Fostamatinib, an oral spleen tyrosine kinase inhibitor, in the treatment of rheumatoid arthritis: a meta‑analysis of randomized controlled trials

Sumit Kunwar1 · Ashok Raj Devkota2 · Dipesh K. C. Ghimire3

Received: 17 December 2015 / Accepted: 15 April 2016
© Springer-Verlag Berlin Heidelberg 2016

Abstract Fostamatinib is a selective inhibitor of spleen tyrosine kinase which has a role in the pathogenesis of RA. Multiple RCTs have been performed to study the effects of fostamatinib. The objective of this study was to perform a meta-analysis to analyze the efficacy and safety of fos- tamatinib in the management of RA. We searched PubMed, EMBASE and Cochrane CENTRAL through 11/9/15. Ran- dom effect model was used to estimate odds ratio (OR) and 95 % confidence interval. We measured outcomes with efficacy analysis using ACR20/50/70 response criteria and safety with adverse events. Five studies were included in the meta-analysis with total of 2105 patients including 1419 in fostamatinib group and 686 in placebo. Fostamatinib was effective in achieving ACR20, ACR50 and ACR70 responses compared to placebo (48 vs. 32.8 %, OR 1.86, 95 % CI 1.32–2.62, P 0.0004, I2 63 %; 26.4 vs. 12.5 %,
OR 2.50, 95 % CI 1.93–3.23, P < 0.00001, I2 0 % and 12.7
vs. 4.4 %, OR 3.00, 95 % CI 1.99–4.51, P < 0.00001, I2
0 %, respectively). Response to fostamatinib was rapid and significant effect on ACR20 response was seen by week 1 (OR 3.70, 95 % CI 2.33–5.87, P < 0.00001, I2 42 %).

Electronic supplementary material The online version of this article (doi:10.1007/s00296-016-3482-7) contains supplementary material, which is available to authorized users.

Safety analysis showed an increased risk of infection (OR 1.59, 95 % CI 1.2–2.11; P 0.001; I2 0 %), diarrhea (OR
3.54; 95 % CI 2.43–5.16; P < 0.00001; I2 2 %), hyperten-
sion (OR 2.55, 95 % CI 1.54–4.22, P 0.0003; I2 42 %)
and neutropenia (OR 5.68, 95 % CI 1.97–16.42, P 0.001, I2 35 %) and showed a trend toward the increase in ALT 3 times ULN (OR 1.76, 95 % CI 0.99–3.13; P 0.05;
I2 0 %). This meta-analysis concludes that fostamatinib has moderate effect in the treatment of RA with mostly mild- to-moderate adverse events and dose-dependent, transient neutropenia and hypertransaminasemia.

Keywords Rheumatoid arthritis · Fostamatinib · SYK inhibitor

Introduction

Rheumatoid arthritis (RA) is a chronic multisystem autoim- mune disease that affects 1 % of the population worldwide [1]. It causes inflammation leading to joint destruction and significant deformity. Initiation of effective therapy early in the disease improves outcome and prevents disease pro- gression. Currently available biologics are not effective enough to achieve remission in one-third of the patients

with RA [2, 3]. Several therapeutic options are being

* Sumit Kunwar [email protected]

1 Department of Medicine, Lakes Region General Hospital, University of New England, 80 Highland Street, Laconia, NH 03246, USA
2 Division of Hospital Medicine, Rhode Island Hospital, Alpert Medical School of Brown University, Providence, RI, USA
3 Department of Medicine, Interfaith Medical Center, Brooklyn, NY, USA

explored including inhibitors of interleukin (IL)-17 and cell signaling protein kinases like Janus kinase and spleen tyrosine kinase (Syk).
Syk is a non-receptor, intracellular protein tyrosine kinase which is present in cells of hematopoietic system including immune cells like B cells, mast cells, mac- rophages and neutrophils [4, 5]. It is involved in immu- noreceptor signaling and when activated leads to induc- tion of proinflammatory cytokines [6] and degranulation

