Clinical remission and/or minimal disease activity in patients receiving adalimumab treatment in a multinational, open-label, twelve-week study

Authors

  • Gerd R. Burmester,

    Corresponding author
    1. Charité University Medicine Berlin, Berlin, Germany
    • Department of Rheumatology and Clinical Immunology, Charité University Medicine Berlin, Charitéplatz 1, 10117 Berlin, Germany
    Search for more papers by this author
    • Dr. Burmester has received consultant fees, speaking fees, and honoraria (more than $10,000 each) from Abbott, Schering-Plough, Wyeth, and Roche. Dr. Ferraccioli has received consultancy fees and speaking fees (less than $10,000 each) from Abbott, Wyeth, and Schering-Plough. Dr. Unnebrink holds stock options in Abbott. Dr. Kary is a contractor for Abbott (Germany). Dr. Kupper holds stock options in Abbott.

  • Gianfranco Ferraccioli,

    1. Catholic University of Rome, Rome, Italy
    Search for more papers by this author
    • Dr. Burmester has received consultant fees, speaking fees, and honoraria (more than $10,000 each) from Abbott, Schering-Plough, Wyeth, and Roche. Dr. Ferraccioli has received consultancy fees and speaking fees (less than $10,000 each) from Abbott, Wyeth, and Schering-Plough. Dr. Unnebrink holds stock options in Abbott. Dr. Kary is a contractor for Abbott (Germany). Dr. Kupper holds stock options in Abbott.

  • René-Marc Flipo,

    1. University Hospital, Lille, France
    Search for more papers by this author
  • Indalecio Monteagudo-Sáez,

    1. Hospital General Universitario Gregorio Marañón, Madrid, Spain
    Search for more papers by this author
  • Kristina Unnebrink,

    1. Abbott GmbH & Co. KG, Ludwigshafen, Germany
    Search for more papers by this author
    • Dr. Burmester has received consultant fees, speaking fees, and honoraria (more than $10,000 each) from Abbott, Schering-Plough, Wyeth, and Roche. Dr. Ferraccioli has received consultancy fees and speaking fees (less than $10,000 each) from Abbott, Wyeth, and Schering-Plough. Dr. Unnebrink holds stock options in Abbott. Dr. Kary is a contractor for Abbott (Germany). Dr. Kupper holds stock options in Abbott.

  • Sonja Kary,

    1. Abbott GmbH & Co. KG, Ludwigshafen, Germany
    Search for more papers by this author
    • Dr. Burmester has received consultant fees, speaking fees, and honoraria (more than $10,000 each) from Abbott, Schering-Plough, Wyeth, and Roche. Dr. Ferraccioli has received consultancy fees and speaking fees (less than $10,000 each) from Abbott, Wyeth, and Schering-Plough. Dr. Unnebrink holds stock options in Abbott. Dr. Kary is a contractor for Abbott (Germany). Dr. Kupper holds stock options in Abbott.

  • Hartmut Kupper

    1. Abbott GmbH & Co. KG, Ludwigshafen, Germany
    Search for more papers by this author
    • Dr. Burmester has received consultant fees, speaking fees, and honoraria (more than $10,000 each) from Abbott, Schering-Plough, Wyeth, and Roche. Dr. Ferraccioli has received consultancy fees and speaking fees (less than $10,000 each) from Abbott, Wyeth, and Schering-Plough. Dr. Unnebrink holds stock options in Abbott. Dr. Kary is a contractor for Abbott (Germany). Dr. Kupper holds stock options in Abbott.


  • ClinicalTrials.gov identifier: NCT00448383.

Abstract

Objective

To evaluate the effect of adalimumab treatment on clinical remission and/or minimal disease activity (MDA) in 6,610 patients with active rheumatoid arthritis (RA) who were enrolled in the Research in Active RA trial, a multinational, open-label, 12-week study with an optional extension period.

Methods

Clinical remission was defined as a Disease Activity Score in 28 joints (DAS28) <2.6, Simplified Disease Activity Index (SDAI) score ≤3.3, or Clinical Disease Activity Index (CDAI) score ≤2.8. MDA required absence of tender and swollen joints plus erythrocyte sedimentation rate (ESR) ≤10 mm/hour; DAS28 score ≤2.85; or achievement of 5 of 7 core criteria for pain, swollen/tender joints, physical function, physician/patient global assessment, and ESR. Time to and time in remission/MDA and response predictors were analyzed using Kaplan-Meier estimates and Cox proportional hazards regression analysis, respectively.

