Rheumatoid arthritis (RA) is a chronic inflammatory disease characterized by progressive joint destruction. As joints deteriorate, patients suffer from pain and loss of function, often accompanied by decreasing quality of life and increasing mortality (1). Depending on the severity of the disease at onset, the risk of disability may be up to 30%, and mortality can be increased by up to 52%, frequently as a result of infection or circulatory disease (2).

RA treatment aims to minimize disease activity, thereby preventing or controlling joint damage and reducing the risk of other serious co-morbidities, such as heart disease or stroke. Early intervention is vital in patients with confirmed RA to preserve joint function (3–5).

Non-steroidal anti-inflammatory drugs (NSAIDs) and glucocorticoids are used to control pain and the inflammatory process (6). Following the diagnosis of RA, patients receive disease-modifying antirheumatic drugs (DMARDs), which reduce the signs and symptoms of the disease, and potentially inhibit radiographic progression (6). While a number of RA patients respond to DMARDs, a large proportion of RA remains active despite such treatments. The approach of targeting cytokines has considerably improved the success in the treatment of RA. In the clinical application, five tumor necrosis factor (TNF)-α inhibitors are available: infliximab (IFX), etanercept (ETN), adalimumab (ADA) (7–10), golimumab (GOLI) and certolizumab pegol (CZP).

This review focuses on the development of these agents regarding their effects on symptoms evaluated by the American College of Rheumatology (ACR) response criteria, structure (in the light of the erosion, joint-space narrowing and Sharp scores), and physical function [based on standardized questionnaires, such as the Health Assessment Questionnaire (HAQ)].

TNF-α is a significant cytokine that mediates inflammation in RA. Elevation of TNF-α levels have been observed in synovial fluid and the synovium of patients with RA (11). TNF-α plays an extremely central role in driving inflammation and bone degradation (12). Due to its influence on various cells in synovial membrane, such as synoviocytes, macrophages, chondrocytes and osteoclasts, which are able to produce metalloproteinases, collagenase and stromelysin, TNF-α induces local inflammation and pannus formation, eventually leading to further erosion of cartilage and bone destruction. Introduction of TNF-α inhibitors has revolutionized RA treatment options resulting in the development of further biologic DMARDs (13). The effects of a TNF-α blockade are partially dependent on synovial TNF-α expression and infiltration by TNF-α-producing inflammatory cells (14). The progress in biotechnology contributes to the development of biological agents, such as anti-TNF-α monoclonal antibodies, as a strategy for the treatment of chronic inflammatory diseases.

IFX is a recombinant IgG1 monoclonal antibody specific for TNF-α that hinders the cytokine from triggering the cellular TNF receptor complex (15). IFX needs to be administered by intravenous infusion and has a terminal half-life of 8–10 days. Thus, it is administered every 4–8 weeks and the dosage varies from 3 to 6 (to 10) mg/kg.

The efficacy of IFX with MTX was previously demonstrated in several trials (Table I). Patients receiving combination therapy exhibited evidently higher median improvements compared to patients in the MTX plus placebo (PBO) group (16–18). In addition, the clinical efficacy was similar at varying dosages in the IFX group (16–18). In terms of radiographic images, the combination of IFX plus MTX prevented the radiographic progression and led to lasting clinical amelioration (16). IFX treatment inhibited joint damage progression even in patients who did not receive MTX in the RISING study (18). Compared to the MTX-only-treated patients, erosions and joint space narrowing evidently reduced from baseline in the IFX plus MTX-treated patients, with the exception of IFX 3 mg/kg every 8 weeks. There were fewer newly eroded joints per patient in the IFX plus MTX treatment groups compared to the MTX-only group (17). In their study, St. Clair et al demonstrated that HAQ scores showed improvement in the IFX group compared to the group receiving MTX alone (16).

The most common adverse events found in clinical trials of IFX are infusion reactions and infection. The therapy of IFX might increase the risk of malignancies and cardiovascular conditions (19). The incidence of serious infections, acute infusion reactions and death was similar in patients treated with IFX plus MTX and those who received MTX only (17). Among the serious infections, pneumonia and tuberculosis occurred more frequently in the IFX-treated patients compared to those treated with MTX alone (16,19).

ETN is a genetically engineered protein comprising two molecules of the extracellular domain of TNF receptor II (p75) and the Fc portion of IgG1 (20). Due to its half-life of 3–5.5 days, ETN is administered subcutaneously (s.c), either on a weekly basis (50 mg) or twice a week (25 mg) (21).

The superiority of the combination therapy of ETN plus MTX over ETN or MTX monotherapy in patients with RA has been demonstrated (Table II) (22–24). The 2-year data from the TEMPO study confirmed that a larger proportion of patients treated with combination therapy exhibited clinical response compared to those receiving either ETN or MTX monotherapy (22). Moreover, the combination-treated patients had predominantly lower erosion change scores (−0.67) compared to patients treated with ETN (0.39) or MTX (3.25) alone (25). Therefore, treatment with a combination of ETN and MTX stopped joint damage and patients exhibited disease remission (25). Sustained efficacy and decreased rate of radiographic progression was observed in patients with early aggressive RA who underwent long-term treatment with ETN (26). Patients adopting combination therapy enhanced to a greater extent in function status compared to those in the monotherapy group (27). Additionally, ETN (50 mg) once weekly was an optimal treatment in most patients with RA. Increasing the dosage of ETN from 50 mg once a week to 50 mg twice a week in suboptimal responders did not markedly improve response rates (28). With regard to safety and efficacy, no obvious improvement was observed between ETN as monotherapy at 50 and 25 mg twice weekly (29).