of effector cells. Immunoreceptor tyrosine-based activa- tion motif (ITAM) is present in cytoplasmic domains of immune receptors which is phosphorylated upon recep- tor engagement (Fc and B cell receptors) [7–9]. Phos- phorylated ITAM then causes recruitment and activa- tion of Syk which in turn activates adaptor proteins, leading to a various gene and cytoskeletal arrangements [10]. Significantly higher levels of phosphorylated Syk expression were seen in synovial tissues in RA as compared to osteoarthritis [11]. Moreover, Fc recep- tor (FcR)-deficient mice were protected from collagen induced arthritis indicating a need of FcR [12] and pos- sibly emphasizing a role of Syk in inflammation and autoimmune diseases. Syk activation plays an important role in TNF alpha-induced cytokine and metallopro- teinase production and IL-6 as well as matrix metallo- proteinase (MMP)-3 protein release decreased in dose- dependent manner when treated with Syk inhibitor [11]. Therefore, Syk might be a therapeutic target for suppres- sion of inflammation and treatment of autoimmune dis- eases like RA [11, 13].
Fostamatinib (R788) is a prodrug, and when taken
orally it is hydrolyzed rapidly and completely to active metabolite R406 by intestinal alkaline phosphatase [14– 17]. A potent and selective inhibitor of Syk, R406, reduced inflammation causing slower progression of arthritis in an antibody-induced arthritis model when compared to con- trol [18]. Arthritic rats when treated with R788/R406 had decreased tissue swelling, joint space narrowing, bone erosion and accumulation of cartilage degradation prod- uct in serum [17]. Similarly, R788/R406 showed potent anti-inflammatory action in cell culture studies of human synoviocytes.
Human trials have shown mixed results in terms of efficacy of this drug in treatment of RA. Two phase II clinical trials (TASKi-1 and TASKi-2) showed signifi- cant clinical improvement when taken in combination with methotrexate or other disease-modifying anti-rheu- matic drugs (DMARDs) [19, 20]; however, a third phase II trial (TASKi-3) failed to show efficacy in patients who had no response to biologic agents [21]. OSKIRA 4, a phase IIb study, found a modest clinic efficacy of fos- tamatinib monotherapy compared to placebo but showed an inferior response when compared to adalimumab [22]. Two phase III trials (OSKIRA-1 and OSKIRA-3) showed statistically significant improvement in ACR20 response; however, it failed to show improvement on progression of structural joint damage [23, 24]. There- fore, to study the role of this novel therapeutic agent and its safety profile in patients with RA, we performed a meta-analysis of all published human randomized con- trolled trials (RCTs).

Methods

Data sources and search strategy

We performed meta-analysis per PRISMA guideline [25, 26]. We searched electronic databases of PubMed, EMBASE and CENTRAL with no language restriction from inception through November 9, 2015. We used search terms “Fostam- atinib” OR “Tyrosine Kinase Inhibitor” OR “Syk” AND “RA” OR “Rheumatoid Arthritis” AND “Randomly” OR “Rand- omized Controlled Trial” OR “RCT” OR “Controlled Clini- cal Trial” OR “Comparative Study.” Database search was per- formed by two investigators (S. K. and D. G.) independently.

Study selection

Initial screening of articles by titles and abstracts was done by two reviewers (S. K. and D. G.), and disagreement was resolved by consensus. A complete reading was performed on the screened articles, and studies meeting inclusion criteria were selected for final analysis. Additional thor- ough hand search was performed for relevant references. Our eligibility criteria included human placebo-controlled RCTs in adults ( 18 years of age) that evaluated effi- cacy (ACR20/50/70 response) and safety outcomes (any and serious adverse events) of fostamatinib in active RA patients. We included published trials only if study duration was at least 10 weeks and included 10 or more patients. Exclusion criteria were nonrandomized design, non-com- parative study, healthy volunteers, pediatric patients, ani- mal studies and abstracts without full-text publication.

