Introduction

Rheumatoid arthritis (RA) is systemic inflammatory disease, which usually proceeds to a chronic disease state and can cause considerable morbidity and loss of function and is associated with increased risk for mortality. There is consensus that treatment with disease-modifying anti-rheumatic drugs (DMARDs) is the best way to suppress the progression of RA. In addition, there is increasing evidence that early institution of DMARD therapy in patients with recent onset of RA increases the frequency of remission and retards joint destruction . To reach this goal, various interventions have been applied. The use of combinations of individual DMARDs in many , although not in all studies , yielded better results compared with DMARD monotherapy. Tight control of the treatment strategies aiming for a target such as remission with frequent treatment modifications (as in daily clinical practice) according to disease activity have also given considerably better results versus routine care . In 2010, the concept of ‘Treat to Target’ was published as being integral to the RA treatment paradigm. According to this, the best way to control the inflammation of RA is to start the treatment early and to adjust treatment every 1–3 months aiming for a target of remission. In the case of chronic RA, where remission may not be achievable, it is reasonable to aim for low disease activity as is clinically available. The tight control and targeted therapy principle was validated in the CAMERA (computer-assisted management in early rheumatoid arthritis) study on DMARD- and oral glucocorticoid-naive patients with early RA .

With the availability of several conventional DMARDs which have been used as monotherapy or in different combinations (including bridging therapy with glucocorticoids), and the 10-year experience on biologic therapies, it is time to reconsider what is the best treatment strategy for early RA. This topic has been recently reviewed and concepts have been incorporated in this chapter. The focus of this work is to highlight recent studies of patients with early untreated RA (duration of symptoms up to 2 years) who were DMARD- and systemic glucocorticoid-naive patients prior to being enrolled in clinical studies. Several fundamental studies have, in part, addressed the approach to treatment of early RA patients in general and will be briefly mentioned .

When to start therapy?

Currently, the European League Against Rheumatism (EULAR) recommends starting treatment in RA after the diagnosis . A meta-analysis by Finckh et al. showed that in studies and cohorts of early RA (defined as a disease duration of <2 years), a delay in the start of treatment has an impact on the progression of radiological damage. Patients with more aggressive disease seemed to benefit most from early initiation of DMARDs. A recent review also concluded that there is strong evidence for an association between symptom duration and radiographic progression. In addition, symptom duration was independently associated with DMARD-free sustained remission. When the treatment is started very early (within 12 weeks from symptom onset), the patients have better change to respond to the therapy and to achieve remission . Thus, we can extrapolate that the treatment should be started at best within 12 weeks from the onset of symptoms for optimal benefit.

• A.
  

  How to start?

  
  
• B.
  

  How to use conventional DMARDs?

  
  
• C.
  

  Monotherapy or combination therapy?

Table 1 presents studies where conventional DMARDs have been started in combinations or with glucocorticoids. The treatments have been modified either using tight control of disease activity or with preset targets versus a conventional treatment schedule.

The Finnish Rheumatoid Arthritis Combination Therapy (FIN-RACo) study was the first one to set the target to strict remission . In the study, 195 DMARD- and oral glucocorticoid-naive patients with recent onset were included to investigate whether combination therapy would be more effective than monotherapy in induction of remission. The patients were required to have early active disease with duration of symptoms <2 years, active disease with ≥3 swollen joints and at least three of the following: erythrocyte sedimentation rate (ESR) ≥28 mm h ⁻¹ or C-reactive protein (CRP) > 19 mg l ⁻¹ , morning stiffness ≥29 min, >5 swollen joints and <10 tender joints. The patients had active disease (Disease Activity Score (DAS28): 5.6), 70% were seropositive for rheumatoid factor (RF) and 48% had erosions. The median duration of symptoms was 6 months . The patients were treated from the diagnosis with a combination of three DMARDs (starting with methotrexate (MTX), sulphasalazine (SSZ) and hydroxychloroquine (HCQ)) and low-dose prednisolone (PRED) (the FIN-RACo strategy) or with DMARD monotherapy (starting with SSZ) with PRED used at the discretion of the treating physician. The target in both treatment arms was strict American College of Rheumatology (ACR) remission (modified from Pinals et al. ), where the fatigue criterion was omitted and no swollen and tender joints were allowed. Intra-articular (i.a.) glucocorticoids were part of the treatment protocol. The DMARDs were changed in the case of insufficient treatment response or in the case of intolerance. The treatment arms were continued for 2 years. After that, the use of combinations and monotherapy was not restricted to the original groups, the treatments still having remission as the target. The patients were monitored every 3 months. Patients in the FIN-RACo combination arm reached strict ACR remission significantly more often at 2 years (37%) than the patients starting with monotherapy (18%). Remission according to the DAS28 criteria was reached by 68% versus 41% of the patients . Despite the option to treat the patients with combinations freely after 2 years, the difference in the clinical outcomes remained up till 11 years . In addition, a sustained retardation of radiological progression up to 11 years was observed . By 5 years, none of the patients were on biologics , and by 11 years, 9% of the patients were on biologic treatment and 19% patients in the original FIN-RACo treatment arm and 14% in the single arm were without treatment .

