Advertisement
Research Article

Systematic Review and Meta-Analysis of the Efficacy and Safety of Existing TNF Blocking Agents in Treatment of Rheumatoid Arthritis

  • Kalle J. Aaltonen,

    Affiliation: Faculties of Pharmacy and Medicine, University of Helsinki, Helsinki, Finland

    X
  • Liisa M. Virkki,

    Affiliation: Faculty of Medicine, University of Helsinki, Helsinki, Finland

    X
  • Antti Malmivaara,

    Affiliation: Centre for Health and Social Economics, National Institute for Health and Welfare (THL), Helsinki, Finland

    X
  • Yrjö T. Konttinen mail,

    yrjo.konttinen@helsinki.fi

    Affiliations: Faculty of Medicine, University of Helsinki, Helsinki, Finland, COXA Hospital for Joint Replacement, Tampere, Finland

    X
  • Dan C. Nordström,

    Affiliation: Helsinki University Central Hospital (HUCH), Helsinki, Finland

    X
  • Marja Blom

    Affiliation: Faculty of Pharmacy, University of Helsinki, Helsinki, Finland

    X
  • Published: January 17, 2012
  • DOI: 10.1371/journal.pone.0030275

Abstract

Background and Objectives

Five-tumour necrosis factor (TNF)-blockers (infliximab, etanercept, adalimumab, certolizumab pegol and golimumab) are available for treatment of rheumatoid arthritis. Only few clinical trials compare one TNF-blocker to another. Hence, a systematic review is required to indirectly compare the substances. The aim of our study is to estimate the efficacy and the safety of TNF-blockers in the treatment of rheumatoid arthritis (RA) and indirectly compare all five currently available blockers by combining the results from included randomized clinical trials (RCT).

Methods

A systematic literature review was conducted using databases including: MEDLINE, SCOPUS (including EMBASE), Cochrane library and electronic search alerts. Only articles reporting double-blind RCTs of TNF-blockers vs. placebo, with or without concomitant methotrexate (MTX), in treatment of RA were selected. Data collected were information of patients, interventions, controls, outcomes, study methods and eventual sources of bias.

Results

Forty-one articles reporting on 26 RCTs were included in the systematic review and meta-analysis. Five RCTs studied infliximab, seven etanercept, eight adalimumab, three golimumab and three certolizumab. TNF-blockers were more efficacious than placebo at all time points but were comparable to MTX. TNF-blocker and MTX combination was superior to either MTX or TNF-blocker alone. Increasing doses did not improve the efficacy. TNF-blockers were relatively safe compared to either MTX or placebo.

Conclusions

No single substance clearly rose above others in efficacy, but the results of the safety analyses suggest that etanercept might be the safest alternative. Interestingly, MTX performs nearly identically considering both efficacy and safety aspects with a margin of costs.

Introduction

Rheumatoid arthritis (RA) is an inflammatory autoimmune disease with a prevalence of 0.5–1.0 per cent in Northern Europe [1]. A recent epidemiological study from Sweden reported that 0.77% of the population have been diagnosed with RA while a survey from UK found the prevalence to be 0.82% [2], [3]. RA is usually diagnosed before the age of 60 and is more common in women than men. Both genetic and environmental factors play a role [4]. Symptoms include joint destruction, pain and impaired movement.

Since the discovery of the role of tumour necrosis factor (TNF) in chronic inflammation in RA, five drugs based on blocking TNF have entered clinical use. Infliximab, adalimumab, golimumab and certolizumab pegol (certolizumab) are monoclonal antibodies targeted against TNF whereas etanercept is a soluble TNF-receptor [5]. However, only few clinical trials compared one TNF-blocker to other TNF-blockers. Previous systematic reviews and meta-analyses have studied the subject in various settings and comparisons [6][14]. These studies concluded that while TNF-blockers are efficacious but it may still be beneficial to use them in combination therapies. Only few differences in efficacy and safety between individual substances were discovered. However, more randomized clinical trials have been published lately with additional data available to systematic reviews and most importantly, two new substances, certolizumab and golimumab, have been introduced to clinical use.

The purpose of this systematic review and meta-analysis is to study the efficacy and safety of all five currently available TNF-blockers in the treatment of RA compared to either methotrexate (MTX) and placebo or placebo alone and to perform an indirect comparison between individual substances in different drug combinations and doses and at different time points. We test the assumption that it is more efficacious and comparatively safer to use MTX in combination with a TNF-blocker in the treatment of RA compared to TNF-blocker monotherapy. We study if high doses of TNF-blockers differ from regular doses in efficacy and safety. Primary efficacy endpoint is the risk ratio between intervention and control group in American College of Rheumatology (ACR) 50% improvement at 6 months [15], [16]. Secondary efficacy endpoints include risk ratios in ACR 20%, 50% and 70% improvements at 3, 6 and 12 months in several comparisons. Primary safety endpoint is the risk ratio between intervention and control group in the number of discontinuations due to adverse events. Secondary safety endpoints include risk ratios in the number of adverse events, serious adverse events, infections, serious infections and injection site reactions.

Methods

Study selection criteria

We performed a search for randomized clinical trials of five TNF-blockers in treatment of RA. Systematic review was conducted in accordance to methods and recommendations from the Cochrane handbook [17].

According to inclusion criteria patients had to be at least 16 years of age; be diagnosed with RA using ACR 1987 criteria; and be randomized either to intervention or control group. Studies were to have one (or more) of the TNF-blockers as intervention and either placebo or combination of placebo and methotrexate as control. The TNF-blocker had to be delivered through the same route as the commercial drug and be within the dose range recommended for the commercially available products. Efficacy was measured in terms of ACR 20%, 50% and 70% improvements and thus, at least one of these had to be reported at some time point. Information regarding safety had to be reported. Previously published systematic reviews were searched for, but excluded from the systematic review due to the inclusion criteria. The protocol of the study was not published online.

Search strategy

Search strategy was designed and performed by a librarian by our request. We used the search terms rheumatoid arthritis, anti-TNF, infliximab, etanercept, adalimumab, golimumab, certolizumab, randomized clinical trials and systematic review. Variations in spelling were taken into account. References from (Ovid®) Medline, Cochrane library (Cochrane Central register of Controlled Trials, Cochrane Database for Systematic Reviews, Health Technology Assessment, Database of Abstracts of Reviews of Effects, NHS Economic Evaluation, Cochrane Methodology Register), SCOPUS (including Embase), ISI web of knowledge and several other databases were extracted and imported to reference management software (RefWorks). Clinical trial register (clinicaltrials.gov) was hand searched for unpublished trials. Duplicate entries were removed using an automated feature. There were no restrictions on study language. For search strategy, see table S1.

Study selection

References were evaluated by two individual investigators (KA, LV) using pre-defined inclusion and exclusion criteria. Decision for inclusion was made on consensus. A third investigator (YTK) made the final decision in case of disagreement. Evaluation was based on title and abstract whenever available. Full text articles from potentially relevant references were obtained in electronic or printed format and re-evaluated for inclusion by the same investigators as before. The acronym PICOS (patients, interventions, comparators, outcomes and settings) was used to assess if the references fully complied with the inclusion and exclusion criteria. As full-text article was required for the systematic review and meta-analysis, references whose full-texts we could not acquire either electronically or as printed copies from the University of Helsinki medical library were excluded. Multiple reports from a single study were considered as one study.

