SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES
  9. Supporting Information

Objective

To summarize the potential predictors of remission in patients with rheumatoid arthritis (RA).

Methods

We performed a systematic review of prognostic studies that identified the predictors of remission in RA patients. Studies were identified in Medline, EMBase, and the Cochrane Registry, and by hand search. We included only studies performing multivariate analysis.

Results

A total of 18 studies from 2,062 citations were included. The following variables were found to be the independent predictors of RA remission: male sex; young age; late-onset RA; short disease duration; nonsmoker; low baseline disease activity; mild functional impairment; low baseline radiographic damage; absence of rheumatoid factor and anti–citrullinated peptide; low serum level of acute-phase reactant, interleukin-2, and RANKL at baseline; MTHFR 677T alleles and 1298C alleles in the methotrexate (MTX)–treated patients; magnetization transfer ratio 2756A allele ± either the SLC 19A180A allele or the TYMS 3R-del6 haplotype in the MTX plus sulfasalazine combination–treated patients; early treatment with nonbiologic disease-modifying antirheumatic drug (DMARD) combinations; the use of anti–tumor necrosis factor (anti-TNF); the concurrent use of DMARDs in anti-TNF–treated patients; and moderate or good response to treatments at the first 6 months. The magnitude of the association in the individual predictor was diverse among the studies depending on the patient characteristics, the study characteristics, and the variables used to adjust for in the models.

Conclusion

A number of independent predictors of remission, i.e., baseline clinical and laboratory characteristics and genetic markers, were summarized. The predictive value of prognostic factors recently identified needs to be confirmed.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES
  9. Supporting Information

Rheumatoid arthritis (RA) is a chronic systemic inflammatory disease leading to joint inflammation, functional disability, deformities, and a shortened life expectancy. In the past, the treatment in RA aimed at a reduction of disease activity to a low disease activity state due to a lack of effective treatment. With the advent of biologic agents, the ultimate goal of treatment in RA is now remission in the early stage of the disease before patients develop permanent deformities, functional disability, and RA-related systemic morbidity and mortality (1–4). Although biologic agents are effective in suppressing inflammation and halting radiographic progression, some patients experienced adverse effects and had to discontinue these medications (5). Furthermore, they are costly. Therefore, it is important to identify predictors of remission in the early stage of the disease, so that physicians can individualize the treatment plan. Perhaps patients who have poor prognostic factors should receive more aggressive treatment. It is also important from the patient's perspective to understand their clinical course and prognosis. The purpose of our study was to summarize the potential predictors of remission in RA patients.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES
  9. Supporting Information

Literature search.

To identify the relevant studies, medical subject headings and keywords related to “rheumatoid arthritis,” “remission,” and “prognostic study” were used. The search strategy is listed in Supplementary Appendix A (available in the online version of this article at http://www3.interscience.wiley.com/journal/77005015/ home). All of the subheadings under each of these headings were combined with “OR.” Subsequently, the set of articles related to “rheumatoid arthritis” was combined with the other 2 sets of “remission” and “prognostic study” using the combination term “AND.” The following bibliographic databases were searched: Medline (1950 to November 2008), EMBase (1980 to November 2008), and the Cochrane Central Register of Controlled Studies (second quarter 2008). Reference lists of all relevant studies from the electronic search were manually searched to identify additional eligible studies. Nonhuman and pediatric (age <18 years) studies were excluded. All languages were included.

Study selection.

Two reviewers (WK, VP) independently screened the titles and abstracts of the retrieved articles for selection criteria. The studies were included if they met the following criteria: 1) participants: the patients studied were people who met the American College of Rheumatology (ACR; formerly the American Rheumatism Association) 1987 revised criteria for RA and were age ≥18 years (6); 2) outcome: assessment of the potential predictive or prognostic factors related to remission was required, and all definitions of remission were included; and 3) statistical analysis: studies had to use multivariate analysis, e.g., multiple linear logistic regression or Cox proportional hazards model, to identify the potential predictive factors of remission. We finally excluded 2 Polish studies because translation for this language was not available.

Data abstraction and quality assessment.

Two reviewers independently collected the data (WK, CS) and assessed the quality of the studies (WK, SJ). Any discrepancies were resolved by consensus. The methodologic quality of the included studies was assessed using the Quality Assessment for Prognostic Studies, a critical appraisal index developed and validated for prognosis studies in systematic reviews (see Supplementary Appendix B, available in the online version of this article at http://www3.interscience.wiley.com/journal/77005015/home) (7). This scale comprises an assessment of the risk of potential bias in 6 domains: study participation, study attrition, prognostic factor measurement, outcome measurement, confounding measurement, and analysis. Each domain consists of 3–7 criteria. Each criterion was rated as “yes” (if they met the criterion), “no” (if they did not meet the criterion), or “unsure” (if the data were not clear). The more the study met the criteria, the lower the risk of bias. The quality of the studies was assessed without masking of trial identifiers.