Results

A total of 38%, 24%, and 27% of patients achieved remission defined as DAS28 <2.6, SDAI ≤3.3, and CDAI ≤2.8, respectively. MDA was observed in 45% of patients by DAS28 ≤2.85, in 43% by the core set of criteria, and in 13% by absence of tender/swollen joints plus ESR ≤10 mm/hour. Median times in continuous remission and MDA were 3.4 and 4.4 months, respectively. Predictors for remission (DAS28 <2.6) and MDA (DAS28 ≤2.85) were male sex; younger age; concomitant disease-modifying antirheumatic drug use; lower baseline DAS28, CRP concentration, and Health Assessment Questionnaire disease index score; ≤1 comorbidity; and tumor necrosis factor antagonist naivety.

Conclusion

During adalimumab treatment, 25% of patients experienced clinical remission and nearly half achieved MDA. To our knowledge, this analysis represents the largest prospective clinical trial data set to be assessed using Outcome Measures in Rheumatology Clinical Trials MDA criteria.

INTRODUCTION

Since the introduction of biologic response modifiers that inhibit tumor necrosis factor (TNF), the expectations of medical treatment for active rheumatoid arthritis (RA) have changed markedly. Several clinical trials have demonstrated the efficacy of TNF antagonists (adalimumab, etanercept, and infliximab) in significantly reducing disease activity and limiting progression of joint destruction and subsequent disability in patients with RA (1–4). Appropriate management of RA requires more than the assessment of relative improvement or change in signs and symptoms. Being able to assess the actual state of disease activity is important in making informed treatment decisions (5–9).

The general therapeutic goal in RA is to obtain a state of low disease activity, or even clinical remission, thereby preventing further joint destruction with optimal treatment. Historically, several criteria for assessing clinical remission have been developed (10, 11). The definition and therapeutic usefulness of clinical remission or of low disease activity/minimal disease activity (MDA) is complex and recently has been the subject of intense discussion (12–19). One well-established definition is achievement of a disease activity score <2.6 based on the erythrocyte sedimentation rate (ESR) and Disease Activity Score in 28 joints (DAS28) assessment (5, 20, 21). Because the DAS28 threshold can be achieved by patients who may be quite symptomatic in the few joints that are swollen, a more restrictive DAS28 score limit of <2.4 has been proposed but not widely accepted (22, 23). Furthermore, 2 other criteria for identifying clinical remission, Simplified Disease Activity Index (SDAI) and Clinical Disease Activity Index (CDAI) cut points of ≤3.3 and ≤2.8, respectively, are also being investigated and applied in clinical practice (22, 24, 25).

The concept of MDA, which aims for an acceptable state of RA that is near remission and is adequate for many patients, has been developed by the Outcome Measures in Rheumatology Clinical Trials (OMERACT) MDA working group (26). By definition, the OMERACT MDA criteria include clinical remission as well as a state of low disease activity. In a stepwise process, the proposed OMERACT definition for identifying MDA first considers all patients who have a tender joint count (TJC) of 0 and a swollen joint count (SJC) of 0 plus an ESR ≤10 mm/hour; this state of RA is, in fact, clinical remission. In the second step, patients are considered in MDA either when a DAS28 score ≤2.85 is achieved or, alternatively, when 5 of the 7 following core criteria are met: pain score ≤2 on a 0–10 visual analog scale (VAS), SJC ≤1 in a 28-joint assessment, TJC ≤1 in a 28-joint assessment, Health Assessment Questionnaire (HAQ) disability index (DI) score ≤0.5 on a 0–3 scale, physician global assessment of disease activity ≤1.5 on a 0–10 VAS, patient global assessment of disease activity ≤2 on a 0–10 VAS, and ESR ≤20 mm/hour.

The recently completed Research in Active RA (ReAct) trial provides a large database of 6,610 patients who received adalimumab, a fully human monoclonal antibody against TNF, in an open-label study designed to mimic rheumatologic care in typical clinical practices (27, 28). We analyzed the percentage of patients in the ReAct trial who achieved remission or MDA by different scores, the time until remission or MDA after the initiation of adalimumab treatment, and the time in sustained remission or MDA. Also, we investigated possible predictors for achievement and maintenance of clinical remission and MDA.

MATERIALS AND METHODS

Ethics.

The ReAct trial was conducted in accordance with principles of the Declaration of Helsinki and the International Conference on Harmonization Guidelines for Good Clinical Practice, and compliance with all local laws and customs was ensured by the investigators. Written informed consent was obtained from each participant before initiation of any study-related procedures.

Patients.