Evaluation for bias

As instructed in the Cochrane handbook for systematic reviews of interventions, the investigators performed an evaluation of bias rather than of methodological quality. Studies included were evaluated for an eventual bias using methods described in the Cochrane handbook. The study was to be considered “possibly biased” in case a possible source of bias was found in any of the seven dimensions evaluated. The following dimensions were considered in the bias assessment tool: Allocation sequence generation, allocation concealment, blinding of participants, personnel and outcome, incomplete outcome data, selective outcome reporting and other sources of bias. Evaluation was done by two independent assessors (KA, LV) to improve the validity. The effect of possible bias on results was studied by performing all meta-analysis twice with possibly biased RCTs included and excluded.

Data extraction

Data on study design, patient status and background, efficacy and safety were extracted from the publications using an Excel data extraction form by two independent researchers (KA, LV) to improve validity.

Meta-analyses

Data were analyzed using the intention to treat results from the included studies. Meta-analyses were performed using Cochrane Collaboration Review Manager 5.0 software. Sensitivity analyses were employed to account for the possible bias. In some settings several time points were combined to increase the power. Efficacy and safety were analyzed using dichotomous data to obtain risk ratios. Dichotomous efficacy data included ACR 20%, 50% and 70% improvements whereas dichotomous safety data was composed of the proportion of patients who experienced an adverse outcome or discontinued the treatment due to adverse events. The efficacy and safety of TNF-blockers was analyzed in six different main comparisons. Random effects model was used to account for the diversity of the studies. Heterogeneity was evaluated via subgroup analysis using Chi square and I2-statistics.

Results

Search results

5308 references were identified from electronic databases by a systematic literature search performed 5.-26.2.2010. 1613 were identified as duplicates by an automated feature in RefWorks. Additionally, 146 references were added via “search alerts”, which extended time coverage of the search to 30.6.2010. No additional references were identified from alternative sources including clinical trial registers.

Study selection

After removing duplicate entries, 3841 references were evaluated for inclusion based on title and/or abstract. Seventy six potentially relevant references were included in the next stage, where the publication was to be re-evaluated based on full text (figure 1). Full text was unavailable for 12 studies most of which were conference abstracts identified from ISI Web of Knowledge [18][29]. Patients, interventions, controls, outcomes or design of the studies did not meet the inclusion criteria of the systematic review in 17 publications [30][46]. Five review articles, one letter to the editor [47] and one erratum [48] were excluded. Several of the remaining 40 publications were reporting on a single study and were thus merged into one (table S2). Publications included in the systematic review and meta-analysis are listed in the bibliography with numbers 49–88. From the 26 clinical trials included in the systematic review, 8 used adalimumab, 7 etanercept , 5 infliximab, 3 golimumab and 3 certolizumab for intervention. The included trials have 9862 patients of which 6780 and 3082 were in intervention and control groups, respectively (table S2).

thumbnail

Figure 1. Flowchart of the selection process of the RCTs for the systematic review and meta-analysis.

doi:10.1371/journal.pone.0030275.g001

Evaluation for bias

A potential source of bias was discovered in five trials included in the systematic review (table 1). In many clinical trials there was an early escape route for patients with insufficient treatment response to avoid rapid disease progression. In some studies this was implemented by considering all patients failing to meet a pre-defined treatment response criteria (e.g. ACR 20% improvement) as “non-responders” before the actual efficacy assessment. While this may be for the best interest of the study subjects, it may introduce a bias to the evaluation of the efficacy results. Another bias was caused by switching the control group to active medication.

thumbnail

Table 1. The results of an assessment for bias in accordance to a tool by Cochrane Collaboration.*

doi:10.1371/journal.pone.0030275.t001

Efficacy

TNF-blocker vs. control.

The primary efficacy endpoint of our study was the risk ratio of 50% improvements in the ACR-treatment response criteria at six months between intervention and control group. Fourteen trials were included and of them 2 used infliximab, 2 etanercept, 5 adalimumab, 2 golimumab and 3 certolizumab for intervention. As a group, TNF-blockers reached a risk ratio of 4.07 (95% CI 2.70–6.13) regarding the achievement of the efficacy endpoint compared to controls. For infliximab, etanercept, adalimumab, golimumab and certolizumab the corresponding figures were 3.08 (0.91–10.43), 8.61 (3.55–20.86), 4.34 (3.30–5.70), 1.56 (0.93–2.60) and 5.95 (3.97–8.92), respectively (figure 2). These results suggest that infliximab and golimumab do not differ significantly from the control. In this comparison golimumab appears to be inferior in efficacy compared to etanercept, adalimumab and certolizumab even after accounting for the possible bias. TNF-blockers as a group were found to be significantly more efficacious than control at all time points using ACR 20, 50 or 70 as outcome measures. The risk ratios observed at 12 months were significantly lower than those at three or six months.

thumbnail

Figure 2. Forest plot of the ACR 50 response at 6 months.

doi:10.1371/journal.pone.0030275.g002

We found some evidence that the duration of RA predicts the efficacy of TNF-blocker treatment. Patients on either infliximab or adalimumab with disease duration more than 2 years were more likely to reach ACR 20, 50 and 70 at 12 months compared to controls than patients with disease duration less than two years (table 2).

thumbnail

Table 2. Meta-analysis of the efficacy of TNF-blockers compared to control(RR, 95% CI).

doi:10.1371/journal.pone.0030275.t002
TNF-blocker + MTX vs. MTX.

Patients on combination therapy had significantly higher ACR outcomes than ones treated with MTX alone at all time points (table 3). A statistically significant difference was revealed between ACR 20 risk ratios of certolizumab (CI 95% 5.08, 3.46–7.48) and golimumab (1.61, 0.94–2.76). However, all certolizumab studies in this comparison were potentially biased. In a subanalysis of trials with patients who had previously used MTX, the results were similar. In comparison to MTX, golimumab combination therapy was still inferior in ACR 20 efficacy at 6 months to certolizumab combination therapy, with risk ratios of 2.14 (1.59–2.89) and 5.08 (3.46–7.48), respectively. At six months patients previously naïve to MTX are statistically significantly less likely to reach either ACR 20, 50 or 70 treatment responses compared to patients who had already been previously treated with MTX. The combination of TNF-blocker and MTX was superior in efficacy to monotherapy with a TNF-blocker at almost all time points (table 4).

thumbnail

Table 3. Meta-analysis of the efficacy of combination TNF-blocker and MTX compared to MTX (RR, 95% CI).

doi:10.1371/journal.pone.0030275.t003
thumbnail

Table 4. Meta-analysis of the efficacy of combination TNF-blocker and MTX compared to TNF-blocker monotherapy (RR, 95% CI).

doi:10.1371/journal.pone.0030275.t004
TNF-blocker monotherapy vs. MTX.