Data synthesis.

We summarized the data by stratifying predictors into 2 groups: 1) clinical predictors and 2) laboratory and radiographic predictors. The association between predictors and remission was expressed as the odds ratio (OR), hazard ratio (HR), or coefficient (β), with its corresponding 95% confidence interval (95% CI) as appropriate.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES
  9. Supporting Information

Our search retrieved 2,062 citations after removal of duplicates within and across databases. After review of the titles and abstracts, 48 full-text articles (44 from electronic databases and 4 from hand search) were retrieved for further evaluation, of which 18 articles were retained for our analysis (Figure 1). The excluded studies and reasons for exclusion are summarized in Supplementary Appendix C (available in the online version of this article at http://www3.interscience.wiley.com/journal/77005015/home).

thumbnail image

Figure 1. Results of the literature search and disposition of the potentially relevant studies. RA = rheumatoid arthritis.

Download figure to PowerPoint

Characteristics of included studies.

The study characteristics and detailed quality of the included studies are shown in Tables 1 and 2, respectively. Most studies (16 of 18) were prospective cohort studies of nonbiologic disease-modifying antirheumatic drug (DMARD)–treated RA (8–17) or anti–tumor necrosis factor (anti-TNF)–treated RA (18–23). Four of these 16 studies used the data from randomized controlled trials: the Finnish Rheumatoid Arthritis Combination Therapy (FIN-RACo) trial (9, 13), the Trial of Etanercept and Methotrexate with Radiographic Patient Outcomes (TEMPO) study (20), and the study by Verstappen et al (12). The other 2 studies were retrospective cohort (24) and case–control studies (25). Study duration ranged between 6 months and 5 years. The following 8 studies had more than 20% dropout or missing data: the study by Gossec et al (11) had 30% noncompleters who had significantly higher antikeratin antibody positivity. Kuuliala et al (13) and Verstappen et al (12) did not report the data comparing between completers and noncompleters. In the studies by Hyrich et al (18, 19), the excluded patients (21%) were significantly older, had a longer disease duration, and were more likely to have a comorbid disease. Mancarella et al (24) reported that the excluded cohort was slightly different from the included cohort in the percentage of patients' steroid takings (54% versus 49%). Van der Heijde et al (20) reported that more patients in the methotrexate (MTX) group significantly withdrew than the etanercept group or the combination of MTX plus etanercept group. In the study by Burmester et al, or the Research in Active RA (ReAct) trial (23), the baseline characteristics between completers and noncompleters were similar.

Table 1. Characteristics of included studies*
Author, year (ref.)Country or trialStudy designTrial durationSample sizeQuality of study (risk of bias)Remission criteriaTreatment
  • *

    SJC = swollen joint count; DMARDs = disease-modifying antirheumatic drugs; ACR = American College of Rheumatology; SSZ = sulfasalazine; MTX = methotrexate; FIN-RACo = Finnish Rheumatoid Arthritis Combination Therapy trial; RCT = randomized controlled trial; HCQ = hydroxychloroquine; pred = prednisone; DAS = Disease Activity Score; anti-TNF = anti–tumor necrosis factor; ESR = erythrocyte sedimentation rate; IM = intramuscular; NSAIDs = nonsteroidal antiinflammatory drugs; DAS28 = Disease Activity Score in 28 joints; etan = etanercept; inflix = infliximab; ada = adalimumab; TEMPO = Trial of Etanercept and Methotrexate with Radiographic Patient Outcomes; MP = methylprednisolone; TJC = tender joint count; ReAct = Research in Active rheumatoid arthritis.

  • ACR remission criteria: 1) morning stiffness of 15 minutes, 2) no joint pain (by history), 3) no joint tenderness or pain on motion, 4) no soft tissue swelling in joint or tendon sheaths, 5) no fatigue, and 6) ESR of 30 mm/hour in women and 20 mm/hour in men. A patient must have ≥5 of the 6 criteria for at least 2 months and must not have any of the following clinical manifestations of active disease: vasculitis, pericarditis, pleuritis, myositis, weight loss, or fever attributable to rheumatoid arthritis.

  • DAS remission criteria: DAS ≤1.6.

  • §

    DAS28 remission criteria: DAS28 <2.6.