Data extraction

Two investigators (A. D. and D. G.) extracted data from the selected studies in duplicate using standardized data extrac- tion table. Extracted data were further reviewed by third investigator (S. K.) for accuracy. Data were extracted on study characteristics (number of patients, study design, study location, doses and frequency, follow-up duration, inclusion/ exclusion criteria, primary and secondary endpoints); patient characteristics [age, sex, disease duration, disease sever- ity at baseline]; efficacy outcome measures (ACR20/50/70 response) and adverse events. The studies measured baseline severity with CRP, ESR, Disease Activity Score at 28 joints (DAS28), tender and swollen joint counts, Patient’s and Phy- sician’s Global Assessment of Disease Activity and Health Assessment Questionnaire-Disability Index (HAQ-DI). One study was placebo controlled till 24 weeks only, and all the participants received fostamatinib for the remaining of the study period (total study duration 52 weeks) [23]. We included data from the first 24 weeks only.

Rheumatol Int

Major outcomes

We pooled data from individual studies to analyze efficacy outcome (outcome measures: ACR20, ACR50 and ACR70 responses, ACR20 response at week 1, DAS28 reduc- tion to <2.6 and HAQ reduction of 0.22). We added data from all doses of fostamatinib to calculate the final effect on ACR responses. We also performed dose-specific sub- group analysis with a head-to-head comparison between 100 mg BID and 150 mg daily for efficacy outcome meas- ures (ACR20/50/70 and DAS28 <2.6). Two studies used 100 mg BID for initial 4 weeks before starting 150 mg dose which was continued for next 20 weeks [23, 24]. For safety outcomes, we analyzed composite (any and serious adverse events) and individual adverse events including infections, headache, diarrhea, neutropenia, increased ala- nine aminotransferase (ALT) and drug discontinuation due to adverse events.

Statistical analysis

Categorical variables were pooled as odds ratio (OR) with 95 % confidence interval (CI). We used crude events from each study to calculate OR with 95 % CI when appropri- ate. The P value <0.05 (2 tailed) was considered statisti- cally significant. We used Review Manager (RevMan 5.3) and STATA 14 for statistical analyses and performed the meta-analysis using random effect model (DerSimonian and Laird). I2 statistic was used to identify heterogeneity between studies, and value of more than 50 was used to indicate significant heterogeneity [27, 28]. We performed sensitivity analysis to identify the possible causes when significant heterogeneity was present. Publication bias was examined at the outcome level with Begg’s funnel plot, and asymmetry was tested with Egger’s test. Because of small number of studies, any test for publication bias will have low power to distinguish chance from real asymmetry [29– 31]. We assessed risk of bias in individual studies using Cochrane Collaboration’s tool [32]. GRADE approach was used to assess the levels of quality of the body of evidence for each individual efficacy outcome using within-study risk of bias, directness of evidence, heterogeneity, precision of effect estimates and risk of publication bias [33, 34].

Results

Description of included studies

We retrieved 184 citations from electronic databases and manual search as shown in Fig. 1. After removal of dupli- cates, 161 articles were screened and 26 full-text articles were reviewed for inclusion. Seven potential studies were

identified; however, two studies were further excluded as they did not meet inclusion criteria (duration was less than 10 weeks) [22, 35]. Five studies met eligibility criteria and were included in meta-analysis [19–21, 23, 24]. Excluded articles with reasons have been included in the supplemen- tary file.
There were 1419 (67.4 %) patients in fostamatinib group and 686 (32.6 %) in placebo with total of 2105 patients in five included studies. In total, 83.6 % (n 1761) were females (F) and 16.3 % (n 344) were males (M) with similar distribution between two groups [fostamatinib (F/M)—84 %/16 % and placebo (F/M)—83 %/17 %). Baseline characteristics in all studies were comparable in both groups. Table 1 describes the patient and the study characteristics.
All included patients met criteria for RA per Ameri- can College of Rheumatology. Active RA was defined by 6 or more swollen joints (28 joint count), 6 or more ten- der joints (28 joint count) and either CRP more than upper limit of normal (ULN) or ESR 28 mm/h (or more than ULN). Exclusion criteria were active or latent infection, presence of hepatitis B or C, history of cancer within pre- vious 5 years, uncontrolled hypertension, major medical conditions, ALT more than 1.2 times ULN, hemoglobin less than 10 mg/dl, platelet count less than 125,000/mm3 or renal failure. One study excluded patients if they were previously treated with biologic agents other than TNF-α antagonists [24]. Most of the studies allowed methotrexate (if taken at stable doses prior randomization), prednisone (or equivalent) at dose 10 mg or less per day and prior use of biologics with appropriate washout period before rand- omization. Two studies even allowed parenteral steroid for symptoms control, if needed with restrictions [23, 24]. One injection of steroid was permitted between weeks 4 and 16 (none between weeks 16 and 24), and nonresponse was imputed for 8 weeks following the injection [24].
A significant proportion of patients failed to complete
the studies with 23.2 % (n 329) and 32.4 % (n 222) in fostamatinib and control groups, respectively. Major reasons for withdrawal were lack of efficacy [10.7 % (n = 152) in fostamatinib group and 23 % (n = 158) in placebo] and adverse events [7 % (n = 99) in fostamatinib group and 3.8 % (n = 26) in placebo].
Efficacy outcome