There are no trials comparing monotherapy of infliximab to MTX, but combined results with the remaining four other TNF-blockers show that while the risk ratios favour the TNF-blocker, the results do not reach statistical significance at any time point using ACR 20, 50 or 70 as outcome measures. Stratifying RTCs by previous exposure to MTX does not show any statistically significant differences in the treatment response to TNF-blocker monotherapy between these two groups.

TNF-blocker monotherapy vs. placebo.

All four TNF-blockers were more efficacious than placebo with the estimates of risk ratios ranging from 2.74 (CI 95% 1.76–4.26) – 12.31 (1.64–92.41). There were no statistically significant differences in efficacy between individual substances in this comparison or, alternatively, the meta-analysis was underpowered to reveal them.

High doses of TNF-blockers vs. normal doses.

The final meta-analysis compared higher than regular doses of TNF-blockers to normal doses (table 5). Both patients using high and normal doses had to be on concomitant MTX or on TNF-blocker monotherapy. Increasing the dose of TNF-blocker provided no additional efficacy compared to regular doses except 12 months with possibly biased results excluded.

thumbnail

Table 5. Meta-analysis of the efficacy of high doses of TNF-blockers compared to normal doses of TNF-blockers.

doi:10.1371/journal.pone.0030275.t005

Sensitivity analyses

The sensitivity analyses based on the results of the bias assessments did not reveal any statistically significant bias on the efficacy results. Occasionally, however, the statistical significance between intervention and control groups disappeared due to reduced number of studies. In the sensitivity analyses, the estimate of the risk ratio decreased, increased or remained the same in 52%, 45% and 3% of cases, respectively. In some cases there were no clearly unbiased RCTs in a comparison, thus making it impossible to perform the sensitivity analysis. Significant heterogeneity was present in the first analysis comparing any intervention to any control. Heterogeneity diminished as the comparisons were stratified into smaller comparisons.

To investigate the possible effect of study patients' baseline disease activity on efficacy, two additional analyses were performed. Using ACR 50 at six months as outcome and stratifying trials into two categories by the number of swollen joints or Health Assessment Questionnaire (HAQ) score revealed no statistical differences between the subgroups. Trials with low swollen joint count and low HAQ score had risk ratios of 3.43 (CI 95% 2.03–5.78) and 3.68 (2.11–6.42), respectively, whereas trials with high swollen joint count and high HAQ score had risk ratios of 5.15 (2.72–9.75) and 4.64 (2.59–8.31), respectively.

Safety

TNF-blocker vs. control.

The primary safety endpoint of the systematic review was the discontinuation of study due to adverse events. There were 25 studies with 6292 patients in the intervention and 2994 in the control group in this analysis (table 6). As a group, the TNF-blockers did not statistically significantly differ from the control (RR 1.26, CI 95% 0.93–1.71). While the patients on infliximab (3.22, 1.76–5.91), adalimumab (1.59, 1.13–2.23), and certolizumab (2.72, 1.23–6.01), had an increased risk to discontinue, the patients on etanercept (0.71, 0.54–0.92) had a decreased risk (figure 3). Patients using certolizumab had a higher risk to experience a serious adverse event than patients on etanercept with risk ratios of 2.24 (1.38–3.63) and 0.90 (0.68–1.20), respectively. Infliximab, etanercept and golimumab increased the likelihood of an injection or infusion reaction while adalimumab and certolizumab did not statistically significantly differ from the controls in this respect although their risk ratios leaned to the same direction.

thumbnail

Figure 3. Forest plot of the number of discontinuations due to an adverse event.

doi:10.1371/journal.pone.0030275.g003
thumbnail

Table 6. Meta-analysis of the safety of TNF-blockers in different comparisons (RR, 95% CI).

doi:10.1371/journal.pone.0030275.t006
TNF-blocker + MTX vs. MTX.

In this comparison, etanercept no longer significantly decreased the likelihood of discontinuation due to adverse event (RR 0.78, CI 95% 0.56–1.09). Combined results from all substances now reached statistical significance (1.37, 1.01–1.87). In an analysis comparing the combination treatment to monotherapy there were only few differences between treatment groups. There was a trend of elevated risk ratios of multiple safety endpoints excluding injection reactions but these findings did not reach statistical significance. Golimumab increased the odds for an unspecified adverse event (1.14, 1.03–1.26) while others did not.

TNF-blocker monotherapy vs. MTX.

TNF-blocker and MTX were comparable in all respects other than injection and infusion reactions (RR 5.20, CI 95% 2.62–10.31).

TNF-blocker monotherapy vs. placebo.

The comparison of TNF-blockers and placebo showed a trend of increased risk of adverse events from TNF-blockers, but only the increase in the frequency of injection reactions was statistically significant (RR 3.69, CI 95% 1.03–13.23). In addition, certolizumab seemed to increase the risk to experience an adverse event compared to placebo (1.31, 1.08–1.26).

High doses of TNF-blockers vs. normal doses.

Increased dose of the TNF-blockers did not increase the frequency of discontinuations due to adverse events (RR 0.98, CI 95% 0.72–1.35), but the likelihood to experience an unspecified adverse event was reduced compared to normal doses (0.93, 0.89–0.97). Patients on high doses of infliximab were less likely to suffer from infusion reactions compared to those on regular doses (0.73, 0.56–0.94).

Discussion

Findings of this review

Our systematic literature search identified 40 publications reporting 26 randomized controlled trials studying the efficacy and safety of TNF-blockers. The included trials were published 1999–2010. Five trials published used infliximab [49][59], seven etanercept [60][72], eight adalimumab [73][82], three golimumab [83][85] and three certolizumab [86][88] for the intervention. Overall, there were 6780 patients in the intervention and 3082 in the control group. The patients' characteristics varied across the included trials with mean time since diagnosis ranging from 0.5 to 13 years, HAQ score from 1.25 to 1.88 and the number of swollen and tender joints from 11 and 14.03 to 24 and 37.2, respectively.

The results of the primary efficacy endpoint suggest that infliximab and golimumab do not statistically significantly differ from control regarding efficacy in a comparison between any combination of TNF-blocker and any control. Even though the different settings and heterogeneity among the studies could have accounted for the result, the finding still raises questions. Golimumab appears to be inferior in efficacy to etanercept, adalimumab and certolizumab even after accounting for the eventual bias. Patients in golimumab trials have lower count of swollen and tender joints as well as lower HAQ score, which may explain the results to some extent, although an ad hoc analysis on the effect of patient characteristics on efficacy showed no statistical significance. Disease duration seems to predict treatment response to adalimumab and infliximab at 12 months.

In the second and third meta-analysis the efficacy of MTX and TNF-blocker combination was found to be superior to either MTX or TNF-blocker alone, respectively. The increase in the number of discontinuations due to adverse events (RR 1.37 95% 1.01–1.87) compared to MTX alone is likely to be acceptable. Patients with previous exposure to MTX were more likely to benefit from the combination therapy compared to MTX naïve patients. Compared to monotherapy with a TNF-blocker the safety of the combination treatment was equal or even improved regarding some aspects.