Pease et al, 1999 (8)UKCohort/prospective3.6 years400ModerateRitchie Articular Index 0, no SJC, >3 monthsDMARDs
Molenaar et al, 2002 (10)The NetherlandsCohort/prospective2 years167HighModified ACR, excluded fatigueDMARDs: SSZ, gold, MTX, antimalarial drugs; no steroids
Möttönen et al, 2002 (9)FIN-RACoCohort (used data from RCT)24 months178ModerateModified ACR, excluded fatigue and duration criteriaMTX + HCQ + SSZ + pred vs. single DMARD
Gossec et al, 2004 (11)FranceCohort/prospective5 years134ModerateDASMTX vs. SSZ, vs. MTX + SSZ
Liang et al, 2005 (25)USCase–control322ModerateModified ACR, excluded duration criterionAnti-TNF ± DMARDs vs. no anti-TNF ± DMARDs
Kuuliala et al, 2005 (13)FIN-RACoCohort (used data from RCT)6 months157LowModified ACR, excluded fatigue and duration criterionMTX + SSZ + HCQ + prednisolone vs. SSZ ± prednisolone
Verstappen et al, 2005 (12)The NetherlandsCohort (used data from RCT)4 years425ModerateMorning stiffness ≤15 minutes, pain ≤10 minutes, joint score ≤10, ESR ≤30 mm/hourHCQ, IM gold, MTX vs. pyramid (NSAIDs [RIGHTWARDS ARROW] DMARDs)
Hyrich et al, 2006 (18)UKCohort/prospective6 months2,711LowDAS28§Etan, inflix, ada
Hyrich et al, 2006 (19)UKCohort/prospective6 months2,879LowDAS28§Etan, inflix, ada
Mancarella et al, 2007 (24)ItalyCohort/retrospective6 months591ModerateDAS28§Etan, inflix, ada
Van der Heijde et al, 2007 (20)TEMPOCohort (used data from RCT)3 years332ModerateDASEtan + MTX vs. etan vs. MTX
Vázquez et al, 2007 (14)SpainCohort/prospective2 years105LowDAS28§MTX + gold salts step strategy + MP 4 mg/day
González-Alvaro et al, 2007 (21)SpainCohort/prospective30 weeks75HighDAS28§Ada, inflix
Kurzawski et al, 2007 (15)PolandCohort/prospective1 year174HighESR <25 mm/hour, SJC <3, TJC <3MTX + MP 4 mg/day
Forslind et al, 2007 (16)SwedenCohort/prospective5 years608LowDAS28§DMARDs or combination biologic drugs
Burmester et al, 2008 (23)ReAct trialCohort/prospective5 years1,257ModerateDAS28§Ada ± DMARDs
James et al, 2008 (17)AustraliaCohort/prospective12 months98HighDAS28§MTX + HCQ + SSZ, MTX + HCQ, MTX + SSZ, MTX
Kristensen et al, 2008 (22)SwedenCohort/prospective6 months1,506LowDAS28§Etan, inflix, ada
Table 2. Quality of included studies
Author, year (ref.)Research question clearly statedRisk of bias
Study participationStudy attritionPrognostic factor measurementOutcome measurementConfounding accounted forAnalysisOverall
Pease et al, 1999 (8)YesLowLowLowModerateLowModerateModerate
Molenaar et al, 2002 (10)YesHighHighModerateLowModerateLowHigh
Möttönen et al, 2002 (9)YesLowLowLowLowModerateModerateModerate
Gossec et al, 2004 (11)YesLowLowLowLowModerateModerateModerate
Liang et al, 2005 (25)YesLowLowLowModerateLowModerateModerate
Kuuliala et al, 2005 (13)YesLowLowLowLowLowLowLow
Verstappen et al, 2005 (12)YesLowLowLowModerateLowModerateModerate
Hyrich et al, 2006 (18)YesLowLowLowLowModerateLowLow
Hyrich et al, 2006 (19)YesLowLowLowLowLowModerateLow
Mancarella et al, 2007 (24)YesLowModerateLowLowLowModerateModerate
Van der Heijde et al, 2007 (20)YesLowHighLowLowLowLowModerate
Vázquez et al, 2007 (14)YesLowLowLowLowLowModerateLow
González-Alvaro et al, 2007 (21)YesLowHighLowLowModerateModerateHigh
Kurzawski et al, 2007 (15)YesLowHighModerateModerateModerateModerateHigh
Forslind et al, 2007 (16)YesLowLowLowLowLowModerateLow
Burmester et al, 2008 (23)YesLowModerateModerateLowModerateModerateModerate
James et al, 2008 (17)YesLowHighModerateLowModerateModerateHigh
Kristensen et al, 2008 (22)YesLowLowLowLowModerateLowLow

Included studies varied in remission criteria used and treatment. Some used well-established criteria, e.g., the Disease Activity Score (DAS) (11, 20) or Disease Activity Score in 28 joints (DAS28) remission criteria (14, 16–19, 21–24). Some used their own criteria (8, 12, 15) or modified ACR remission criteria (9, 10, 13, 25). Treatment included nonbiologic, biologic, single, and combined DMARDs. The patient characteristics and remission rate of the included studies are shown in Table 3. The remission rate reported varied depending on the remission criteria used, time to assess remission, and subgroup of interest. The remission rate was higher in “point remission” (remission at one followup visit) than in “persistent remission” (remission at ≥2 consecutive followup visits).