The study was conducted in adult patients with RA as defined by the 1987 revised criteria of the American College of Rheumatology (formerly the American Rheumatism Association) (29). Eligible patients had RA for ≥3 months; a DAS28 score ≥3.2, indicating at least moderate disease activity (20); and treatment failure with at least 1 traditional disease-modifying antirheumatic drug (DMARD). Patients were allowed to continue their current antirheumatic therapy if the treatment regimen was not modified until week 12. Previous or concurrent treatment with alkylating agents was not allowed.

Study design.

All patients self-administered a subcutaneous injection of adalimumab 40 mg (Abbott Laboratories, Abbott Park, IL) every other week during this 12-week, open-label, multicenter study. After the initial 12-week phase, patients could enter an optional extension until adalimumab became commercially available. Subsequently, patients could also enter an ongoing postmarketing observational study for up to 5 years. Patients who had received adalimumab without combination DMARD use and did not achieve at least a moderate response according to the European League Against Rheumatism response criteria (30) at week 12 could change to weekly injections of adalimumab. Efficacy and routine safety evaluations were conducted at weeks 2, 6, and 12, and every 8 weeks thereafter.

Statistical analysis.

Clinical remission was defined as achieving a DAS28 score <2.6, an SDAI score ≤3.3, or a CDAI score ≤2.8. The SDAI consists of the sum of the number of tender and swollen joints, a global assessment of disease activity by both the patient and the clinician (10-cm VAS), and C-reactive protein (CRP) concentration in mg/dl (22). The CDAI refers to the same clinical variables as the SDAI but omits the CRP component.

MDA was evaluated using the preliminary OMERACT definition as described by Wells and colleagues (26), although we did not perform the analyses in 2 consecutive steps. All patients were entered into the evaluation of actual clinical remission, defined as no tender or swollen joint counts and an ESR <10 mm/hour. Irrespective of the result of this analysis, all patients were also included in the analyses of MDA by either a DAS28 score ≤2.85 or 5 of the 7 core set criteria as described earlier.

All measures were assessed at week 12 and at the last individual observation for each patient, irrespective of the point in time when the patient discontinued the study. In addition, outcome criteria were assessed at any individual time point during the treatment period. The percentage of patients who fulfilled each outcome criterion was calculated using the total number of patients who had the required data available at the defined time points (observed values) as the denominator. In addition, the percentage of patients achieving remission or MDA at any time during treatment was stratified by baseline patient characteristics and treatment conditions.

Analyses of time to remission and time to MDA and of time in remission and time in MDA (in patients who had achieved remission) were limited to the respective DAS28 criteria. Kaplan-Meier survival estimates were used to evaluate the median time from the first adalimumab injection to remission and to MDA. If remission or MDA had not been achieved at the time of the last visit, the observation was censored in the Kaplan-Meier analysis. For patients who reached remission, both the time in remission and the time in at least MDA after remission were analyzed using Kaplan-Meier methods. The time in remission and the time in at least MDA were calculated from time of remission to the first visit where remission (or MDA) was lost (the “event” in the Kaplan-Meier analysis) or to the last available visit. If remission (or MDA) was ongoing at the time of the last visit, the observation was censored in the Kaplan-Meier analysis.

Predictors of achievement and maintenance of remission or MDA were identified using Cox proportional hazards regression analysis. Again, remission and MDA for this analysis were defined by the DAS28 criterion exclusively. Baseline characteristics entered into the model as possible predictors were DAS28 score (≤5.1, >5.1–7, >7), CRP concentration (<20, ≥20 mg/liter), physical function (HAQ DI score <1, 1–<1.5, 1.5–<2, ≥2), rheumatoid factor (RF; positive/negative), age (<40 years, 40–≤65 years, >65–≤75 years, >75 years), sex (male/female), smoking history (never/ever), duration of RA (≤1 year, >1–2 years, >2–5 years, >5–10 years, >10 years), number of previously failed DMARDs (1, 2, 3, 4, ≥5), history of treatment with etanercept and/or infliximab (yes/no), concomitant use of DMARDs (yes/no), concomitant use of glucocorticoids (yes/no), number of comorbidities (≤1, >1), and time to remission (for sustained remission/MDA only) (31–35). Finally, in a descriptive analysis, the DAS28 values for patients who achieved remission were followed up to the last observation of each patient at study discontinuation.