Injection-site reactions and hypertension were more common with ETN compared to MTX or combination therapy (22). These events were mostly mild or moderate. Nausea and vomiting were more often associated with MTX compared to ETN or combination therapy. No statistically significant differences were observed in the groups regarding the incidence of serious adverse events (infectious and non-infectious) (22).

Thus, ETN is beneficial for patients with RA. However, the combination of ETN with MTX is superior to a monotherapy with each drug. The combination regimen may reduce disease activity, retard radiographic progression and improve function. Furthermore, the treatment with ETN plus MTX is well-tolerated and does not increase serious adverse events.

ADA is a monoclonal antibody of recombinant immunoglobulin (IgG1) containing only human sequences of peptides. It is an antagonist of TNF-α, which is able to prevent the binding of TNF-α to its receptors (6). It has a half-life of 10–20 days and may be used as monotherapy or in combination with several other DMARDs, preferably MTX (30–31). The recommended dose of ADA is 25 mg s.c. twice a week.

Treatment with ADA plus MTX was found to be statistically superior to PBO plus MTX, according to the ACR20/50/70 response rates at week 26 (Table III) (32). When receiving ADA plus MTX in early RA, patients exhibited rapid clinical and functional improvements (32). ADA regimens decreased the risk of radiographic disease progression (33). In a 5-year, open-label extension study, the addition of ADA led to greater inhibition of structural damage compared to patients who continued with MTX monotherapy (Table III) (34). The PREMIER study confirmed that treatment with ADA plus MTX initiated earlier contributed to higher improvements in clinical, functional and radiographic responses as compared to treatment with MTX or ADA alone (35). In addition, ADA plus MTX ameliorated physical function for patients with RA (33,36).

ADA exhibited a good overall tolerance. Findings of a previous study demonstrated that the rate of adverse events (serious and nonserious) was similar in the ADA and PBO groups, although the proportion of patients reporting serious infections was higher in patients receiving ADA (3.8%) compared to those receiving PBO (0.5%) (P<0.02), and was the highest in the patients administered 40 mg every other week (33). The common adverse events were injection site reactions and serious infections, such as military tuberculosis and cellulitis (35). However, ADA was safe and well-tolerated. Adverse events were not serious and severe side effects were relatively seldom.

GOLI is a human anti-TNF-α monoclonal antibody that is generated and matured in an in vivo system (37). GOLI has a high affinity and specificity for human TNF-α and effectively neutralizes TNF-α bioactivity in vitro (38).

The efficacy of GOLI has been demonstrated in several groups (Table IV) (37,39–40). The combination of GOLI and MTX was significantly better at improving the signs and symptoms of RA and physical function (37). No difference has been observed in the efficacy between the two GOLI dose groups (50 and 100 mg) (37). Compared individually with the PBO group, GOLI in combination with MTX in patients with RA showed greater clinical response, while the response rates did not show a clear dose-response pattern in the groups of GOLI plus MTX (Table IV) (39).

In the multicenter, randomized, PBO-controlled GO-FORWARD study, the mean improvement from baseline in HAQ-DI was significantly greater for GOLI 50 mg + MTX and 100 mg + MTX vs. PBO + MTX (41). However, GOLI + MTX also elicited a significant better response compared to PBO + MTX in various efficacy parameters, including disease activity score (DAS28) response. Additionally, the combination of GOLI and MTX limited radiographic progression (42).

The safety of GOLI was demonstrated in various trials (39–41). However, adverse events were reported in the process of treatment. The most frequent adverse events in the combined GOLI groups were nausea, headache and injection site reaction (39–41). Most events were mild or moderate (43). In general, GOLI in combination with MTX may alleviate the signs and symptoms of RA and improve physical function.

CZP is a humanized anti-TNF-α antibody with high affinity to TNF-α (44). In managing patients with RA, the recommended dose of CZP is 400 mg (given as two s.c. injections of 200 mg) initially and at weeks 2 and 4, followed by 200 mg every other week.

An international, multicenter, phase 3, randomized, double-blind, PBO-controlled study had assessed the efficacy of CZP in RA patients for MTX non-responders (45). Compared to PBO treatment, CZP plus MTX effectively reduced the signs and symptoms of RA, and inhibited progression of joint damage (Table V) (45–46). No evident differences were observed in clinical efficacy between the two CZP dose groups (45). Additionally, treatment with CZP monotherapy also provided a rapid, meaningful and durable clinical response and acceptable safety profile (Table V) (47). Increasing the CZP dose from 200 to 400 mg did not result in an additional benefit in RA patients (48). A study showed that the mean tender joint count (−24.8 vs. −24.6) or swollen joint count (−18.6 vs. −18.7) was similar between the dose-escalation (200 mg increased to 400 mg every other week) and stable-dose subgroups (400 mg every other week) (49). The most common adverse reactions were tuberculosis, injection site pain and injection site reaction (46).