Table 3. Baseline characteristics of the patients*
Author, year (ref.)GroupAge, mean or median ± SD or (range)Women, %Disease duration, mean or median ± SD or (range)RF+, %Anti-CCP+, %Disease activityHAQ score, mean or median ± SD or (range)Remission at the end of followup, %
MeasureMean or median ± SD or (range)
  • *

    RF = rheumatoid factor; anti-CCP = anti–citrullinated peptide; HAQ = Health Assessment Questionnaire; RA = rheumatoid arthritis; RAI = Ritchie Articular Index; DAS = Disease Activity Score; SJC = swollen joint count; anti-TNF = anti–tumor necrosis factor; HCQ = hydroxychloroquine; IM = intramuscular; MTX = methotrexate; DAS28 = Disease Activity Score in 28 joints; etan = etanercept; DMARD = disease-modifying antirheumatic drug; inflix = infliximab; ada = adalimumab.

Pease et al, 1999 (8)Younger-onset RA (age 18–65 years at onset)52 (19–64)687 (0.6–24.7) months57 RAI6 (0–48)1.1 (0–2.9)20
 Late-onset RA (age >65 years at onset)73 (65–90)654 (0.24–24.9) months43 RAI1 (0–51)1.8 (0–3.0)46
Molenaar et al, 2002 (10)All cohort50  69 DAS1 (0.0–2.9) 53
Möttönen et al, 2002 (9)Combination therapy, short delay (0–4 months)48 ± 1054 77 SJC14 (8–16)1 (0.5–1.4)42
 Combination therapy, long delay (>4 months)46 ± 1060 72 SJC13 (9–16)0.8 (0.4–1)42
 Single therapy, short delay (0–4 months)50 ± 965 70 SJC14 (10–17)0.6 (0.4–1.3)35
 Single therapy, long delay (>4 months)47 ± 1167 70 SJC13 (10–15)0.9 (0.4–1.3)11
Gossec et al, 2004 (11)All cohort51 ± 1573.33 ± 2.6 months8159DAS4.1 ± 0.81.3 ± 0.720
Liang et al, 2005 (25)Anti-TNF53 ± 128314 ± 10 years     18 (total cohort)
Non–anti-TNF58 ± 148411 ± 8.3 years
Kuuliala et al, 2005 (13)Combination therapy47 (23–65)617 (2–22) months69 SJC13 (9–17)0.9 (0.4–1.1)23
 Single therapy48 (20–65)679 (3–23) months67 SJC13 (10–16)0.9 (0.4–1.3)9
Verstappen et al, 2005 (12)All cohort56 ± 14       33
HCQ57 ± 1471 68   1.4 ± 0.7
 IM gold55 ± 1569 61   1.3 ± 0.8
 MTX56 ± 1469 68   1.2 ± 0.7
 Pyramid56 ± 1472 60   1.3 ± 0.7
Hyrich et al, 2006 (18)All cohort57 ± 127815 ± 9 years  DAS286.7 ± 12.2 ± 0.5
Etan58 ± 128016 ± 10 years  DAS286.8 ± 12.2 ± 0.55
 Etan + MTX54 ± 127613 ± 8 years  DAS286.6 ± 0.92.1 ± 0.512
 Etan + non-MTX DMARD55 ± 127915 ± 9 years  DAS286.6 ± 0.92.1 ± 0.510
 Inflix59 ± 127916 ± 11 years  DAS286.8 ± 1.12.2 ± 0.57
 Inflix + MTX55 ± 127714 ± 9 years  DAS286.7 ± 0.92.1 ± 0.58
 Inflix + non-MTX DMARD58 ± 127414 ± 9 years  DAS286.8 ± 1.12.2 ± 0.64
Hyrich et al, 2006 (19)Etan56 ± 127815 ± 9 years72 DAS286.7 ± 12.1 ± 0.58
Inflix55 ± 127714 ± 9 years72 DAS286.7 ± 12.1 ± 0.59
Mancarella et al, 2007 (24) 53 ± 136712 ± 7.6 years79 DAS285.9 ± 1.21.6 ± 0.726
Van der Heijde et al, 2007 (20)MTX53 ± 13797 ± 5.5 years71 DAS  18
Etan53 ± 14776 ± 5.1 years75    22
 Etan + MTX53 ± 12747 ± 5.4 years76    41
Vázquez et al, 2007 (14)All cohort55 ± 157710 ± 6.7 months7470DAS285.7 ± 0.91 ± 0.632
González-Alvaro et al, 2007 (21)All cohort56 ± 14818 (4–15) years87 DAS285.9 ± 11.6 (1.1–2.1)13
Ada57 ± 14799 (4–15) years89 DAS285.9 ± 1.11.6 (1.1–2.3)
 Inflix50 ± 141008 (6–24) years70 DAS286 ± 11.6 (1.3–2)
Kurzawski et al, 2007 (15)All cohort55 ± 117410 ± 7.8 years71 DAS286.3 ± 2.7 15
Forslind et al, 2007 (16)All cohort58 ± 15646 ± 3.2 months6056DAS285.3 ± 1.31 ± 0.739
Burmester et al, 2008 (23)All cohort548111 years73 DAS2861.638
James et al, 2008 (17)All cohort57 (44–69)7512 (8–20) weeks5149DAS285.9 (5.1–6.4) 40
Kristensen et al, 2008 (22)Men58 ± 12 11 ± 10 years  DAS285.4 ± 1.31.1 ± 0.618
Women55 ± 14 12 ± 10 years  DAS285.6 ± 1.21.4 ± 0.717

Predictors of remission.