Compared to placebo, fostamatinib was more effective in achieving ACR20 response [48 vs. 32.8 %; OR 1.86; 95 %
CI 1.32–2.62; P 0.0004; I2 63 %], ACR50 response
[26.4 vs. 12.5 %; 2.50; 1.93–3.23; P < 0.00001, I2 0 %]
and ACR70 response [12.7 vs. 4.4 %; 3.00; 1.99–4.51;
P < 0.00001; I2 0 %] (Fig. 2). Because of significant het- erogeneity observed in ACR20 response, we performed a

Fig. 1 Flow diagram of study selection

sensitivity analysis removing one study at a time. We found that Weinblatt et al. [20] study contributed to heterogene- ity. Sensitivity analysis performed after removing this

study showed similar effect with reduced heterogeneity [1.62; 1.29–2.04; P < 0.0001; I2 5 %]. Response to fostam- atinib was rapid, and significant effect on ACR20 response

Table 1 Characteristics of patients at baseline and selected studies

Ref. Duration Study design Dose Total

Agea, yrs Sex (n)

(Race)

RF positive

Previous

Prednisone

DAS28b CRP
b

Tender joint Swollen
b b

(n)

(F/M)

White

(n)

biologics

≤10 mg/day

(mg/L)

count

joint count

parallel-group study

daily

Ref. Reference; n number, D drug, P Placebo, F Female, M Male, RF rheumatoid factor, NA not available, DAS28 Disease Activity Score in 28 joints, CRP C-reactive protein, yrs years
a Values are in means and range unless specified
b Values are in mean (SD)
c Median

was seen by week 1 [23.6 vs. 7.7 %; 3.70; 2.33–5.87;
P < 0.00001; I2 42 %] [19, 20, 23, 24].
Fostamatinib also showed improvement on Disease Activity Scores (DAS28) and greater reduction in Health Assessment Questionnaire-Disability Index (HAQ-DI). A greater proportion of patients achieved DAS28 <2.6 and HAQ-DI reduction of 0.22 on fostamatinib compared to placebo [(15.5 vs. 5.8 %; 2.82; 1.80–4.41; P < 0.00001, I2
31 %) and (53 vs. 37 %; 1.98; 1.61–2.45; P < 0.00001; I2
0 %) [20, 23, 24], respectively]. Sensitivity analysis per- formed on DAS28 <2.6 after removing Weinblatt et al. [20] study showed similar effect with no heterogeneity [I2 0 %] (data not shown).
Data were pooled from three studies to perform dose- specific analysis (Fig. 3); 100 mg BID was more effec- tive than 150 mg daily in achieving ACR20, ACR50 and ACR70 responses [(1.33; 1.04–1.68; P 0.02; I2 0 %),
(1.56; 1.18–2.05; P 0.002; I2 0 %) and (2.46; 1.58–3.85;
P < 0.0001; I2 15 %), respectively] [20, 23, 24].