RESULTS

A total of 6,610 patients were enrolled in the ReAct trial: 81% were female, 73% were RF positive, 74% were taking at least 1 concomitant DMARD, and 71% were receiving glucocorticoids. Additional baseline characteristics included a mean age of 54 years, mean disease duration of 11 years, mean DAS28 score of 6.0, and mean HAQ DI score of 1.64 (27). During the complete adalimumab treatment period, the mean exposure to adalimumab was 33.3 weeks (maximum of 95.6 weeks), resulting in 4,210 patient-years of adalimumab exposure. At week 12, 93% of enrolled patients continued in the study. The overall withdrawal rate was 20.8%, with 6.8% of patients withdrawing due to lack of effectiveness (27). The number of patients in the ReAct trial gradually decreased as adalimumab was approved for commercial use in participating countries. After 1 year, 1,257 patients remained in the study. The baseline characteristics of patients who discontinued the study before week 52 were very similar to the baseline characteristics of patients who remained in the study after 1 year. Statistically significant but clinically irrelevant differences between the ≥1-year treatment subset versus the <1-year treatment subset were concomitant DMARD use (77% versus 73%), HAQ DI score (1.7 versus 1.6), mean number of previous DMARDs (3.1 versus 2.7), mean duration of RA (11 years versus 10 years), and mean age (54 years versus 53 years).

The median DAS28 score of all enrolled patients at their last visit was 3.6; the first quartile DAS28 score at the last visit was 2.6, indicating that one-quarter of all patients discontinued the study nearly in remission. In the extension phase, 414 (6%) of 6,610 patients, most of whom were receiving adalimumab monotherapy, underwent dose escalation to adalimumab 40 mg weekly. For these patients, the first quartile DAS28 score 16 weeks after dose increase was 3.4, indicating that only some of these patients experienced remission/MDA after dose increase.

Clinical remission and MDA at various time points.

At week 12, a total of 1,187 (20%) of 6,051 patients were in remission as measured by a DAS28 score <2.6, and 1,494 (25%) of 6,051 were in a state of MDA as measured by a DAS28 score ≤2.85. Compared with week 12, higher percentages of patients achieved both definitions of response at the last observation (25% and 31%, respectively) and at any time during adalimumab treatment (38% and 45%, respectively) (Figure 1).

Figure 1.

Clinical remission and minimal disease activity (MDA) at various time points in the Research in Active Rheumatoid Arthritis (ReAct) trial. A, Percentage of patients achieving clinical remission measured by Disease Activity Score in 28 joints (DAS28), Simplified Disease Activity Index (SDAI), and Clinical Disease Activity Index (CDAI). B, Percentage of patients achieving MDA measured by Outcome Measures in Rheumatology Clinical Trials criteria. Open bars = week 12; shaded bars = last observation; solid bars = any time. TJC = tender joint count; SJC = swollen joint count; ESR = erythrocyte sedimentation rate.

With respect to clinical remission by the SDAI score ≤3.3 and CDAI score ≤2.8 definitions, 11% of patients met the threshold for each measure of remission at week 12. Similar to the DAS28 results, the percentages of patients meeting the SDAI and CDAI remission criteria increased at the last observation and at any time during the study (Figure 1A). MDA was observed in 43% of patients according to the core set of criteria at any time during adalimumab treatment, compared with 24% and 30% at week 12 and the last observation, respectively. Absence of tender/swollen joints plus ESR ≤10 mm/hour resulted in the lowest percentage of patients achieving MDA at all time points (5% of all patients at week 12, 7% of all patients at the last observation, and 13% at any time) (Figure 1B). In addition, the remission and MDA data as achieved at any time during the study were stratified by baseline patient and treatment characteristics (Table 1). Irrespective of the assessment method, patients with lower baseline DAS28 score, HAQ DI score, CRP concentration, age, and number of previous DMARDs and who were TNF antagonist naive and RF negative were more likely to achieve remission/MDA. These univariate analyses were then scrutinized using the Cox proportional hazards regression model.

Table 1. Observed percentages of patients in remission or in minimal disease activity measured by various scores at any time during adalimumab treatment*
SubgroupClinical remissionMinimal disease activity
DAS28 <2.6SDAI ≤3.3CDAI ≤2.8No TJC or SJC and ESR ≤10 mm/hourDAS28 ≤2.855 of 7 core criteria
  • *

    DAS28 = Disease Activity Score in 28 joints; SDAI = Simplified Disease Activity Index; CDAI = Clinical Disease Activity Index; TJC = tender joint count; SJC = swollen joint count; ESR = erythrocyte sedimentation rate; HAQ DI = Health Assessment Questionnaire disability index; DMARD = disease-modifying antirheumatic drug; ACR50 = American College of Rheumatology 50% improvement criteria (44).