Tables 4 and 5 summarize the clinical predictors and laboratory and radiographic predictors for remission in patients with RA, respectively.

Table 4. Clinical predictors for remission in patients with rheumatoid arthritis (RA)*
PredictorStudyNTime to assess remissionTreatmentEffect sizePAdjusted for
  • *

    DMARDs = disease-modifying antirheumatic drugs; OR = odds ratio; RF = rheumatoid factor; anti-CCP = anti–citrullinated peptide; DAS28 = Disease Activity Score in 28 joints; HAQ = Health Assessment Questionnaire; etan = etanercept; MTX = methotrexate; 95% CI = 95% confidence interval; DAS = Disease Activity Score; TSS = total Sharp/van der Heijde score; ada = adalimumab; HR = hazard ratio; inflix = infliximab; CRP = C-reactive protein; NSAIDs = nonsteroidal antiinflammatory drugs; ESR = erythrocyte sedimentation rate; ACR20 = American College of Rheumatology 20% improvement criteria; ACR50 = ACR 50% improvement criteria; EULAR = European League Against Rheumatism; SE = shared epitope; SJC = swollen joint count; TJC = tender joint count; anti-TNF = anti–tumor necrosis factor.

Male sexForslind et al, 2007 (16)60818–60 monthsNonbiologic DMARDsOR 1.5–2.9≤ 0.04RF, Anti-CCP, DAS28, HAQ, smoking status, disease duration
 Van der Heijde et al, 2007 (20)3323 yearsBiologic DMARDs (etan ± MTX)OR 1.9 (95% CI 1.3–2.8)< 0.05Disease duration, RF, previous MTX use, DAS, HAQ, etan, etan + MTX, baseline TSS
 Burmester et al, 2008 (23)1,2573 yearsBiologic DMARDs (ada)HR 1.3 (95% CI 1.2–1.4)< 0.0001RF, smoking status, disease duration, concomitant DMARDs and steroids, history of previously failed DMARDs, time to remission
Female sexHyrich et al, 2006 (19)2,87912 monthsBiologic DMARDs (etan, inflix)OR 0.6 (95% CI 0.4–0.9)No dataAge, smoking, comorbidity, RF, previous use of MTX or steroids
Late-onset RA (age >65 years)Pease et al, 1999 (8)400Any point within 5 yearsNonbiologic DMARDsOR 3 (95% CI 1.8–5)< 0.05RF, HLA–DR1, HLA–DR4, myalgia, acute-phase reactants, anemia, body weight
Age ≥53 yearsMancarella et al, 2007 (24)5916 monthsBiologic DMARDs (inflix, etan, ada)OR 0.6 (95% CI 0.4–0.9)No dataSex, HAQ, RF
Age >40 yearsBurmester et al, 2008 (23)1,2573 yearsBiologic DMARDs (ada)HR 0.6–0.8≤ 0.0006Sex, comorbidities, DAS, HAQ, RF, CRP level, smoking status, disease duration, concomitant DMARDs and steroids, history of previously failed DMARDs, previous etan/inflix use, time to remission
Disease durationForslind et al, 2007 (16)60818–60 monthsNonbiologic DMARDsOR 0.91–0.93< 0.02Sex, RF, Anti-CCP, DAS28, HAQ, smoking status
SmokingForslind et al, 2007 (16)60818–60 monthsNonbiologic DMARDsOR 0.6–0.7< 0.04Sex, RF, Anti-CCP, DAS28, HAQ, disease duration
 Hyrich et al, 2006 (19)2,87912 monthsBiologic DMARDs (etan)OR 1.1 (95% CI 0.8–1.4)No dataAge, sex, comorbidity, RF, DAS28, HAQ, NSAID or MTX use, no. of previous DMARDs
    Biologic DMARDs (inflix)OR 0.8 (95% CI 0.6–0.99)No data 
ComorbiditiesBurmester et al, 2008 (23)1,2573 yearsBiologic DMARDs (ada)HR 0.9 (95% CI 0.8–0.9)0.0005Sex, age, DAS, HAQ, RF, smoking status, disease duration, concomitant DMARDs and steroids, history of previously failed DMARDs, previous etan/inflix use, time to remission