Safety outcome

Fostamatinib increased the risk of any adverse effect [72 vs. 58 %; 1.96; 1.41–2.71; P < 0.0001; I2 37 %] [21, 23,
24] and had a trend toward the increase in serious adverse effect [4.4 vs. 2.4 %; 1.80; 0.92–3.52; P 0.09; I2 0 %]
[19, 23, 24]. Except for serious infection (the commonest serious adverse event), we did not perform subgroup analy- sis for individual serious adverse events because of small number of occurrence.
Fostamatinib had no effect on nausea [6 vs. 4.7 %; 1.21; 0.71–2.08; P 0.48; I2 25 %] [19, 20, 23, 24], headache
[6.2 vs. 5.4 %; 1.12; 0.75–1.67; P 0.58, I2 0 %] and seri-
ous infection [1.4 vs. 1 %; 0.65; 0.21–1.98; P 0.45; I2
23 %] [19, 20, 23, 24]; however, it increased the risk of
any infection [25 vs. 14.7 %; 1.59; 1.2–2.11; P 0.001; I2
0 %] [21, 23, 24], diarrhea [16.7 vs. 5 %; 3.54; 2.43–5.16;
P < 0.00001; I2 2 %], hypertension [17 vs. 7.6 %; 2.55;
1.54–4.22; P 0.0003; I2 42 %] [19, 20, 23, 24], neutro-
penia [6.8 vs. 0.8 %; 5.68; 1.97–16.42; P 0.001; I2 35 %]
and showed a trend toward the increase in ALT 3 times ULN [4.2 vs. 2.3 %; 1.76; 0.99–3.13; P 0.05; I2 0 %]
[20, 21, 23, 24] (Fig. 4).
There were total of 4 deaths, 1 due to cardiopulmonary arrest in the treatment group [24] and 3 due to septic shock, diabetes/hyponatremia and pulmonary embolism, each in placebo [21, 23, 24]. Four neoplasms (renal cell cancer, bladder cancer, gastric and thyroid cancer) were reported by three studies in the treatment group and none in pla- cebo [19, 23, 24]. Because these neoplasms were identified within short period of time from exposure (within 12 or 24 weeks), it is less likely to have any causal relationship.

Risk of bias and quality of evidence

Risk of bias was assessed using Cochrane Collabora- tion’s tool. Incomplete outcome data due to failure to complete the study by a significant proportion of patients have increased the risk of bias in the individual studies. The studies were found to be of low to unclear risk of bias (Fig. 5). Quality of the evidence was found to be moder- ate for outcome measures ACR50, ACR70 and DAS28 <2.6 (quality level decreased due to high drop-out rate in indi- vidual studies) and low for ACR20 (high drop-out rate and significant heterogeneity).
We tested funnel plot asymmetry for efficacy outcomes using Egger’s test which did not show publication bias. The Egger’s bias coefficients were 0.43 (P 0.88); 0.67
(P 0.43); 1.1 (P 0.47); and 0.088 (P 0.97) for effi-
cacy outcomes ACR20, ACR50, ACR70 and DAS28 <2.6, respectively. Because of very small number of studies, Egger’s test might not have had sufficient power to detect publication bias.

Discussion

This meta-analysis showed that fostamatinib is superior to placebo in achieving ACR20/50/70 responses with sig- nificant improvement on Disease Activity Scores (DAS28
<2.6) and greater reduction in HAQ-DI with acceptable safety profile. The clinical response was rapid with sig- nificant improvement on ACR20 response by week 1. There was, however, a significant heterogeneity in ACR20 response at treatment completion. Weinblatt et al. [20] study showed a higher ACR20 response rate than other studies which might be due to higher proportion of His- panic population (around 50 %) and relatively small per- centage of patients with prior biologic use (around 15 %), contributing to heterogeneity. Hispanic patients have higher response to fostamatinib [19, 20], while patients with prior biologic failure have been found to have poor response and difficult to treat with fostamatinib [21, 24].
In a post hoc analysis, patient-reported outcomes (PRO) were compared between fostamatinib and placebo [36] which showed a significant improvement in health-related quality of life (HRQoL) outcomes measured by pain, physical function, fatigue and patient’s Global Assess- ment of Disease Activity. This was consistent with clinical improvement (ACR20) seen in primary analysis [20]. Also, Strand et al. [37] found a clinically meaningful improve- ment in physical function, fatigue and health-related qual- ity of life (HRQoL) even from a small benefit in clinical endpoint (ACR20/50/70). Interestingly, this improvement in PRO was incremental when achieving higher threshold