Proudman et al. studied 82 patients with early (≤12 months of symptoms of arthritis) untreated RA with the question whether a regimen of MTX, cyclosporin and corticosteroids introduced at onset in poor-prognosis RA can produce a significant improvement in outcome compared with standard monotherapy with SSZ. The patients fulfilled the ACR criteria for RA. In addition, they had poor-prognosis disease (at least three of the following criteria present: CRP >20 mg l ⁻¹ , RF positivity, presence of the conserved sequence of DR β in the third allelic hypervariable region, abnormal findings in the Health Assessment Questionnaire (HAQ) and female sex). At baseline, the patients had a median duration of the disease of 8.5 months, 80% were seropositive and 62% had erosive disease. The mean DAS28 at baseline was 5.1–5.4 in the treatment arms, and the HAQ score was 1.4. The patients were randomised to monotherapy or combination therapy including active i.a. injections with glucocorticoids. The monotherapy arm was treated with SSZ (maximum dosage 3 g day ⁻¹ ). Clinically significant, painful joint effusions were treated with i.a. corticosteroid injections. The combination arm started with cyclosporin (maximum dosage 4.2 mg kg ⁻¹ ) + MTX (maximum dosage 20 mg week ⁻¹ after 8 weeks of study) + methylprednisolone injections i.a. in all active joints at study entry. During the study, clinically active joints could also be injected. An intramuscular (i.m.) injection of 120 mg methylprednisolone could be given as rescue therapy after 3 months. There was no switch to other DMARDs. Oral glucocorticoids were not allowed during the study. The 1-year results showed that there were no statistically significant differences between clinical or radiological outcomes. The treatment responses were modest. DAS28 and ACR remissions in the combination group were of the same magnitude as those in the single-arm treatment in the FIN-RACo study ( Table 1 ). The radiological progression during the first year was, however, quite slow, although higher than in CIMESTRA (Ciclosporine, Methotrexate, Steroid in RA), another study focussing on the use of i.a. glucocorticoids as part of the active treatment strategy. The comparison between the study by Proudman et al. and other studies on early RA patients is not, however, straightforward, because the Proudman study included only patients with markers of poor outcome.