Baseline DAS28Forslind et al, 2007 (16)60818–60 monthsNonbiologic DMARDsOR 0.7–0.8< 0.006Sex, disease duration, RF, anti-CCP, HAQ, smoking status
 Hyrich et al, 2006 (19)2,87912 monthsBiologic DMARDs (etan)OR 0.6 (95% CI 0.5–0.7)No dataAge, sex, comorbidity, smoking status, RF, HAQ, NSAID or MTX use, no. of previous DMARDs
    Biologic DMARDs (inflix)OR 0.7 (95% CI 0.5–0.8)No data
 González-Alvaro et al, 2007 (21)7530 weeksBiologic DMARDs (inflix, ada)OR 0.3 (95% CI 0.1–0.9)0.032Serum RANKL
 Kristensen et al, 2008 (22)1,5063 and 6 monthsBiologic DMARDs (inflix, etan, ada)OR 0.8 (no data)< 0.01Sex, disease duration, HAQ, history of DMARDs and NSAIDs, CRP level, current use of prednisolone
Baseline DAS28 >5.1Burmester et al, 2008 (23)1,2573 yearsBiologic DMARDs (ada)HR 0.3–0.8< 0.0001Sex, age, comorbidities, RF, smoking status, disease duration, concomitant DMARDs and steroids, history of previously failed DMARDs, time to remission, HAQ
Baseline DAS <4Gossec et al, 2004 (11)1343–5 yearsNonbiologic DMARDsOR 5.5–5.7< 0.0001Sharp score <4, morning stiffness <60 minutes, CRP level <14.5 mg/liter
Baseline DAS28 <5.1Vázquez et al, 2007 (14)1052 yearsNonbiologic DMARDsOR 4.1–4.7≤ 0.004Sex, pain, ESR, hemoglobin, RF, ± ACR20 and ACR50 or EULAR response at 6 months
Lower baseline DASVan der Heijde et al, 2007 (20)3323 yearsBiologic DMARDs (etan ± MTX)OR 1.7 (95% CI 1.4–2)< 0.05Sex, disease duration, RF, previous MTX use, HAQ, etan, etan + MTX, baseline TSS
Ritchie's score <17Gossec et al, 2004 (11)1343–5 yearsNonbiologic DMARDsOR 2.7–4.2≤ 0.02HAQ, initial Sharp score <4, morning stiffness, CRP level <14.5 mg/liter
PainVerstappen et al, 2005 (12)4254 yearsNonbiologic DMARDsβ = 0.98 (95% CI 0.97–0.99)0.0001Joint score, RF, good response to treatment at the first 6 months
HAQForslind et al, 2007 (16)60818–60 monthsNonbiologic DMARDsOR 0.6≤ 0.03Sex, RF, anti-CCP, DAS28, disease duration, smoking status
 Hyrich et al, 2006 (19)2,87912 monthsBiologic DMARDs (etan)OR 0.5 (95% CI 0.3–0.8)No dataAge, sex, comorbidity, smoking status, RF, DAS, NSAID or MTX use, no. of previous DMARDs
    Biologic DMARDs (inflix)OR 0.4 (95% CI 0.3–0.6)No data 
 Kristensen et al, 2008 (22)1,5063 and 6 monthsBiologic DMARDs (inflix, etan, ada)Inversely associated< 0.01Sex, disease duration, DAS, history of DMARD and NSAID use, CRP level, current use of prednisolone
HAQ score >1Burmester et al, 2008 (23)1,2573 yearsBiologic DMARDs (ada)HR 0.6–0.9≤ 0.02Sex, age, comorbidities, RF, smoking status, disease duration, concomitant DMARDs and steroids, history of previously failed DMARDs, time to remission, DAS
HAQ score ≥1.63Mancarella et al, 2007 (24)5916 monthsBiologic DMARDs (inflix, etan, ada)OR 0.6 (95% CI 0.4–0.8)No dataAge, sex, RF
Low baseline HAQ scoreVan der Heijde et al, 2007 (20)3323 yearsBiologic DMARDs (etan ± MTX)OR 1.6 (95% CI 1.2–2.1)< 0.05Sex, disease duration, RF, previous MTX use, DAS, etan, etan + MTX, baseline TSS
Delayed treatment (>4 months)Möttönen et al, 2002 (9)1782 yearsNonbiologic DMARDs11% vs. 35%0.021Age, sex, RF, SE, no. of other ACR criteria fulfilled