The fourth meta-analysis found no statistical difference between MTX and TNF-blocker monotherapy and the fifth one confirmed that TNF-blocker monotherapy was more efficacious than placebo. The last secondary efficacy meta-analysis found little benefit from increasing the dose of TNF-blockers.

In the first safety comparison between TNF-blockers and control the risk ratios reached statistical significance only in the number of patients experiencing injection or infusion reactions. Interestingly, infliximab, adalimumab and certolizumab increased the risk of discontinuation of treatment due to adverse events, but etanercept made it less likely. Certolizumab was the only TNF-blocker which increased the likelihood of experiencing a serious adverse event. While TNF-blockers as a group increased the odds to experience an injection or infusion reaction this may not be the case with adalimumab and certolizumab.

Strengths and limitations

It could be asked, whether TNF-blocker naive and switchers should be included in the same review, because these patients could be very different. However, fifteen trials included in this systematic review stated previous TNF-blocker use as an exclusion criterion. In eight more trials it was unclear if switchers were included and only two certolizumab trials included switchers but excluded those who had had insufficient response to previous TNF-blocker treatment. However, the percentage of previous TNF-blocker users in these two trials was small (2–4%) and a sensitivity analysis was performed.

While broader comparisons with larger number of trials may be more likely to reach statistically significant results (1.00 not included in the confidence interval), their validity may be questioned. Heterogeneity introduced by combining the results of trials with different settings causes random effects model to calculate wider confidence intervals than fixed effects model would do. While reducing the possibility of type I error, it may introduce a type II error. Hence, the efficacies of TNF-blockers were compared with different controls, combinations and dosages in smaller, but more homogenous comparisons.

Results of the sensitivity analyses revealed that the source of bias in the RCTs is as likely to lead to underestimation as overestimation of the risk difference between intervention and control groups. However, the homogeneity of study population, intervention, control, outcomes and study settings are likely to be more crucial to the validity of the meta-analysis. Length of exposure was not taken into account in the safety analyses, only the difference in risk ratios between intervention and control group.

The methods used in the study were derived from the Handbook for systematic reviews of interventions by Cochrane Collaboration. The team involved in study design and execution included clinicians, methodology experts and pharmacists. Two researchers independently worked at each step and afterwards combined their results to improve the validity of the study. The meta-analyses were done using Review Manager 5.0 –software. The report was written in accordance to the PRISMA-statement (table S3).

Our systematic review and meta-analysis has some limitations. The authors of the included trials were not contacted to retrieve unpublished data. Many studies that lasted for one year or more only reported results at 12 months. The meta-analyses would have been more powered if the efficacy results had been reported at all time points. Selective reporting was included in the evaluation for bias, but we were unable to identify any bias here.

Findings in comparison to other systematic reviews

Another systematic review and meta-analysis pooled efficacy results from different time points and found slightly different estimates for the efficacy of TNF-blockers, which is likely due to differences in study designs [12]. Several large clinical trials have been published since the aforementioned review along with the introduction of two novel TNF-blockers, certolizumab and golimumab. Our study distinguishes itself from previous systematic reviews by including larger number of clinical trials and by presenting efficacy results separately at three, six and twelve months. Our results reveal that the new substances do not offer improved efficacy or safety profile over the already existing ones. A recent systematic review concluded that certolizumab is at least as efficacious compared to older TNF-blockers [14]. However, the study did not include studies with MTX comparison nor evaluate the safety of biologic treatments. Certolizumab may be more efficacious than golimumab but may also be associated with a greater risk of serious adverse events.

A previous systematic review reached the same conclusion as we did, regarding the increased dose of TNF-blockers. Contradicting our results, they found high doses leading to two-fold risk of serious infections. In contrast to our direct approach, they however separately compared recommended and high doses of TNF-blockers to placebo [10]. Another systematic review concluded that infliximab might require an increased dosage level to reach similar efficacy as etanercept and adalimumab have [13].

Implications for practise and research

The novel TNF-blockers may offer an alternative to older substances but do not make them obsolete. On the contrary, etanercept may be the best choice when taking into account safety profiles of the TNF-blockers. Infliximab, etanercept and adalimumab have been in clinical use for years with extensive amount of post-marketing data available. More post-marketing information is needed on certolizumab and golimumab for comprehensive pharmacovigilance.

The annual medication costs of TNF-blockers are more than 10 000€ while the MTX treatment costs less than 100€ per year. Subgroup analysis in table 3 suggests that considering the high expenses of biologics, the treatment of RA could be initiated with MTX while combining TNF-blockers to ongoing treatment in patients with insufficient response to MTX. Even though safety was not compromised, it might not be cost-effective to use high doses of TNF-blockers. Given the limited resources in healthcare systems our results may help clinicians and decision makers to get most out of the expensive, but efficacious treatment.

The next step could be to analyze the efficacy and safety of not just TNF-blockers, but all biologics in a large systematic review and meta-analysis. One randomized clinical trial included in our systematic review actually compared abatacept to infliximab [55]. However, a systematic review is indicated to summarise the evidence.

Supporting Information

Table S1.

Search strategy to (Ovid®) Medline.

doi:10.1371/journal.pone.0030275.s001

(DOCX)

Table S2.

Description of studies included in the systematic review and meta-analysis. 1 = Evaluation based on 28 joints. 2 = Baseline data. 3 = Evaluation based on 71 joints. 4 = Values in median. 5 = placebo switched to active medication at 6 months. Ada = Adalimumab. Cer = Certolizumab pegol. Eta = Etanercept. Gol = Golimumab. Inf = Iinfliximab. MTX = methotrexate.

doi:10.1371/journal.pone.0030275.s002

(DOCX)

Table S3.

PRISMA Checklist of items to include when reporting a systematic review or meta-analysis.

doi:10.1371/journal.pone.0030275.s003

(DOCX)

Acknowledgments

We would like to thank librarian Terhi Sandgren for the design and execution of the systematic literature search.

Author Contributions

Conceived and designed the experiments: KA LV AM YTK DN MB. Performed the experiments: KA LV. Analyzed the data: KA LV. Wrote the paper: KA LV AM YTK DN MB.