As shown above, CZP monotherapy or the combination therapy with MTX as an effective treatment provides a rapid, meaningful and durable clinical response and an acceptable safety profile.

It is widely accepted that patients with RA have low quality of life. Clinical trials have shown that TNF-α blocking agents, such as ETN, IFX and ADA, relieve joint inflammations and slow the radiographic progression of joint damage and improve physical function in advanced RA (50–52). The availability of newer agents, including CZP and GOLI, has increased treatment options for patients with RA. Furthermore, anti-TNF-α agents are more efficacious in promoting the clinical signs and symptoms of RA compared to MTX alone. Anti-TNF-α agents plus MTX show sustained efficacy and remain more effective compared to anti-TNF-α monotherapy (53). Compared to MTX and PBO, the ACR20, 50 and 70 response rates for 1-year treatment with MTX plus any of the TNF inhibitors were 60 vs. 25%, 40 vs. 10% and 20 vs. 5%, respectively (54).

However, they have distinct clinical pharmacokinetic and -dynamic properties that must be considered when selecting one drug for therapy (55). For example, there are evident differences in the half-lives of the three agents (IFX, ETN and ADA), with ETN having the shortest (3–5.5 days) and ADA the longest (2 weeks) (21). The three types of biological agents also differ from each other in their dosing regimens (55). The larger, yet less frequently administered dose of IFX may result in higher peak serum concentrations compared to the smaller but more commonly administered dose of ETN and ADA, resulting in higher tissue concentrations (55). Total efficacies of varying biologics are highly similar, which have been observed in most of the studies and adopted by several investigators (55). Nevertheless, a recent study indicated significant differences in the efficacy of and adherence to therapy with ADA, ETN and IFX (56). IFX had the lowest treatment responses, disease remission and drug adherence rates. ADA had the highest treatment response and remission rate, while ETN had the longest drug survival rate (56) (Table VI).

In their study, Singh et al (57) demonstrated that patients administered ADA plus IFX were at a markedly higher risk compared to those administered PBO. Indirect companies showed that ADA had a higher tendency to withdraw compared to ETN (OR 1.89; 95% CI, 1.18–3.04) and ETN was less likely than IFX (OR 0.37; 95% CI, 0.19–0.70). Additionally, there seemed to be differences in the risk of tuberculosis (TB) among varying biologics, and this might influence the selection of patients likely to receive the biological agent. TB occurred more frequently in monoclonal antibodies-treated patients (i.e., IFX and ADA) compared to those treated with soluble TNF receptor therapy (i.e., ETN) (58–59). The rate of hospitalized infection in patients treated with other agents was lower compared to that for patients treated with IFX (60). Among these biological agents, the incidence of serious infections was higher in the CZP group compared to others. ADA, ETN and GOLI were associated with a low incidence of treatment discontinuation due to adverse events, whereas the IFX was not (61). Moreover, the biological agents increased the risk of infections. Consequently, patients with tuberculosis should be excluded and should receive pneumococcal, influenza and hepatitis B vaccinations prior to undergoing the therapy with biological agents.

Patients with RA may discontinue their initial drug and switch to a second anti-TNF-α agent due to shortage of drug efficacy. Regarding the effect of the second biological agent, in a retrospective study (62), certain patients (n=20) switched from ETN to IFX, while others (n=73) received IFX with no prior anti-TNF-α therapy. The C-reactive protein, swollen and tender joint counts as well as the morning stiffness ameliorated in the two groups, while no statistically significant difference was observed in the degree of benefit between the groups (62). However, IFX may be of additional clinical profit for patients with an incomplete response to ETN. In particular, patients receiving IFX exhibited a better amelioration in the HAQ score compared to those receiving ETN (Table VII) (63).

Another study concluded that patients switching to ADA exhibited a good clinical response when the therapy of IFX or ETN was ineffective (64). Patients who do not respond to an initial anti-TNF drug may also improve their HAQ score, subsequent to switching to a second agent (65). Patients with RA may be successfully treated with another TNF-α agent, especially those withdrawing due to inefficacy and adverse events (66). The above results demonstrated that switching among various biological agents was beneficial.

Biological agents render the treatment of RA a new era, especially for patients with an insufficient response to DMARDs. Biological agents can quickly relieve clinical symptoms and delay bone destruction. When the TNF-α inhibitors are applied in clinical practice, the combination with DMARDs are conductive to ease the symptoms and prevent bone structural damage and elevate physical function. Moreover, the conversion between various agents may have the same function. Certain drugs, such as ETN, in combination with MTX are better compared to monotherapy regarding long-term efficacy. Most adverse events of agents are infection-site reactions. Although severe side-effects may be treated appropriately, they still prevent clinical remedy. Physicians should prescribe various treatment regimens according to the patient’s symptoms as well as constantly explore the immune mechanism of RA, and develop novel biological agents. In the future, immunotherapy is likely to bring fundamental changes for patients with RA.