Combination therapyKuuliala et al, 2005 (13)1576 monthsNonbiologic DMARDsOR 4.4 (95% CI 1.6–12.2)No dataSex, RF, ESR, SJC, TJC
Use of anti-TNFLiang et al, 2005 (25)322Biologic DMARDsOR 2.7 (95% CI 1.4–5.3)No dataAge, sex, race, disease duration, NSAID and prednisolone use, history of previous DMARDs
Concurrent use of MTX or non-MTX DMARDs with anti-TNFHyrich et al, 2006 (18)2,71112 monthsBiologic DMARDs (etan + MTX vs. etan)OR 2.24 (95% CI 1.2–4.1)No dataAge, sex, comorbidity, smoking status, RF, DAS, NSAID or MTX use, no. of previous DMARDs used
 Hyrich et al, 2006 (19)2,87912 monthsBiologic DMARDs (etan + MTX vs. etan)OR 1.8 (95% CI 1.1–2.9)No dataAge, sex, smoking, comorbidity, RF, previous use of MTX or steroids or NSAIDs, DAS, HAQ
 Van der Heijde et al, 2007 (20)3323 yearsBiologic DMARDs (etan + MTX vs. etan)OR 2.4 (95% CI 1.6–3.6)< 0.0001Sex, disease duration, DAS, HAQ, RF, previous MTX use, DAS, baseline TSS
    Biologic DMARDs (etan + MTX vs. etan)OR 2.7 (95% CI 1.9–4)< 0.0001 
 Kristensen et al, 2008 (22)1,5063 and 6 monthsBiologic DMARDs (inflix, etan, ada)OR 2< 0.01Sex, disease duration, DAS, HAQ, history of DMARD and NSAID use, CRP level, current use of prednisolone
 Burmester et al, 2008 (23)1,2573 yearsBiologic DMARDs (ada)HR 1.3 (95% CI 1.2–1.5)< 0.0001Sex, age, comorbidities, RF, smoking status, disease duration, concomitant steroids, history of previously failed DMARDs, time to remission, DAS, HAQ
No. of previous DMARDsLiang et al, 2005 (25)322Biologic DMARDsOR 0.8 (95% CI 0.6–0.9)No dataAge, sex, race, disease duration, NSAIDs, prednisolone, previous DMARDs
 Hyrich et al, 2006 (19)2,87912 monthsBiologic DMARDs (etan and inflix)OR 0.8–0.9No dataAge, sex, smoking, comorbidity, RF, previous use of MTX or steroids or NSAIDs, DAS, HAQ
Previously failed etan/inflixBurmester et al, 2008 (23)1,2573 yearsBiologic DMARDs (ada)HR 0.8 (95% CI 0.7–1)0.01Sex, age, comorbidities, RF, smoking status, disease duration, concomitant steroids, history of previously failed DMARDs, time to remission, DAS, HAQ
NSAID useHyrich et al, 2006 (19)2,87912 monthsBiologic DMARDs (etan and inflix)OR 1.8–1.9No dataAge, sex, smoking, comorbidity, RF, previous use of MTX or steroids or NSAIDs, DAS, HAQ
 Kristensen et al, 2008 (22)1,5063 and 6 monthsBiologic DMARDs (etan, inflix, ada)OR 1.50.04Sex, disease duration, DAS, HAQ, history of DMARDs, CRP level, current use of prednisolone
EULAR (good response) at 1 yearVerstappen et al, 2005 (12)4254 yearsNonbiologic DMARDsβ = 4.7 (95% CI 3.2–7)0.0001Joint score, RF, pain
ACR50 at first 6 monthsVázquez et al, 2007 (14)1052 yearsNonbiologic DMARDsOR 3.9 (95% CI 1.1–13.4)0.03Sex, pain, ESR, hemoglobin, RF, DAS <5.1, ACR20
EULAR (good response) at first 6 monthsVázquez et al, 2007 (14)1052 yearsNonbiologic DMARDsOR 6.2 (95% CI 1.6–24)0.008Sex, pain, ESR, hemoglobin, RF, DAS <5.1
Table 5. Laboratory and radiographic predictors for remission in patients with rheumatoid arthritis*
PredictorStudyNTime to assess remissionTreatmentEffect sizePAdjusted for
  • *