References

  1. 1. Aho K, Kaipiainen-Seppänen O, Heliövaara M, Klaukka T (1998) Epidemiology of rheumatoid arthritis in Finland. Seminars in Arthritis and Rheumatism 27: 325–334.
  2. 2. Neovius M, Simard JF, Askling J (2011) Nationwide prevalence of rheumatoid arthritis and penetration of disease-modifying drugs in Sweden. Annals of the rheumatic diseases 70: 624–629. doi:10.1136/ard.2010.133371.
  3. 3. Symmons D, Turner G, Webb R, Asten P, Barrett E, et al. (2002) The prevalence of rheumatoid arthritis in the United Kingdom: new estimates for a new century. Rheumatology 41: 793–800. doi:10.1093/rheumatology/41.7.793.
  4. 4. Gregor AJMAC, Snieder H, Rigby AS, Koskenvuo M, Kaprio J, et al. (2000) Characterizing the quantitative genetic contribution to rheumatoid arthritis using data from twins. Arthritis & Rheumatism 43: 30–37.
  5. 5. Tracey D, Klareskog L, Sasso EH, Salfeld JG, Tak PP (2008) Tumor necrosis factor antagonist mechanisms of action: a comprehensive review. Pharmacology & therapeutics 117: 244–279. doi:10.1016/j.pharmthera.2007.10.001.
  6. 6. Singh JA, Christensen R, Wells GA, Suarez-Almazor ME, Buchbinder R, et al. (2009) Biologics for rheumatoid arthritis: an overview of Cochrane reviews. Cochrane database of systematic reviews (Online): CD007848. Available: http://www.ncbi.nlm.nih.gov/pubmed/19821​440. Accessed 2 May 2011.
  7. 7. Wiens A, Correr CJ, Venson R, Grochocki MC, Otuki MF, et al. (2009) A meta-analysis of the efficacy and safety of using infliximab for the treatment of rheumatoid arthritis. Clinical rheumatology 28: 1365–1373. doi:10.1007/s10067-009-1233-9.
  8. 8. Wiens A, Correr CJ, Pontarolo R, Venson R, Quinalha JV, et al. (2009) A systematic review and meta-analysis of the efficacy and safety of etanercept for treating rheumatoid arthritis. Scandinavian journal of immunology 70: 337–344. doi:10.1111/j.1365-3083.2009.02296.x.
  9. 9. Chen Y-F, Jobanputra P, Barton P, Jowett S, Bryan S, et al. (2006) Adalimumab, etanercept and infliximab for the treatment of rheumatoid arthritis. Health Technology Assessment 10: 248.
  10. 10. Alonso-Ruiz A, Pijoan JI, Ansuategui E, Urkaregi A, Calabozo M, et al. (2008) Tumor necrosis factor alpha drugs in rheumatoid arthritis: systematic review and metaanalysis of efficacy and safety. BMC musculoskeletal disorders 9. Available: http://www.pubmedcentral.nih.gov/article​render.fcgi?artid=2377247&tool=pmcentrez​&rendertype=abstract. Accessed 14 September 2011.
  11. 11. Leombruno JP, Einarson TR, Keystone EC (2009) The safety of anti-tumour necrosis factor treatments in rheumatoid arthritis: meta and exposure-adjusted pooled analyses of serious adverse events. Annals of the rheumatic diseases 68: 1136–1145. doi:10.1136/ard.2008.091025.
  12. 12. Wiens A, Venson R, Correr CJ, Otuki MF, Pontarolo R (2010) Meta-analysis of the efficacy and safety of adalimumab, etanercept, and infliximab for the treatment of rheumatoid arthritis. Pharmacotherapy 30: 339–353. doi:10.1592/phco.30.4.339.
  13. 13. Kristensen LE, Christensen R, Bliddal H, Geborek P, Danneskiold-Samsøe B, et al. (2007) The number needed to treat for adalimumab, etanercept, and infliximab based on ACR50 response in three randomized controlled trials on established rheumatoid arthritis: a systematic literature review. Scandinavian journal of rheumatology 36: 411–417. doi:10.1080/03009740701607067.
  14. 14. Launois R, Avouac B, Berenbaum F, Blin O, Bru I, et al. (2011) Comparison of certolizumab pegol with other anticytokine agents for treatment of rheumatoid arthritis: a multiple-treatment Bayesian metaanalysis. The Journal of rheumatology 38: 835–845. doi:10.3899/jrheum.100665.
  15. 15. Felson DT, Anderson JJ, Boers M, Bombardier C, Chernoff M, et al. (1993) The American Colloge of Rheumatolgy preliminary core set of disease activity measures for rheumatoid arthritis clinical trials. Arthritis and rheumatism 36: 729–740.
  16. 16. Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF, et al. (1988) The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis and rheumatism 31: 315–324.
  17. 17. Higgins JP, Green S, editors. (2008) Cochrane Handbook for Systematic Reviews of Interventions. Chichester: John Wiley & Sons Ltd. England. 672 p.
  18. 18. Bankhurst A (1999) Etanercept and methotrexate combination therapy. Clinical and experimental rheumatology 17: 69–72.
  19. 19. Elliott MJ, Maini RN, Feldmann M, Kalden JR, Antoni C, et al. (1994) Randomised double-blind comparison of chimeric monoclonal antibody to tumour necrosis factor alpha (cA2) versus placebo in rheumatoid arthritis. Lancet 344: 1105–1110.
  20. 20. Ericson ML, Wajdula J, European Etanercept Investigators (1999) A double-blind, placebo-controlled study of the efficacy and safety of four different doses of etanercept in patients with rheumatoid arthritis. Arthritis and Rheumatism 42: Suppl 582.
  21. 21. Fleischmann RM, Emery P, Moreland LW, Hsia EC, Strusberg I, et al. (2008) Golimumab, a new human Anti-TNF-alpha monoclonal antibody, administered subcutaneously everv 4 weeks in methotrexate-naive patients with active rheumatoid arthritis: A randomized, double-blind, placebo-controlled, GO-BEFORE study. Arthritis and Rheumatism 58: 530.
  22. 22. Gao GH, Li J, Xie HW, Lu Z (2010) [Therapeutic effect of infliximab on moderate and severe active rheumatoid arthritis]. Nan fang yi ke da xue xue bao = Journal of Southern Medical University 30: 724–726.
  23. 23. Huang F, Zhang FC, Bao CD, Tao Y, Gu JR, et al. (2009) [Adalimumab plus methotrexate for the treatment of rheumatoid arthritis: a multi-center randomized, double-blind, placebo-controlled clinical study.]. Chung-Hua Nei Ko Tsa Chih Chinese Journal of Internal Medicine 48: 916–921.
  24. 24. Kavanaugh AF, Cush JJ, St Clair EW, McCune WJ, Braakman TAJ, et al. (1996) Anti-TNF-alpha monoclonal antibody (mAb) treatment of rheumatoid arthritis (RA) patients with active disease on methotrexate (MTX): Results of a double-blind, placebo controlled multicenter trial. Arthritis and Rheumatism 39: 575.
  25. 25. Kavanaugh A, St Clair EW, McCune WJ, Braakman T, Lipsky P (2000) Chimeric anti-tumor necrosis factor-alpha monoclonal antibody treatment of patients with rheumatoid arthritis receiving methotrexate therapy. The Journal of rheumatology 27: 841–850.