  • Fulfillment of 5 of the following 7 core criteria: pain score ≤2 (0–10 visual analog scale), swollen joint count ≤1 (0–28 joints), tender joint count ≤1 (0–28 joints), HAQ DI score ≤0.5 (0–3), physician global assessment of disease activity ≤1.5 (0–10), patient global assessment of disease activity ≤2 (0–10 visual analog scale), and ESR ≤20 mm/hour.

  • n = number of patients with available DAS28 data; other subgroups have slightly different numbers of patients and are not shown but can be provided upon request.

  • §

    Includes valvular heart disease, congestive heart failure, ischemic heart disease, hypertension, liver disease, peptic ulcer disease, kidney disorder, diabetes mellitus, hyperlipidemia, hypothyroidism, osteoarthritis, osteoporosis, cerebral vascular accident, depression, asthma, emphysema/chronic obstructive pulmonary disease, past tuberculosis, past pneumonia, and drug allergies/reactions.

All patients (n = 6,473)382427134543
Baseline DAS28 score      
 ≤5.1 (n = 1,266)633739257262
 >5.1 (n = 5,207)322124113938
 >7.0 (n = 1,275)18141662426
Baseline C-reactive protein, mg/liter      
 <20 (n = 3,837)442829165146
 ≥20 (n = 2,562)291923103738
Baseline HAQ DI score, 0–3      
 <1 (n = 1,092)584243236567
 1–<1.5 (n = 1,334)473134165454
 1.5–<2 (n = 1,711)372224134441
 ≥2 (n = 2,313)25141783227
Previous DMARD use, no.      
 ≤3 (n = 4,068)422831164948
 >3 (n = 2,210)311719103434
Previous etanercept and/or infliximab use      
 Yes (n = 878)26131573127
 No (n = 5,595)402629144845
Duration of rheumatoid arthritis, years      
 ≤5 (n = 1,973)433033174948
 >5 (n = 4,466)362224124440
Rheumatoid factor      
 Positive (n = 4,714)372326124442
 Negative (n = 1,748)422729174944
Age, years      
 <40 (n = 980)513437235857
 40–≤65 (n = 4,046)392426134643
 >65–≤75 (n = 1,222)29202183635
 >75 (n = 225)26171972827
Sex      
 Male (n = 1,246)462730195248
 Female (n = 5,227)362426124442
Comorbidities, no.§      
 ≤1 (n = 3,991)432730165048
 >1 (n = 2,482)31202193835
Concomitant DMARD use, no.      
 0 (n = 1,672)28182093534
 ≥1 (n = 4,801)422729154946
ACR50 response at week 6      
 Yes (n = 1,699)634954267274
 No (n = 4,345)29141793632

Time to clinical remission or time to MDA and predictors for response.

For all patients enrolled, the Kaplan-Meier estimates were 10.4 months (95% confidence interval [95% CI] 10.1–11.7) for median time to remission and 8.2 months (95% CI 7.2–8.3) for MDA (Figure 2). In the 38% and 45% of all patients who achieved remission or MDA during adalimumab therapy, the median time to response was 2.8 months for both measures (mean 3.7 months and 3.4 months, respectively). We identified the following predictors for clinical remission response by Cox regression: lower baseline HAQ DI score, lower baseline DAS28 score, younger age, male sex, concomitant DMARD treatment, lower CRP concentration, ≤1 comorbidity, and no previous TNF antagonist treatment (Table 2). The predictors for MDA were identical with the addition of a weak positive impact of never smoking and fewer previous DMARDs (data not shown).

Figure 2.

Inverted Kaplan-Meier curves of time to first occurrence of clinical remission (Disease Activity Score in 28 joints [DAS28] <2.6; median 10.4 months) and of minimal disease activity (MDA; DAS28 score ≤2.85; median 8.2 months). * Number of patients still under observation excluding those who already achieved remission. + Number of patients still under observation excluding those who already achieved MDA. Solid line = time to MDA; broken line = time to remission.

Table 2. Predictors of time to clinical remission (DAS28 score <2.6)*
Possible confounderHazard ratio95% CIP
  • *

    95% CI = 95% confidence interval; see Table 1 for additional definitions.

  • Overall P value for all subcategories.

  • Includes valvular heart disease, congestive heart failure, ischemic heart disease, hypertension, liver disease, peptic ulcer disease, kidney disorder, diabetes mellitus, hyperlipidemia, hypothyroidism, osteoarthritis, osteoporosis, cerebral vascular accident, depression, asthma, emphysema/chronic obstructive pulmonary disease, past tuberculosis, past pneumonia, and drug allergies/reactions.