The CIMESTRA study examined whether disease control can be achieved by treatment with MTX and i.a. betamethasone, and whether the addition of cyclosporine to the regimen has any additional effect. This study included 160 patients with early untreated RA (duration of active disease <6 months). The patients fulfilled the revised ACR criteria for RA and had active disease (DAS28: 5.4) with at least two swollen joints at study entry. About 47% had erosive disease, 65% were RF-positive and 59% were cyclic citrullinated peptide (CCP)-positive. The patients were randomised to MTX (dose increased to maximum 20 mg week ⁻¹ from week 8 if swollen joints were present) + cyclosporin (2.5 mg kg ⁻¹ ) or only MTX. Patients who did not reach the ACR20 response were switched to parenteral MTX for 3 months, followed by triple therapy (MTX + SSZ + HCQ) for 3 months and then were switched to a tumour necrosis-factor (TNF) inhibitor if disease activity persisted . Patients in both treatment arms received i.a. betamethasone at weeks 0, 2, 4 and 8 and thereafter every 4 weeks in all swollen joints (maximum four joints/visit) during the first year. Oral glucocorticoids were not allowed. At week 78, cyclosporin was tapered down. HCQ was added in all patients at week 68. The treatment outcome was very good in both treatment arms, with very limited radiological progression. There were no significant differences between the treatment groups with respect to the proportion of patients achieving remission and radiological progression during the 5-year follow-up period ( Table 1 ). By 5 years, 19% of the patients were receiving biologics and 16% were on sustained remission without treatment. The results of the CIMESTRA and FIN-RACo studies were very similar during the first 2 years ( Table 1 ). However, the 5-year rates of remission were numerically higher in the CIMESTRA study. The baseline characteristics of the patients in the two studies with respect to the disease activity and rates of erosive disease were roughly similar. Although the studies cannot be directly compared, there were differences between the duration of symptoms at entry (≤1 year in CIMESTRA vs. ≤2 years in the FIN-RACo study) and in the use of biologics in the CIMESTRA (19%) compared with the FIN-RACo study (0%) during the first 5 years.

The CAMERA study addressed the question whether intensive treatment with MTX according to a strict protocol and a computerised decision program is more beneficial compared with conventional treatment with MTX in early RA. The study compared a computerised system versus a conventional approach to aid in the decision of changing the dose of MTX in early RA . The study included 299 DMARD- and prednisone-naive patients with early RA (disease duration <1 year). Of the patients, 64% were positive for RF. They had active RA, with 14 swollen and 14.5 tender joints, the mean HAQ was 1.2 and the radiologic damage score (modified Total Sharp Score (TSS)) was 1.9. The treatment was started with MTX (starting at 7.5 mg week ⁻¹ ), and in the case the patient did not respond (≤20% improvement of number of swollen joints and ≤20% improvement in two out of three criteria: ESR, number of tender joints and general well-being by a visual analogue scale (VAS)), doses were increased by 5 mg week ⁻¹ up to 30 mg week ⁻¹ . The route of MTX could be switched to a subcutaneous (s.c.) one. In patients with inadequate response (≤50% improvement by the above-mentioned criteria), cyclosporin was added. In the case of no response after this treatment strategy, other DMARD(s) were started. Intra-articular and oral corticosteroids were avoided as far as possible. Those in the intensive group visited the outpatient department once a month for adjustment of the MTX dosage as guided by a computer algorithm that was based on clinical disease activity. Those in the conventional treatment group visited the outpatient department every 3 months and they were treated as per usual practice. The outcome was remission for at least 3 months, defined as: no swollen joints (0–38 joints) and at least two of the following: number of tender joints ≤3 (0–38 joints), ESR ≤20 mm h ⁻¹ and patient’s general well-being (VAS 0–100) ≤20. By 2 years, 39% of patients in the intensive treatment arm and 24% in the conventional treatment arm had discontinued the treatment, mostly due to adverse events. At 2 years, more patients in the intensive group compared with the conventional group (50% vs. 37%) had achieved at least one period of remission and had better clinical outcome in nearly all clinical parameters. The radiological progression was minimal, with no significant difference between the treatment arms. After 2 years, the treatments in both patient groups were no longer tightly controlled. At the 5-year visit , the mean disease activity was still low in both groups of patients, but some radiological progression had developed ( Table 1 ). The positive effect of initial tight control of the MTX dose was diluted and there were no significant difference for clinical and radiographic outcomes between the original study groups.