  26. 26. Kay J, Matteson EL, Dasgupta B, Nash P, Durez P, et al. (2007) Improvement in DAS28 response through one-year of golimumab treatment in patients with active RA despite treatment with methotrexate: A phase II, randomized, double blind, placebo controlled, dose ranging trial. Annals of the Rheumatic Diseases 66: 175–176.
  27. 27. Kay J, Matteson EL, Dasgupta B, Nash P, Durezs P, et al. (2006) Das28 responses in patients with active rheumatoid arthritis who received golimumab and methotrexate: A randomized, double-blind, placebo-controlled phase 2 trial. Annals of the Rheumatic Diseases 65: 323.
  28. 28. Keystone E, Weinblatt M, Furst D, Weisman M, Moreland L, et al. (2001) The ARMADA trial: A double-blind placebo controlled trial of the fully human anti-TNF monoclonal antibody, adalimumab (D2E7), in patients with active RA on methotrexate (MTX). Arthritis and Rheumatism 44: 965.
  29. 29. Zhang W, Shi Q, Wu D-H, Bao C-D, Yang N-P, et al. (2009) [Efficacy and safety of infliximab in patients with rheumatoid arthritis]. Zhonghua yi xue za zhi 89: 1876–1880.
  30. 30. Allaart CF, Goekoop-Ruiterman YP, de Vries-Bouwstra JK, Breedveld FC, Dijkmans BA, et al. (2006) Aiming at low disease activity in rheumatoid arthritis with initial combination therapy or initial monotherapy strategies: the BeSt study. Clinical & Experimental Rheumatology 24: 77–82.
  31. 31. Antoni C, Kalden JR (1999) Combination therapy of the chimeric monoclonal anti-tumor necrosis factor alpha antibody (infliximab) with methotrexate in patients with rheumatoid arthritis. Clinical & Experimental Rheumatology 17: 73–77.
  32. 32. Durez P, Malghem J, Nzeusseu Toukap A, Depresseux G, Lauwerys BR, et al. (2007) Treatment of early rheumatoid arthritis: a randomized magnetic resonance imaging study comparing the effects of methotrexate alone, methotrexate in combination with infliximab, and methotrexate in combination with intravenous pulse methylprednisolone. Arthritis & Rheumatism 56: 3919–3927.
  33. 33. Durez P, Nzeusseu Toukap A, Lauwerys BR, Manicourt DH, Verschueren P, et al. (2004) A randomised comparative study of the short term clinical and biological effects of intravenous pulse methylprednisolone and infliximab in patients with active rheumatoid arthritis despite methotrexate treatment. Annals of the Rheumatic Diseases 63: 1069–1074.
  34. 34. Furst DE, Schiff MH, Fleischmann RM, Strand V, Birbara CA, et al. (2003) Adalimumab, a fully human anti tumor necrosis factor-alpha monoclonal antibody, and concomitant standard antirheumatic therapy for the treatment of rheumatoid arthritis: results of STAR (Safety Trial of Adalimumab in Rheumatoid Arthritis). Journal of Rheumatology 30: 2563–2571.
  35. 35. Goekoop-Ruiterman Y, de Vries-Bouwstra JK, Allaart CF, van Zeben D, Kerstens P, et al. (2008) Clinical and radiographic outcomes of four different treatment strategies in patients with early rheumatoid arthritis (the BeSt study): A randomized, controlled trial. Arthritis and rheumatism 58 2: Suppl126–135.
  36. 36. Goekoop-Ruiterman YPM, de Vries-Bouwstra JK, Allaart CF, van Zeben D, Kerstens PJSM, et al. (2005) Clinical and radiographic outcomes of four different treatment strategies in patients with early rheumatoid arthritis (the BeSt study): a randomized, controlled trial. Arthritis and rheumatism 52: 3381–3390. doi:10.1002/art.21405.
  37. 37. Maini RN, Breedveld FC, Kalden JR, Smolen JS, Davis D, et al. (1998) Therapeutic efficacy of multiple intravenous infusions of anti-tumor necrosis factor alpha monoclonal antibody combined with low-dose weekly methotrexate in rheumatoid arthritis. Arthritis & Rheumatism 41: 1552–1563.
  38. 38. Miceli-Richard C, Comets E, Verstuyft C, Tamouza R, Loiseau P, et al. (2008) A single tumour necrosis factor haplotype influences the response to adalimumab in rheumatoid arthritis. Annals of the Rheumatic Diseases 67: 478–484.
  39. 39. Moreland LW, Baumgartner SW, Schiff MH, Tindall EA, Fleischmann RM, et al. (1997) Treatment of rheumatoid arthritis with a recombinant human tumor necrosis factor receptor (p75)-Fc fusion protein. New England Journal of Medicine 337: 141–147.
  40. 40. van der Bijl AE, Goekoop-Ruiterman YPM, de Vries-Bouwstra JK, Ten Wolde S, Han KH, et al. (2007) Infliximab and methotrexate as induction therapy in patients with early rheumatoid arthritis. Arthritis and rheumatism 56: 2129–2134. doi:10.1002/art.22718.
  41. 41. van Riel PLCM, Taggart AJ, Sany J, Gaubitz M, Nab HW, et al. (2006) Efficacy and safety of combination etanercept and methotrexate versus etanercept alone in patients with rheumatoid arthritis with an inadequate response to methotrexate: the ADORE study. Annals of the rheumatic diseases 65: 1478–1483. doi:10.1136/ard.2005.043299.
  42. 42. Weisman MH, Paulus HE, Burch FX, Kivitz AJ, Fierer J, et al. (2007) A placebo-controlled, randomized, double-blinded study evaluating the safety of etanercept in patients with rheumatoid arthritis and concomitant comorbid diseases. Rheumatology 46: 1122–1125.
  43. 43. Westhovens R, Yocum D, Han J, Berman A, Strusberg I, et al. (2006) The safety of infliximab, combined with background treatments, among patients with rheumatoid arthritis and various comorbidities: a large, randomized, placebo-controlled trial. Arthritis and rheumatism 54: 1075–1086. doi:10.1002/art.21734.
  44. 44. Zhang F-C, Hou Y, Huang F, Wu D-H, Bao C-D, et al. (2006) Infliximab versus placebo in rheumatoid arthritis patients receiving concomitant methotrexate: A preliminary study from China. APLAR Journal of Rheumatology 9: 127–130.
  45. 45. Smolen JS, Kay J, Doyle MK, Landewe R, Matteson EL, et al. (2009) Golimumab in patients with active rheumatoid arthritis after treatment with tumour necrosis factor alpha inhibitors (GO-AFTER study): a multicentre, randomised, double-blind, placebo-controlled, phase III,trial. Lancet 374: 210–221.
  46. 46. Rahman MU, Strusberg I, Geusens P, Berman A, Yocum D, et al. (2007) Double-blinded infliximab dose escalation in patients with rheumatoid arthritis. Annals of the rheumatic diseases 66: 1233–1238. doi:10.1136/ard.2006.065995.
  47. 47. Yocum D, Wolfe F, Rahman MU, Han J, Berman A, et al. (2004) The safety and efficacy of infliximab in RA: 1-year results of a large, randomized, placebo-controlled trial in patients with various comorbidities and background treatments as encountered in clinical practice. Arthritis and Rheumatism 50: 659–659.