Baseline DAS28  <0.0001
 >7 vs ≤5.10.2650.225–0.311<0.0001
 >5.1–≤7 vs ≤5.10.5040.459–0.553<0.0001
Baseline C-reactive protein, mg/liter   
 ≥20 vs <200.8380.769–0.913<0.0001
Baseline HAQ DI score, 0–3  <0.0001
 ≥2 vs <10.6030.533–0.683<0.0001
 1.5–<2 vs <10.7390.659–0.829<0.0001
 1–<1.5 vs <10.8700.778–0.9740.0157
Previous DMARD use, no.  0.0443
 ≥5 vs 10.8790.754–1.0250.1004
 4 vs 11.0560.921–1.2120.4354
 3 vs 11.0340.916–1.1680.5872
 2 vs 11.1000.983–1.2310.0967
Previous etanercept and/or infliximab use   
 Yes vs no0.8250.709–0.9600.0126
Duration of rheumatoid arthritis, years  0.0725
 >10 vs ≤10.8840.720–1.0850.2377
 >5–10 vs ≤10.9010.732–1.1090.3248
 >2–5 vs ≤10.8460.687–1.0410.1144
 >1–2 vs ≤11.0560.837–1.3320.6462
Rheumatoid factor   
 Positive vs negative0.9420.861–1.0310.1943
Age, years  <0.0001
 >75 vs <400.6110.461–0.8100.0006
 >65–≤75 vs <400.6760.584–0.783<0.0001
 40–≤65 vs <400.7790.701–0.865<0.0001
Sex   
 Male vs female1.2841.160–1.422<0.0001
Comorbidities, no.   
 >1 vs ≤10.8530.780–0.9320.0005
Concomitant DMARD use, no.   
 ≥1 vs 01.3401.209–1.484<0.0001
Concomitant glucocorticoids (maximum  dosage 10 mg/day)   
 Yes vs no1.0470.959–1.1440.3034
Smoking history   
 Ever vs never1.0760.987–1.1740.0973

Time in clinical remission or time in MDA and predictors of maintenance.

Overall, of the patients who achieved remission (DAS28 score <2.6) during adalimumab treatment, 47% discontinued the study in a state of remission and 55% discontinued the study in MDA after a period of continuous remission or MDA. The Kaplan-Meier estimate for median time in continuous remission was 3.4 months (95% CI 3.2–3.7) (Figure 3). Similarly, the median time that patients remained in at least continuous MDA following remission was 4.4 months (95% CI 4.1–5.1). Twenty-five percent of patients remained in remission for at least 8.7 months (95% CI 7.3–12.5).

Figure 3.

Kaplan-Meier curves for time in continuous remission (Disease Activity Score in 28 joints [DAS28] <2.6; median 3.4 months) or in minimal disease activity (MDA; DAS28 score ≤2.85; median 4.4 months). * Number of patients still under observation excluding those who already experienced loss of remission. + Number of patients still under observation excluding those who already experienced loss of MDA. Solid line = time in MDA after remission; broken line = time in remission.

The analysis of predictors for sustained remission, measured as hazard ratios for loss of remission, revealed that the following factors were associated with a longer duration of remission: male sex, no previous TNF antagonist treatment, younger age, and short time to clinical remission (Table 3). Additional factors predicting a longer time in at least MDA after clinical remission (measured as time to loss of MDA) were smoking (never), ≤1 comorbidity, and low baseline HAQ DI score and DAS28 score, whereas age had no significant impact on time in MDA after remission (data not shown).

Table 3. Predictors of time to loss of clinical remission (DAS28 score ≥2.6)*
Possible confounderHazard ratio95% CIP
  • *

    95% CI = 95% confidence interval; see Table 1 for additional definitions.

  • Overall P value for all subcategories.

  • Includes valvular heart disease, congestive heart failure, ischemic heart disease, hypertension, liver disease, peptic ulcer disease, kidney disorder, diabetes mellitus, hyperlipidemia, hypothyroidism, osteoarthritis, osteoporosis, cerebral vascular accident, depression, asthma, emphysema/chronic obstructive pulmonary disease, past tuberculosis, past pneumonia, and drug allergies/reactions.