The CAMERAII study investigated whether the addition of 10 mg day ⁻¹ prednisone to the CAMERA-protocol-driven computerised treatment algorithm in patients with early RA would result in few joint erosions. Compared with CAMERA, the CAMERAII study was a randomised, placebo-controlled, double-blind study run for 2 years. The patients visited the rheumatologist monthly and the treatment was guided by predefined criteria, aiming to remission defined as in the original CAMERA study . The primary ‘end’ point (erosions) was accomplished. In both treatment groups, the radiological progression was minimal, but those patients on prednisone had even less progression compared with the placebo group (median erosion scores 0 (interquartile range (IQR), 0–0) and 0 (IQR, 0–2), respectively, p = 0.022). Seventy-eight percent of the patients in the MTX + prednisone arm and 67% of those in the MTX + placebo arm had no erosions at the 2-year visit. There were no significant differences in the clinical outcomes at 2 years. However, in the group treated with prednisone, remission was reached earlier, function as measured by the HAQ was better, fewer patients needed the switch from oral to s.c. MTX and had fewer patients needed to start a biologic drug compared to the placebo-treated patients. The study-defined remission criterion was achieved by a higher number of the patients and DAS28 was lower compared with the original CAMERA study .

In conclusion, triple combination of conventional DMARDs, starting the treatment within the first year of symptoms, active use of i.a. glucocorticoid injections or oral prednisone added to a DMARD and frequent monitoring with defined criteria for dose adjustments all contribute to reaching the target (remission) more often.

When to start therapy?

Currently, the European League Against Rheumatism (EULAR) recommends starting treatment in RA after the diagnosis . A meta-analysis by Finckh et al. showed that in studies and cohorts of early RA (defined as a disease duration of <2 years), a delay in the start of treatment has an impact on the progression of radiological damage. Patients with more aggressive disease seemed to benefit most from early initiation of DMARDs. A recent review also concluded that there is strong evidence for an association between symptom duration and radiographic progression. In addition, symptom duration was independently associated with DMARD-free sustained remission. When the treatment is started very early (within 12 weeks from symptom onset), the patients have better change to respond to the therapy and to achieve remission . Thus, we can extrapolate that the treatment should be started at best within 12 weeks from the onset of symptoms for optimal benefit.

• A.
  

  How to start?

  
  
• B.
  

  How to use conventional DMARDs?

  
  
• C.
  

  Monotherapy or combination therapy?

Table 1 presents studies where conventional DMARDs have been started in combinations or with glucocorticoids. The treatments have been modified either using tight control of disease activity or with preset targets versus a conventional treatment schedule.

The Finnish Rheumatoid Arthritis Combination Therapy (FIN-RACo) study was the first one to set the target to strict remission . In the study, 195 DMARD- and oral glucocorticoid-naive patients with recent onset were included to investigate whether combination therapy would be more effective than monotherapy in induction of remission. The patients were required to have early active disease with duration of symptoms <2 years, active disease with ≥3 swollen joints and at least three of the following: erythrocyte sedimentation rate (ESR) ≥28 mm h ⁻¹ or C-reactive protein (CRP) > 19 mg l ⁻¹ , morning stiffness ≥29 min, >5 swollen joints and <10 tender joints. The patients had active disease (Disease Activity Score (DAS28): 5.6), 70% were seropositive for rheumatoid factor (RF) and 48% had erosions. The median duration of symptoms was 6 months . The patients were treated from the diagnosis with a combination of three DMARDs (starting with methotrexate (MTX), sulphasalazine (SSZ) and hydroxychloroquine (HCQ)) and low-dose prednisolone (PRED) (the FIN-RACo strategy) or with DMARD monotherapy (starting with SSZ) with PRED used at the discretion of the treating physician. The target in both treatment arms was strict American College of Rheumatology (ACR) remission (modified from Pinals et al. ), where the fatigue criterion was omitted and no swollen and tender joints were allowed. Intra-articular (i.a.) glucocorticoids were part of the treatment protocol. The DMARDs were changed in the case of insufficient treatment response or in the case of intolerance. The treatment arms were continued for 2 years. After that, the use of combinations and monotherapy was not restricted to the original groups, the treatments still having remission as the target. The patients were monitored every 3 months. Patients in the FIN-RACo combination arm reached strict ACR remission significantly more often at 2 years (37%) than the patients starting with monotherapy (18%). Remission according to the DAS28 criteria was reached by 68% versus 41% of the patients . Despite the option to treat the patients with combinations freely after 2 years, the difference in the clinical outcomes remained up till 11 years . In addition, a sustained retardation of radiological progression up to 11 years was observed . By 5 years, none of the patients were on biologics , and by 11 years, 9% of the patients were on biologic treatment and 19% patients in the original FIN-RACo treatment arm and 14% in the single arm were without treatment .