Fig. 2 Meta-analysis of fostamatinib versus placebo. CI confidence interval, df degree of freedom

Fig. 3 Meta-analysis of dose-specific comparison (100 mg BID vs. 150 mg daily). CI confidence interval, df degree of freedom

of clinical endpoint indicating a clinical as well as HRQoL improvements from achieving ACR response.
In a difficult to treat patient population who failed biologic agents, Genovese et al. [21] showed response to fostamatinib proportional to degree of synovitis (haz- ard ratios of fostamatinib vs. placebo for 25th and 75th percentiles were 0.97 and 1.95, respectively). It might indicate that one group of RA patients might have bet- ter result than the other depending upon baseline disease severity and a selective approach might improve efficacy. Interestingly, despite significant ACR20 response, Wein- blatt et al. [23] did not find an effect on progression of joint damage. One possibility might be a time lag in clini- cal and radiological responses. Future larger RCTs with longer study duration involving radiological assessment of the progression of joint damage might help differenti- ate disease remission from clinical improvement. Another interesting evaluation would be to compare fostamatinib to currently approved biologic agents. Taylor et al. in a recent phase IIb study found fostamatinib less effective than adalimumab; however, it was limited by a small sam- ple size.
Fostamatinib seemed to have an acceptable safety profile
with majority of mild-to-moderate adverse events. It caused a significant neutropenia which was only transient and nor- malized after drug interruption or dose reduction without further drop. Similarly, hypertransaminasemia was found in a small proportion of patients which also normalized after dose reduction. A significant increase in blood pressure

was seen which was likely due to off target effect prevent- ing vascular endothelial growth factor-mediated vasore- laxation [23]. Hypertension occurred early on the treatment and mostly responded to either dose reduction or addition/ changing of anti-hypertensive medications. Fostamatinib- associated increase in blood pressure was more common in patients with preexisting hypertension. OSKIRA-ABPM showed an increase in 24-h ambulatory blood pressure (systolic and diastolic) which was reversible on drug dis- continuation [35].
The additional advantages of fostamatinib are its oral dosing with good bioavailability [14] and compatibility with methotrexate without affecting its pharmacokinetics [38]. This meta-analysis points toward possible and excit- ing role of this novel protein kinase inhibitor for treatment of RA. However, the response of fostamatinib compared to placebo was only modest with ACR20 of 48 %, compared to 32.8 % with placebo. Similar responses were observed for ACR50 and ACR70.

Study limitation

The major limitations of this meta-analysis are limited number of studies, significant percentage of withdrawal in the individual studies, variable doses of fostamatinib and inability to assess long-term safety profile due to short duration of follow-up. Significant heterogeneity was observed in ACR20 which was further explored with sensi- tivity analysis.

Fig. 4 Meta-analysis of adverse events. CI confidence interval; df degree of freedom

Fig. 5 Risk of bias summary; , low risk of bias; , high risk of bias; empty box unclear risk of bias

Conclusion

Fostamatinib has shown moderate effect in treatment of active RA with increase in risk of infection, diarrhea, hypertension and dose-dependent transient neutropenia as well as hypertransaminasemia. Larger RCTs should further confirm these findings and explore the long-term safety of this interesting novel agent in RA patients.

Authors’ contribution SK had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. SK and AD were involved in study design. SK, AD and DG were involved in acquisition of data. SK was involved in statistical analysis and interpreted the data. SK, AD and DG were involved in manuscript preparation.

Compliance with ethical standards

Conflict of interest Sumit Kunwar, Ashok Raj Devkota and Dipesh
K. C. Ghimire declare that they have no conflict of interest.

Ethical approval This article does not contain any studies with ani- mals or human participants performed directly by any of its authors.