  48. 48. Keystone (2009) Certolizumab Pegol Plus Methotrexate Is Significantly More Effective Than Placebo Plus Methotrexate in Active Rheumatoid Arthritis Findings of a Fifty-Two-Week, Phase III, Multicenter, Randomized, Double-Blind, Placebo-Controlled, Parallel-Group Study. Arthritis and Rheumatism 60: 1249.
  49. 49. Abe T, Takeuchi T, Miyasaka N, Hashimoto H, Kondo H, et al. (2006) A multicenter, double-blind, randomized, placebo controlled trial of infliximab combined with low dose methotrexate in Japanese patients with rheumatoid arthritis. Journal of Rheumatology 33: 37–44.
  50. 50. Bessette L, Schiff MH, Kieserman M (2007) The efficacy and safety of abatacept or infliximab in ra patients with an inadequate response to mtx: results from a 1-year double-blind, randomized, placebo-controlled trial. The Journal of rheumatology 34: 1627.
  51. 51. Lipsky PE, van der Heijde D, St Clair W, Furst DE, Breedveld FC, et al. (2000) Infliximab and methotrexate in the treatment of rheumatoid arthritis. The New England journal of medicine 343: 1594–1602. doi:10.1016/S0167-5699(99)01552-2.
  52. 52. Maini RN, Breedveld FC, Kalden JR, Smolen JS, Furst D, et al. (2004) Sustained improvement over two years in physical function, structural damage, and signs and symptoms among patients with rheumatoid arthritis treated with infliximab and methotrexate. Arthritis and rheumatism 50: 1051–1065. doi:10.1002/art.20159.
  53. 53. Maini R, St Clair EW, Breedveld F, Furst D, Kalden J, et al. (1999) Infliximab (chimeric anti-tumour necrosis factor alpha monoclonal antibody) versus placebo in rheumatoid arthritis patients receiving concomitant methotrexate: a randomised phase III trial. ATTRACT Study Group. Lancet 354: 1932–1939.
  54. 54. Quinn MA, Conaghan PG, O'Connor PJ, Karim Z, Greenstein A, et al. (2005) Very early treatment with infliximab in addition to methotrexate in early, poor-prognosis rheumatoid arthritis reduces magnetic resonance imaging evidence of synovitis and damage, with sustained benefit after infliximab withdrawal: results from a twelve-m. Arthritis and rheumatism 52: 27–35. doi:10.1002/art.20712.
  55. 55. Schiff M, Keiserman M, Codding C, Songcharoen S, Berman A, et al. (2008) Efficacy and safety of abatacept or infliximab vs placebo in ATTEST: a phase III, multi-centre, randomised, double-blind, placebo-controlled study in patients with rheumatoid arthritis and an inadequate response to methotrexate. Annals of the rheumatic diseases 67: 1096–1103. doi:10.1136/ard.2007.080002.
  56. 56. Smolen JS, Han C, Bala M, Maini RN, Kalden JR, et al. (2005) Evidence of radiographic benefit of treatment with infliximab plus methotrexate in rheumatoid arthritis patients who had no clinical improvement: a detailed subanalysis of data from the anti-tumor necrosis factor trial in rheumatoid arthritis with concomi. Arthritis and rheumatism 52: 1020–1030. doi:10.1002/art.20982.
  57. 57. Smolen JS, Van Der Heijde DMFM, St Clair EW, Emery P, Bathon JM, et al. (2006) Predictors of joint damage in patients with early rheumatoid arthritis treated with high-dose methotrexate with or without concomitant infliximab: results from the ASPIRE trial. Arthritis and rheumatism 54: 702–710. doi:10.1002/art.21678.
  58. 58. St Clair EW, van der Heijde DMFM, Smolen JS, Maini RN, Bathon JM, et al. (2004) Combination of infliximab and methotrexate therapy for early rheumatoid arthritis: a randomized, controlled trial. Arthritis and rheumatism 50: 3432–3443. doi:10.1002/art.20568.
  59. 59. van Vollenhoven RF, Schiff M, Keiserman M, Codding C, Songcharoen S, et al. (2008) Safety of abatacept or infliximab in RA patients with an inadequate response to MTX: Results from a 1-year double-blind randomized clinical trial. Scandinavian journal of rheumatology Suppl 123: 52.
  60. 60. Bathon JM, Genovese MC (2003) The Early Rheumatoid Arthritis (ERA) trial comparing the efficacy and safety of etanercept and methotrexate. Clinical and experimental rheumatology 21: 195–197.
  61. 61. Bathon JM, Martin RW, Fleischmann RM, Tesser JR, Schiff MH, et al. (2000) A comparison of etanercept and methotrexate in patients with early rheumatoid arthritis. The New England journal of medicine 343: 1586–1593.
  62. 62. Emery P, Breedveld FC, Hall S, Durez P, Chang DJ, et al. (2008) Comparison of methotrexate monotherapy with a combination of methotrexate and etanercept in active, early, moderate to severe rheumatoid arthritis (COMET): a randomised, double-blind, parallel treatment trial. Lancet 372: 375–382. doi:10.1016/S0140-6736(08)61000-4.
  63. 63. Emery P, Breedveld F, van der Heijde D, Ferraccioli G, Dougados M, et al. (2010) Two-year clinical and radiographic results with combination etanercept-methotrexate therapy versus monotherapy in early rheumatoid arthritis: a two-year, double-blind, randomized study. Arthritis and rheumatism 62: 674–682. doi:10.1002/art.27268.
  64. 64. Genovese MC, Bathon JM, Martin RW, Fleischmann RM, Tesser JR, et al. (2002) Etanercept versus methotrexate in patients with early rheumatoid arthritis: two-year radiographic and clinical outcomes. Arthritis and rheumatism 46: 1443–1450. doi:10.1002/art.10308.
  65. 65. Kavanaugh A, Klareskog L, van der Heijde D, Li J, Freundlich B, et al. (2008) Improvements in clinical response between 12 and 24 weeks in patients with rheumatoid arthritis on etanercept therapy with or without methotrexate. Annals of the rheumatic diseases 67: 1444–1447. doi:10.1136/ard.2008.094524.
  66. 66. Keystone EC, Schiff MH, Kremer JM, Kafka S, Lovy M, et al. (2004) Once-weekly administration of 50 mg etanercept in patients with active rheumatoid arthritis: results of a multicenter, randomized, double-blind, placebo-controlled trial. Arthritis and rheumatism 50: 353–363. doi:10.1002/art.20019.
  67. 67. Klareskog L, van der Heijde D, de Jager JP, Gough A, Kalden J, et al. (2004) Therapeutic effect of the combination of etanercept and methotrexate compared with each treatment alone in patients with rheumatoid arthritis: double-blind randomised controlled trial. Lancet 363: 675–681. doi:10.1016/S0140-6736(04)15640-7.
  68. 68. Lan J-L, Chou S-J, Chen D-Y, Chen Y-H, Hsieh T-Y, et al. (2004) A comparative study of etanercept plus methotrexate and methotrexate alone in Taiwanese patients with active rheumatoid arthritis: a 12-week , double -blind , randomized , placebo -controlled study. Journal Of The Formosan Medical Association 103: 618–623.