Proudman et al. studied 82 patients with early (≤12 months of symptoms of arthritis) untreated RA with the question whether a regimen of MTX, cyclosporin and corticosteroids introduced at onset in poor-prognosis RA can produce a significant improvement in outcome compared with standard monotherapy with SSZ. The patients fulfilled the ACR criteria for RA. In addition, they had poor-prognosis disease (at least three of the following criteria present: CRP >20 mg l ⁻¹ , RF positivity, presence of the conserved sequence of DR β in the third allelic hypervariable region, abnormal findings in the Health Assessment Questionnaire (HAQ) and female sex). At baseline, the patients had a median duration of the disease of 8.5 months, 80% were seropositive and 62% had erosive disease. The mean DAS28 at baseline was 5.1–5.4 in the treatment arms, and the HAQ score was 1.4. The patients were randomised to monotherapy or combination therapy including active i.a. injections with glucocorticoids. The monotherapy arm was treated with SSZ (maximum dosage 3 g day ⁻¹ ). Clinically significant, painful joint effusions were treated with i.a. corticosteroid injections. The combination arm started with cyclosporin (maximum dosage 4.2 mg kg ⁻¹ ) + MTX (maximum dosage 20 mg week ⁻¹ after 8 weeks of study) + methylprednisolone injections i.a. in all active joints at study entry. During the study, clinically active joints could also be injected. An intramuscular (i.m.) injection of 120 mg methylprednisolone could be given as rescue therapy after 3 months. There was no switch to other DMARDs. Oral glucocorticoids were not allowed during the study. The 1-year results showed that there were no statistically significant differences between clinical or radiological outcomes. The treatment responses were modest. DAS28 and ACR remissions in the combination group were of the same magnitude as those in the single-arm treatment in the FIN-RACo study ( Table 1 ). The radiological progression during the first year was, however, quite slow, although higher than in CIMESTRA (Ciclosporine, Methotrexate, Steroid in RA), another study focussing on the use of i.a. glucocorticoids as part of the active treatment strategy. The comparison between the study by Proudman et al. and other studies on early RA patients is not, however, straightforward, because the Proudman study included only patients with markers of poor outcome.

The CIMESTRA study examined whether disease control can be achieved by treatment with MTX and i.a. betamethasone, and whether the addition of cyclosporine to the regimen has any additional effect. This study included 160 patients with early untreated RA (duration of active disease <6 months). The patients fulfilled the revised ACR criteria for RA and had active disease (DAS28: 5.4) with at least two swollen joints at study entry. About 47% had erosive disease, 65% were RF-positive and 59% were cyclic citrullinated peptide (CCP)-positive. The patients were randomised to MTX (dose increased to maximum 20 mg week ⁻¹ from week 8 if swollen joints were present) + cyclosporin (2.5 mg kg ⁻¹ ) or only MTX. Patients who did not reach the ACR20 response were switched to parenteral MTX for 3 months, followed by triple therapy (MTX + SSZ + HCQ) for 3 months and then were switched to a tumour necrosis-factor (TNF) inhibitor if disease activity persisted . Patients in both treatment arms received i.a. betamethasone at weeks 0, 2, 4 and 8 and thereafter every 4 weeks in all swollen joints (maximum four joints/visit) during the first year. Oral glucocorticoids were not allowed. At week 78, cyclosporin was tapered down. HCQ was added in all patients at week 68. The treatment outcome was very good in both treatment arms, with very limited radiological progression. There were no significant differences between the treatment groups with respect to the proportion of patients achieving remission and radiological progression during the 5-year follow-up period ( Table 1 ). By 5 years, 19% of the patients were receiving biologics and 16% were on sustained remission without treatment. The results of the CIMESTRA and FIN-RACo studies were very similar during the first 2 years ( Table 1 ). However, the 5-year rates of remission were numerically higher in the CIMESTRA study. The baseline characteristics of the patients in the two studies with respect to the disease activity and rates of erosive disease were roughly similar. Although the studies cannot be directly compared, there were differences between the duration of symptoms at entry (≤1 year in CIMESTRA vs. ≤2 years in the FIN-RACo study) and in the use of biologics in the CIMESTRA (19%) compared with the FIN-RACo study (0%) during the first 5 years.