Time to remission (per month)1.0241.003–1.0460.0243
Baseline DAS28 score  0.1813
 >7.0 vs ≤5.11.1770.939–1.4760.1576
 >5.1–≤7 vs ≤5.11.1220.984–1.2780.0857
Baseline C-reactive protein, mg/liter   
 ≥20 vs <200.9460.842–1.0640.3544
Baseline HAQ DI score, 0–3  0.0950
 ≥2 vs <11.2131.019–1.4440.0302
 1.5–<2 vs <11.1340.966–1.3310.1252
 1–<1.5 vs <11.0180.868–1.1950.8250
Previous DMARD use, no.  0.3691
 ≥5 vs 11.0080.820–1.2410.9364
 4 vs 10.9950.824–1.2020.9604
 3 vs 10.9290.785–1.1000.3949
 2 vs 10.8690.743–1.0170.0794
Previous etanercept and/or infliximab use   
 Yes vs no1.3361.097–1.6290.0041
Duration of rheumatoid arthritis, years  0.4865
 >10 vs ≤11.2110.907–1.6430.1884
 >5–10 vs ≤11.3020.964–1.7570.0855
 >2–5 vs ≤11.1880.879–1.6070.2627
 >1–2 vs ≤11.2250.879–1.7090.2312
Rheumatoid factor   
 Positive vs negative1.0960.967–1.2430.1525
Age, years  0.0063
 >75 vs <401.5811.084–2.3060.0174
 65–≤75 vs <401.3011.056–1.6030.0136
 40–≤65 vs <401.2801.100–1.4890.0014
Sex   
 Male vs female0.7110.614–0.824< 0.0001
Comorbidities, no.   
 >1 vs ≤11.0650.942–1.2050.3136
Concomitant DMARD use, no.   
 ≥1 vs 00.9750.844–1.1260.7280
Concomitant glucocorticoid use (maximum  dosage 10 mg/day)   
 Yes vs no1.1140.982–1.2630.0924
Smoking history   
 Ever vs never1.1040.980–1.2430.1030

Finally, the DAS28 values of those patients who experienced clinical remission (DAS28 score <2.6) at any time during adalimumab treatment were evaluated until patients discontinued the study. This patient subset entered the study with a mean and median DAS28 score of 5.6. The mean and median DAS28 scores at the last observation were 2.5 and 2.3, respectively, indicating that more than 50% of this subset discontinued the study in a state of clinical remission. At 8, 16, 24, and 32 weeks after the first assessment of clinical remission, the observed mean and median DAS28 scores in 2,049, 1,639, 1,235, and 862 patients, respectively, were 2.5 and 2.4 at each time point.

DISCUSSION

The database from the ReAct trial presented an excellent opportunity to investigate the effectiveness of adalimumab on achievement of clinical remission and MDA in patients with severe RA (mean baseline DAS28 score 6.0) (26, 27). Limitations of the study, including an open-label design and a relatively short study duration, are balanced by a large data set that was obtained in a study designed to reflect usual clinical practice. In the current analyses, we focused on the clinical aspects of remission or MDA during adalimumab therapy and did not attempt to validate the various assessment methods. A separate evaluation of the various methodologic approaches will be published separately.

Regardless of the assessment method, the percentage of patients who met the remission or MDA requirements increased after week 12. Dose escalation was rare during the study and had no impact on the overall rate of remission/MDA. In general, clinical remission rates using the SDAI and CDAI definitions were similar to each other but were more restrictive than the DAS28 definition. We also applied the new OMERACT concept of MDA, which encompasses a restrictive parameter (absence of tender/swollen joints plus normal ESR) that is indicative of remission, as well as 2 other definitions for an acceptable state of low disease activity in which future joint damage can be prevented (26). The core set of MDA criteria tended to produce response profiles similar to the DAS28 score ≤2.85 requirement, whereas the strictest MDA definition yielded fewer responders than the DAS28 definition of clinical remission. Wells and colleagues (36) evaluated MDA outcomes in 2 abatacept studies. In their study of TNF antagonist–naive patients with RA, a lower percentage of patients in MDA were reported at week 12 compared with our study; however, this comparison is limited by differences in study design (36).

Overall, results from the ReAct trial were consistent with previous placebo-controlled, randomized clinical trials of adalimumab as a treatment for RA (37–39). The remission rates documented for clinical studies of etanercept (40–42) and infliximab (4) are similar to or somewhat lower than those observed in the ReAct trial; however, differences in study design limit our ability to directly compare the results of our study with results from randomized controlled studies. The 1-year remission rate associated with biologic treatment was 16% in the German biologics register (pooled data of 4 biologics, with 15% of patients receiving adalimumab) (34). No data on remission during adalimumab treatment have been published by the British biologic register, but the remission rates for etanercept and infliximab were lower than in clinical studies (32, 33). The baseline characteristics of patients in the ReAct trial were similar to the patients in the German register (34), whereas patients in the British register had notably higher disease activity and worse physical function (32, 33). In general, a comparison between the studies is limited by both different response assessments and different statistical methods used to assess predictors.

