) is a selective inhibitor of JAK1 and JAK2. Baricitinib is similar to ruxolitinib in its inhibition of JAK1 and JAK2. Ruxolitinib was the first JAK inhibitor approved by the United States FDA in November of 2011 for treatment of myelofibrosis. Phase 2a trials for ruxolitinib in RA demonstrated significantly improved ACR response criteria, spurring on further investigation of baricitinib. In preclinical trials of baricitinib, inhibition of JAK1 and JAK2 interfered with signaling of inflammatory cytokines such as IL-6 and IL-23. Indeed, baricitinib was found to be effective in several rodent models of inflammatory arthritis without evidence of immunosuppression. The risk of bone marrow suppression expected with JAK1 and JAK2 inhibition was avoided by using periodic and incomplete inhibition. No change from baseline was observed in hemoglobin, neutrophils or lymphocyte counts in rodent models. Baricitinib is currently in phase 3 trials for RA.
Table 5. Clinical stages of development for various kinase inhibitors
JAK, Janus kinase.
Multiple trials ongoing
Multiple trials ongoing
Phase 2b expected 2014
No current trials
VX-509 is a selective JAK3 inhibitor currently in phase 2 and 3 investigation in the treatment of RA. Phase 2 studies compared 12 weeks of VX-509 monotherapy to placebo in patients who had failed a non-biologic DMARD. A significant response based on ACR20 and DAS28-CRP was seen with VX-509 dosed above 50 mg twice daily. Serious infections were noted, including a case of tuberculosis and pneumonia. As seen with tofacitinib and JAK3 inhibition, elevations in LDL, HDL and transaminases were reported. No effect was seen on hemoglobin, neutrophils or creatinine.
GLPG0634 is a selective JAK1 inhibitor. Conceptually, this might lead to anti-inflammatory effects of IL-6 reduction without the side-effect profile of JAK2 and JAK3 inhibition. A 4-week phase 2a trial was performed on 36 RA patients comparing GLPG0634 to placebo in those with inadequate response to MTX. A statistically significant response was seen in ACR20, DAS28 and CRP. Mild decreases in neutrophils and platelets counts were reported without effects on hemoglobin, LDL, creatinine or transaminases. A larger phase 2a study confirmed the efficacy previously seen as well as the safety profile. Phase 2b trials were scheduled to start in 2013.
Spleen tyrosine kinase and fostamatinib
Spleen tyrosine kinase (Syk) is another intracellular cytoplasmic tyrosine kinase. Syk has generated interest in the rheumatology community because it is downstream from the B cell receptor and Fc receptors, which have integral roles in immunoreceptor signaling for macrophages, neutrophils, mast cells and B cells.[40, 41] Additionally, Syk plays an important role in osteoclast development and bone remodeling, adding to its attraction as a target for inhibition in RA treatment. Syk is expressed in the RA synovial tissue and mediates TNF-α-induced production of cytokines such as IL-6 and metalloproteinase. Fostamatinib (R788) is a Syk inhibitor that showed superiority over placebo in attaining ACR20, ACR50, ACR70 and DAS28 responses in a phase 2a trial of patients failing MTX. In a second, 6-month phase 2 trial, fostamatinib continued to show efficacy over placebo in RA patients on background MTX, with statistically significant improvements in ACR20, ACR50, ACR70 and DAS28 responses at 100 mg twice daily and 150 mg daily dosing regimens. Side-effects included diarrhea, neutropenia and transaminitis. Hypertension was also noted as an adverse event, although patients responded to anti-hypertensive therapy with subsequent normalization of blood pressure. A subsequent phase 2 study of fostamatinib 100 mg twice daily in patients with an incomplete response to biologic therapy failed to demonstrate efficacy based on ACR response criteria. A difference was reported on CRP levels and magnetic resonance imaging synovitis score despite the lack of clinical response. However, patients in the fostamatinib group were noted to have more active disease and had failed more biologic therapies. A post hoc analysis demonstrated that patients with increased synovitis and who had failed only one biologic had significantly improved ACR20 responses. In a 24-week phase 2 trial of fostamatinib on background MTX, patient-reported outcomes of pain, disease activity, fatigue and physical function were improved in patients taking 100 mg twice daily. However, recent phase 3 clinical trials have reported mixed results, prompting the manufacturers of fostamatinib to put further drug development trials on hold.
The past 20 years have seen significant advances in the treatment of RA. MTX led the way providing not only symptom control, but also the ability to alter disease progression. Over the past 10 years, biologic DMARDs have expanded on this success and offered an alternative to those unable to achieve positive outcomes with traditional synthetic DMARDs alone.
Despite these triumphs there remains a need for safe and effective alternatives to the currently available RA therapies. Some patients who receive a biologic agent fail to achieve an adequate primary response and others experience a secondary loss of response. Further, biologics depend on an injectable route of administration, which is not appealing to a small subset of RA patients. Unfortunately, no advantage in drug cost has been achieved, as tofacitinib falls near the same price point as current biologic therapies. Novel small-molecule pharmacologic agents, such as JAK and Syk inhibitors, have the potential to become an alternative to biologic drugs for some patients with RA. Many clinical trials have demonstrated their efficacy in patients with inadequate disease control from traditional non-biologic and biologic DMARDs. Longer-term studies will be crucial to understand better the adverse effects and overall safety profile of these drugs.
None of the authors have any competing interests to declare.