The CAMERA study addressed the question whether intensive treatment with MTX according to a strict protocol and a computerised decision program is more beneficial compared with conventional treatment with MTX in early RA. The study compared a computerised system versus a conventional approach to aid in the decision of changing the dose of MTX in early RA . The study included 299 DMARD- and prednisone-naive patients with early RA (disease duration <1 year). Of the patients, 64% were positive for RF. They had active RA, with 14 swollen and 14.5 tender joints, the mean HAQ was 1.2 and the radiologic damage score (modified Total Sharp Score (TSS)) was 1.9. The treatment was started with MTX (starting at 7.5 mg week ⁻¹ ), and in the case the patient did not respond (≤20% improvement of number of swollen joints and ≤20% improvement in two out of three criteria: ESR, number of tender joints and general well-being by a visual analogue scale (VAS)), doses were increased by 5 mg week ⁻¹ up to 30 mg week ⁻¹ . The route of MTX could be switched to a subcutaneous (s.c.) one. In patients with inadequate response (≤50% improvement by the above-mentioned criteria), cyclosporin was added. In the case of no response after this treatment strategy, other DMARD(s) were started. Intra-articular and oral corticosteroids were avoided as far as possible. Those in the intensive group visited the outpatient department once a month for adjustment of the MTX dosage as guided by a computer algorithm that was based on clinical disease activity. Those in the conventional treatment group visited the outpatient department every 3 months and they were treated as per usual practice. The outcome was remission for at least 3 months, defined as: no swollen joints (0–38 joints) and at least two of the following: number of tender joints ≤3 (0–38 joints), ESR ≤20 mm h ⁻¹ and patient’s general well-being (VAS 0–100) ≤20. By 2 years, 39% of patients in the intensive treatment arm and 24% in the conventional treatment arm had discontinued the treatment, mostly due to adverse events. At 2 years, more patients in the intensive group compared with the conventional group (50% vs. 37%) had achieved at least one period of remission and had better clinical outcome in nearly all clinical parameters. The radiological progression was minimal, with no significant difference between the treatment arms. After 2 years, the treatments in both patient groups were no longer tightly controlled. At the 5-year visit , the mean disease activity was still low in both groups of patients, but some radiological progression had developed ( Table 1 ). The positive effect of initial tight control of the MTX dose was diluted and there were no significant difference for clinical and radiographic outcomes between the original study groups.

The CAMERAII study investigated whether the addition of 10 mg day ⁻¹ prednisone to the CAMERA-protocol-driven computerised treatment algorithm in patients with early RA would result in few joint erosions. Compared with CAMERA, the CAMERAII study was a randomised, placebo-controlled, double-blind study run for 2 years. The patients visited the rheumatologist monthly and the treatment was guided by predefined criteria, aiming to remission defined as in the original CAMERA study . The primary ‘end’ point (erosions) was accomplished. In both treatment groups, the radiological progression was minimal, but those patients on prednisone had even less progression compared with the placebo group (median erosion scores 0 (interquartile range (IQR), 0–0) and 0 (IQR, 0–2), respectively, p = 0.022). Seventy-eight percent of the patients in the MTX + prednisone arm and 67% of those in the MTX + placebo arm had no erosions at the 2-year visit. There were no significant differences in the clinical outcomes at 2 years. However, in the group treated with prednisone, remission was reached earlier, function as measured by the HAQ was better, fewer patients needed the switch from oral to s.c. MTX and had fewer patients needed to start a biologic drug compared to the placebo-treated patients. The study-defined remission criterion was achieved by a higher number of the patients and DAS28 was lower compared with the original CAMERA study .

In conclusion, triple combination of conventional DMARDs, starting the treatment within the first year of symptoms, active use of i.a. glucocorticoid injections or oral prednisone added to a DMARD and frequent monitoring with defined criteria for dose adjustments all contribute to reaching the target (remission) more often.