References

1. Rivera J, Colbert RA (2010) Healing the Syk through kinase inhibitors. N Engl J Med 363:1362–1364
2. Storage SS, Agrawal H, Furst DE (2010) Description of the effi- cacy and safety of three new biologics in the treatment of rheu- matoid arthritis. Korean J Intern Med 25:1–17
3. Mewar D, Wilson AG (2011) Treatment of rheumatoid arthri- tis with tumour necrosis factor inhibitors. Br J Pharmacol 162:785–791
4. Okamoto H, Kobayashi A (2011) Tyrosine kinases in rheumatoid arthritis. J Inflamm Lond Engl 8:21
5. Salgado E, Maneiro JR, Carmona L, Gomez-Reino JJ (2014) Safety profile of protein kinase inhibitors in rheumatoid arthri- tis: systematic review and meta-analysis. Ann Rheum Dis 73:871–882
6. Scheinecker C, Smolen JS (2011) Rheumatoid arthritis in 2010: from the gut to the joint. Nat Rev Rheumatol 7:73–75
7. Berton G, Mócsai A, Lowell CA (2005) Src and Syk kinases: key regulators of phagocytic cell activation. Trends Immunol 26:208–214
8. Turner M, Schweighoffer E, Colucci F et al (2000) Tyrosine kinase SYK: essential functions for immunoreceptor signalling. Immunol Today 21:148–154
9. Kyttaris VC, Tsokos GC (2007) Syk kinase as a treatment target for therapy in autoimmune diseases. Clin Immunol Orlando Fla 124:235–237
10. Kaur M, Singh M, Silakari O (2013) Inhibitors of switch kinase “spleen tyrosine kinase” in inflammation and immune-mediated disorders: a review. Eur J Med Chem 67:434–446
11. Cha H-S, Boyle DL, Inoue T et al (2006) A novel spleen tyrosine kinase inhibitor blocks c-Jun N-terminal kinase-mediated gene expression in synoviocytes. J Pharmacol Exp Ther 317:571–578
12. Kleinau S, Martinsson P, Heyman B (2000) Induction and sup- pression of collagen-induced arthritis is dependent on distinct fcgamma receptors. J Exp Med 191:1611–1616
13. Wong BR, Grossbard EB, Payan DG, Masuda ES (2004) Target- ing Syk as a treatment for allergic and autoimmune disorders. Expert Opin Investig Drugs 13:743–762
14. Baluom M, Grossbard EB, Mant T, Lau DTW (2013) Pharma- cokinetics of fostamatinib, a spleen tyrosine kinase (SYK) inhib- itor, in healthy human subjects following single and multiple oral dosing in three phase I studies. Br J Clin Pharmacol 76:78–88
15. Scott DL (2011) Role of spleen tyrosine kinase inhibitors in the management of rheumatoid arthritis. Drugs 71:1121–1132
16. Sweeny DJ, Li W, Grossbard E, Lau DTW (2010) Contribution of gut bacteria to the metabolism of the spleen tyrosine kinase (Syk) inhibitor R406 in cynomolgus monkey. Xenobiotica Fate Foreign Compd Biol Syst 40:415–423
17. Pine PR, Chang B, Schoettler N et al (2007) Inflammation and bone erosion are suppressed in models of rheumatoid arthritis following treatment with a novel Syk inhibitor. Clin Immunol Orlando Fla 124:244–257
18. Braselmann S, Taylor V, Zhao H et al (2006) R406, an orally available spleen tyrosine kinase inhibitor blocks fc receptor sign- aling and reduces immune complex-mediated inflammation. J Pharmacol Exp Ther 319:998–1008
19. Weinblatt ME, Kavanaugh A, Burgos-Vargas R et al (2008) Treatment of rheumatoid arthritis with a Syk kinase inhibitor: a twelve-week, randomized, placebo-controlled trial. Arthritis Rheum 58:3309–3318
20. Weinblatt ME, Kavanaugh A, Genovese MC et al (2010) An oral spleen tyrosine kinase (Syk) inhibitor for rheumatoid arthritis. N Engl J Med 363:1303–1312