However, the results of our univariate subgroup analyses are similar overall to previously reported findings for both biologic registry and clinical trial data (31–35, 43). Irrespective of the assessment method, patients with moderate RA activity achieved higher rates of remission/MDA compared with patients with more severe RA. Moreover, the percentage of patients meeting the most restrictive MDA definition was substantially higher in patients who were less disabled at baseline. In addition, patients who achieved a rapid response to adalimumab treatment achieved higher rates of clinical remission/MDA compared with nonresponders.

Although the median time to remission or MDA was ∼3 months in patients who experienced remission/MDA, other results, including the relatively low rate of remission/MDA observed in the analyses of week 12 data, indicate that many patients experienced remission or MDA later in the course of adalimumab treatment.

Among other factors, younger age, male sex, less severe RA, and concomitant DMARD therapy increased the likelihood of achieving clinical remission or MDA, whereas the number of previous DMARDs, the duration of RA, concomitant glucocorticoid therapy, and smoking history had no influence. Overall, these findings were consistent with the results published in the British register (33). However, we did not assess the impact of nonsteroidal antiinflammatory drug use in our analysis.

Remission is, in general, difficult to maintain in patients with RA, particularly in patients with long-established and severe RA (19). The mean adalimumab treatment duration of approximately 8 months in the ReAct trial restricts the evaluation of maintenance of remission/MDA, but this period is comparable with published data (34, 42). In the German register, 7.7% of patients treated with anti-TNF with a DAS28 score <2.6 achieved this value after 6 months of treatment (34). In the Trial of Etanercept and Methotrexate with Radiographic Patient Outcomes, 18% of patients with a DAS28 score <2.6 sustained this level of response after 6 months of treatment with etanercept and methotrexate (42). In the ReAct trial, approximately 50% of patients who achieved remission experienced sustained remission for a median of 3.4 months, and 25% experienced sustained remission for 7.5 months. We analyzed the time in MDA after remission because MDA is considered an acceptable state from both a patient and physician perspective. Half of the patients maintained at least MDA for a median of 4.4 months. However, approximately one-quarter of the patients discontinued in a state of remission or MDA and could not be evaluated for long-term outcomes. The maintenance of remission was strongly dependent on male sex and, to a lesser degree, on a short time to remission, the absence of previous TNF antagonist therapy, and younger age. By comparison, sustained MDA was dependent not on age but on smoking history (never). Consistent with previous reports, we found a strong relationship between male sex and achievement and maintenance of remission (32, 43).

To our knowledge, this analysis represents the largest, prospective clinical trial data set to be characterized using the OMERACT criteria for MDA. Because remission is difficult to maintain, the MDA data demonstrating the benefit of adalimumab treatment should be of value to rheumatologists treating patients with RA.

Depending on the assessment method, clinical remission was achieved in approximately one-quarter to more than one-third of patients with active RA during adalimumab therapy in the ReAct trial. Nearly half of the patients experienced MDA as defined by OMERACT methods. Initiation of adalimumab treatment in combination with DMARDs in patients who are less disabled and have developed few comorbidities substantially increases the likelihood of remission/MDA. Overall, more than half of the patients who had experienced remission at any time during the study remained in remission at study end. Throughout the complete course of adalimumab therapy, at least MDA was sustained for a median of 4.4 months in patients who achieved remission.

AUTHOR CONTRIBUTIONS

Dr. Burmester had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study design. Burmester, Unnebrink, Kupper.

Acquisition of data. Burmester, Monteagudo-Sáez, Unnebrink.

Analysis and interpretation of data. Burmester, Ferraccioli, Flipo, Monteagudo-Sáez, Unnebrink, Kary, Kupper.

Manuscript preparation. Burmester, Flipo, Monteagudo-Sáez, Unnebrink, Kary, Kupper.

Statistical analysis. Flipo, Unnebrink.

ROLE OF THE STUDY SPONSOR

Abbott GmbH & Co. KG collaborated with Drs. Burmester, Ferraccioli, Flipo, and Monteagudo-Sáez to design and conduct the study. All authors analyzed the data, reviewed the manuscript critically for important intellectual content, approved the final draft, and agreed to its submission.

Acknowledgements

Abbott thanks the physicians and staff members of all study centers for having participated in ReAct. We thank Dr. Uemit Oezer for medical support, Dr. David Webber and Ina Reinhardt for study management, and Dana L. Randall, MS, PharmD, for editorial assistance.

Ancillary