  69. 69. Moreland LW, Schiff MH, Baumgartner SW, Tindall EA, Fleischmann RM, et al. (1999) Etanercept therapy in rheumatoid arthritis. A randomized, controlled trial. Annals of internal medicine 130: 478–486.
  70. 70. van der Heijde D, Klareskog L, Landewé R, Bruyn GAW, Cantagrel A, et al. (2007) Disease remission and sustained halting of radiographic progression with combination etanercept and methotrexate in patients with rheumatoid arthritis. Arthritis and rheumatism 56: 3928–3939. doi:10.1002/art.23141.
  71. 71. van der Heijde D, Klareskog L, Rodriguez-Valverde V, Codreanu C, Bolosiu H, et al. (2006) Comparison of etanercept and methotrexate, alone and combined, in the treatment of rheumatoid arthritis: two-year clinical and radiographic results from the TEMPO study, a double-blind, randomized trial. Arthritis and rheumatism 54: 1063–1074. doi:10.1002/art.21655.
  72. 72. Weinblatt ME, Kremer JM, Bankhurst AD, Bulpitt KJ, Fleischmann RM, et al. (1999) A trial of etanercept, a recombinant tumor necrosis factor receptor:Fc fusion protein, in patients with rheumatoid arthritis receiving methotrexate. The New England journal of medicine 340: 253–259. doi:10.1056/NEJM199901283400401.
  73. 73. Breedveld FC, Weisman MH, Kavanaugh AF, Cohen SB, Pavelka K, et al. (2006) The PREMIER study: A multicenter, randomized, double-blind clinical trial of combination therapy with adalimumab plus methotrexate versus methotrexate alone or adalimumab alone in patients with early, aggressive rheumatoid arthritis who had not had previo. Arthritis and rheumatism 54: 26–37. doi:10.1002/art.21519.
  74. 74. Chen D-Y, Chou S-J, Hsieh T-Y, Chen Y-H, Chen H-H, et al. (2009) Randomized, double-blind, placebo-controlled, comparative study of human anti-TNF antibody adalimumab in combination with methotrexate and methotrexate alone in Taiwanese patients with active rheumatoid arthritis. Journal of the Formosan Medical Association = Taiwan yi zhi 108: 310–319. doi:10.1016/S0929-6646(09)60071-1.
  75. 75. Jamal S, Patra K, Keystone EC (2009) Adalimumab response in patients with early versus established rheumatoid arthritis: DE019 randomized controlled trial subanalysis. Clinical rheumatology 28: 413–419. doi:10.1007/s10067-008-1064-0.
  76. 76. Keystone EC, Haraoui B, Bykerk VP (2003) Role of adalimumab in the treatment of early rheumatoid arthritis. Clinical and experimental rheumatology 21: 198–199.
  77. 77. Keystone EC, Kavanaugh AF, Sharp JT, Tannenbaum H, Hua Y, et al. (2004) Radiographic, clinical, and functional outcomes of treatment with adalimumab (a human anti-tumor necrosis factor monoclonal antibody) in patients with active rheumatoid arthritis receiving concomitant methotrexate therapy: a randomized, placebo-controlled. Arthritis and rheumatism 50: 1400–1411. doi:10.1002/art.20217.
  78. 78. Kim H-Y, Lee S-K, Song YW, Yoo D-H, Koh E-M, et al. (2007) study of the human anti-tumor necrosis factor antibody adalimumab administered as subcutaneous injections in Korean rheumatoid arthritis patients treated with methotrexate. APLAR Journal of Rheumatology 10: 9–16.
  79. 79. Miyasaka N (2008) Clinical investigation in highly disease-affected rheumatoid arthritis patients in Japan with adalimumab applying standard and general evaluation: the CHANGE study. Modern rheumatology/the Japan Rheumatism Association 18: 252–262. doi:10.1007/s10165-008-0045-0.
  80. 80. van de Putte LBA, Rau R, Breedveld FC, Kalden JR, Malaise MG, et al. (2003) Efficacy and safety of the fully human anti-tumour necrosis factor monoclonal antibody adalimumab (D2E7) in DMARD refractory patients with rheumatoid arthritis: a 12 week, phase II study. Annals of the Rheumatic Diseases 62: 1168–1177. doi:10.1136/ard.2003.009563.
  81. 81. van de Putte LBA, Atkins C, Malaise M, Sany J, Russell AS, et al. (2004) Efficacy and safety of adalimumab as monotherapy in patients with rheumatoid arthritis for whom previous disease modifying antirheumatic drug treatment has failed. Annals of the rheumatic diseases 63: 508–516. doi:10.1136/ard.2003.013052.
  82. 82. Weinblatt ME, Keystone EC, Furst DE, Moreland LW, Weisman MH, et al. (2003) Adalimumab, a fully human anti-tumor necrosis factor alpha monoclonal antibody, for the treatment of rheumatoid arthritis in patients taking concomitant methotrexate: the ARMADA trial. Arthritis and rheumatism 48: 35–45. doi:10.1002/art.10697.
  83. 83. Emery P, Fleischmann RM, Moreland LW, Hsia EC, Strusberg I, et al. (2009) Golimumab, a human anti-tumor necrosis factor alpha monoclonal antibody, injected subcutaneously every four weeks in methotrexate-naive patients with active rheumatoid arthritis: twenty-four-week results of a phase III, multicenter, randomized, double-bli. Arthritis and rheumatism 60: 2272–2283. doi:10.1002/art.24638.
  84. 84. Kay J, Matteson EL, Dasgupta B, Nash P, Durez P, et al. (2008) Golimumab in patients with active rheumatoid arthritis despite treatment with methotrexate: a randomized, double-blind, placebo-controlled, dose-ranging study. Arthritis and rheumatism 58: 964–975. doi:10.1002/art.23383.
  85. 85. Keystone EC, Genovese MC, Klareskog L, Hsia EC, Hall ST, et al. (2009) Golimumab, a human antibody to tumour necrosis factor {alpha} given by monthly subcutaneous injections, in active rheumatoid arthritis despite methotrexate therapy: the GO-FORWARD Study. Annals of the rheumatic diseases 68: 789–796. doi:10.1136/ard.2008.099010.
  86. 86. Fleischmann R, Vencovsky J, van Vollenhoven RF, Borenstein D, Box J, et al. (2009) Efficacy and safety of certolizumab pegol monotherapy every 4 weeks in patients with rheumatoid arthritis failing previous disease-modifying antirheumatic therapy: the FAST4WARD study. Annals of the rheumatic diseases 68: 805–811. doi:10.1136/ard.2008.099291.
  87. 87. Keystone E, Heijde DVD, Mason D, Landewé R, Vollenhoven RV, et al. (2008) Certolizumab pegol plus methotrexate is significantly more effective than placebo plus methotrexate in active rheumatoid arthritis: findings of a fifty-two-week, phase III, multicenter, randomized, double-blind, placebo-controlled, parallel-group study. Arthritis and rheumatism 58: 3319–3329. doi:10.1002/art.23964.
  88. 88. Smolen J, Landewé RB, Mease P, Brzezicki J, Mason D, et al. (2009) Efficacy and safety of certolizumab pegol plus methotrexate in active rheumatoid arthritis: the RAPID 2 study. A randomised controlled trial. Annals of the rheumatic diseases 68: 797–804. doi:10.1136/ard.2008.101659.