21. Genovese MC, Kavanaugh A, Weinblatt ME et al (2011) An oral Syk kinase inhibitor in the treatment of rheumatoid arthritis: a three-month randomized, placebo-controlled, phase II study in patients with active rheumatoid arthritis that did not respond to biologic agents. Arthritis Rheum 63:337–345
22. Taylor PC, Genovese MC, Greenwood M et al (2015) OSKIRA-4: a phase IIb randomised, placebo-controlled study of the efficacy and safety of fostamatinib monotherapy. Ann Rheum Dis 74:2123–2129
23. Weinblatt ME, Genovese MC, Ho M et al (2014) Effects of fos- tamatinib, an oral spleen tyrosine kinase inhibitor, in rheumatoid arthritis patients with an inadequate response to methotrexate: results from a phase III, multicenter, randomized, double-blind, placebo-controlled, parallel-group study. Arthritis Rheumatol Hoboken NJ 66:3255–3264
24. Genovese MC, van der Heijde DM, Keystone EC et al (2014) A phase III, multicenter, randomized, double-blind, placebo- controlled, parallel-group study of 2 dosing regimens of fostam- atinib in patients with rheumatoid arthritis with an inadequate response to a tumor necrosis factor-α antagonist. J Rheumatol 41:2120–2128
25. Moher D, Liberati A, Tetzlaff J et al (2009) Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Ann Intern Med 151(264–269):W64
26. Liberati A, Altman DG, Tetzlaff J et al (2009) The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. Ann Intern Med 151:W65–94
27. Higgins JPT, Thompson SG (2002) Quantifying heterogeneity in a meta-analysis. Stat Med 21:1539–1558. doi:10.1002/sim.1186
28. Higgins JPT, Thompson SG, Deeks JJ, Altman DG (2003) Meas- uring inconsistency in meta-analyses. BMJ 327:557–560
29. Tang JL, Liu JL (2000) Misleading funnel plot for detection of bias in meta-analysis. J Clin Epidemiol 53:477–484
30. Ioannidis JPA, Trikalinos TA (2007) The appropriateness of asymmetry tests for publication bias in meta-analyses:

a large survey. CMAJ Can Med Assoc J J Assoc Med Can 176:1091–1096
31. Terrin N, Schmid CH, Lau J (2005) In an empirical evaluation of the funnel plot, researchers could not visually identify publica- tion bias. J Clin Epidemiol 58:894–901
32. Higgins JPT, Altman DG, Gøtzsche PC et al (2011) The Cochrane Collaboration’s tool for assessing risk of bias in ran- domised trials. BMJ 343:d5928
33. Guyatt GH, Oxman AD, Vist GE et al (2008) GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. BMJ 336:924–926
34. Guyatt GH, Oxman AD, Kunz R et al (2008) What is “qual- ity of evidence” and why is it important to clinicians? BMJ 336:995–998
35. Kitas GD, Abreu G, Jedrychowicz-Rosiak K et al (2014) The effects of the spleen tyrosine kinase inhibitor fostamatinib on ambulatory blood pressure in patients with active rheumatoid arthritis: results of the OSKIRA-ABPM (ambulatory blood pres- sure monitoring) randomized trial. J Am Soc Hypertens JASH 8:780–790
36. Weinblatt ME, Kavanaugh A, Genovese MC et al (2013) Effects of fostamatinib (R788), an oral spleen tyrosine kinase inhibitor, on health-related quality of life in patients with active rheuma- toid arthritis: analyses of patient-reported outcomes from a ran- domized, double-blind, placebo-controlled trial. J Rheumatol 40:369–378
37. Strand V, Kosinski M, Gnanasakthy A et al (2014) Secukinumab treatment in rheumatoid arthritis is associated with incremental benefit in the clinical outcomes and HRQoL improvements that exceed minimally important thresholds. Health Qual Life Out- comes 12:31
38. Baluom M, Samara E, Grossbard EB, Lau DT-W (2011) Fostam- atinib, a Syk-kinase inhibitor, does not affect methotrexate phar- macokinetics in patients with rheumatoid arthritis. J Clin Phar- macol 51:1310–1318