    RF = rheumatoid factor; DMARDs = disease-modifying antirheumatic drugs; OR = odds ratio; 95% CI = 95% confidence interval; anti-CCP = anti–citrullinated peptide; DAS28 = Disease Activity Score in 28 joints; HAQ = Health Assessment Questionnaire; inflix = infliximab; etan = etanercept; ada = adalimumab; CRP = C-reactive protein; HR = hazard ratio; DAS = Disease Activity Score; IL-2 = interleukin-2; ESR = erythrocyte sedimentation rate; SJC = swollen joint count; TJC = tender joint count; TSS = total Sharp/van der Heijde score; MTX = methotrexate.

RFPease et al, 1999 (8)400Any point within 5 yearsNonbiologic DMARDsOR 0.5 (95% CI  0.3–0.8)< 0.05Age at onset, HLA–DR1, HLA–DR4,  myalgia, acute-phase reactants,  anemia, body weight
 Molenaar et al, 2002 (10)1672 yearsNonbiologic DMARDsNo association DMARD use, shared epitope  (DRB1*04, DRB1*13, and DQB1)
 Verstappen et al, 2005 (12)4254 yearsNonbiologic DMARDsβ = 1.63 (95%  CI 1.2–2.3) for  RF negative0.061Joint score, pain, good response to  treatment at the first 6 months
 Forslind et al, 2007 (16)60818–60 monthsNonbiologic DMARDsOR 0.5–0.6< 0.01Sex, RF, Anti-CCP, DAS28, HAQ,  smoking status, disease duration
 Mancarella et al, 2007 (24)5916 monthsBiologic DMARDs  (inflix, etan, ada)OR 0.6 (95% CI  0.4–0.96)No dataAge, sex, HAQ, RF
Anti-CCPForslind et al, 2007 (16)60824 monthsNonbiologic DMARDsOR 0.6 (95% CI  0.5–0.9)0.008Sex, disease duration, DAS28, HAQ
CRP level ≥20 mg/literBurmester et al, 2008 (23)1,2573 yearsBiologic DMARDs  (ada)HR 0.8 (95% CI  0.8–0.9)< 0.0001Sex, age, comorbidities, DAS, HAQ,  RF, smoking status, disease  duration, concomitant DMARDs  and steroids, history of  previously failed DMARDs,  previous etan/inflix use, time to  remission
IL-2Kuuliala et al, 2005 (13)1576 monthsNonbiologic DMARDsOR 4.7 (95% CI  1.4–15.2)No dataSex, RF, ESR, SJC, TJC, combination therapy
Serum RANKLGonzález-Alvaro et al, 2007 (21)7530 weeksBiologic DMARDs  (inflix, ada)OR 0.993 (95% CI 0.987–0.999)0.037DAS28
Shared epitope (DRB1*04, DRB1*13, and DQB1)Molenaar et al, 2002 (10)1672 yearsNonbiologic DMARDsNo association DMARD use, RF
MTHFR 677T allelesKurzawski et al, 2007 (15)1741 yearNonbiologic DMARDsOR 5.2 (95% CI  2–13.9)< 0.001
MTHFR 1298C alleles    OR 2.9 (95% CI  1.2–7)< 0.05
MTR 2756A allele ± either the SLC 19A180A allele or the TYMS 3R-del6 haplotypeJames et al, 2008 (17)9812 monthsNonbiologic DMARDsOR 3.4 (95% CI  1.1–10)0.04
Initial Sharp score <4Gossec et al, 2004 (11)1343 yearsNonbiologic DMARDsOR 2.9 (95% CI  1.2–7)≤ 0.02DAS, morning stiffness, HAQ score  <1.25, Ritchie's score
Lower TSS per 10 unitsVan der Heijde et al, 2007 (20)3323 yearsBiologic DMARDs  (etan ± MTX)OR 1.1 (95% CI  1.0–1.1)< 0.05Sex, disease duration, DAS, HAQ,  RF, previous MTX use

Clinical characteristics.

Sex.

Eleven studies examined the effect of sex on remission. Male sex was shown to be an independent predictor of remission in 5 studies, of which patients in one study were treated with nonbiologic DMARDs (16), and the other 4 were anti-TNF cohorts (19, 20, 23, 24). Forslind et al (16) found that male sex was a predictor of both point remission (18, 24, and 60 months) and persistent remission (18 and 24 months, 24 and 60 months, and 18 and 24 and 60 months), with ORs ranging from 1.53–2.87. Hyrich et al (19) found that women were significantly less likely to achieve remission compared with men in both etanercept- and infliximab-treated patients (etanercept: OR 0.6, 95% CI 0.38–0.94; infliximab: OR 0.6, 95% CI 0.4–0.89). Data from the 3-year TEMPO trial showed that male sex was an independent predictor of remission in patients receiving etanercept (OR 1.92, 95% CI 1.32–2.77). In the ReAct trial, a postmarketing cohort of adalimumab (23), male sex was found to be a significant predictor of remission during 3 years of followup (HR 1.28, 95% CI 1.16–1.42). A cohort of 3 anti-TNF agents (etanercept, infliximab, and adalimumab) (24) confirmed the influence of male sex on remission, but the detailed effect size and point estimate were not provided. Six studies found that sex was not an independent predictor of remission when adjusted for other variables (9, 13, 14, 18, 22, 25).

Age and age at onset.

Age has been shown to be a significant predictor of remission in 2 anti-TNF cohorts. Mancarella et al (24) found that patients age ≥53 years were less likely to achieve remission (OR 0.64, 95% CI 0.44–0.94) when adjusted for sex, the presence of rheumatoid factor (RF), and baseline functional status. Burmester et al (the ReAct trial) (23) found that the older the patient, the less likely to achieve remission up to 3 years of followup (HR 0.78, 95% CI 0.7–0.87 for age 40 to ≤65 years versus age <40 years; HR 0.68, 95% CI 0.58–0.78 for age >65 to ≤75 years versus age <40 years; and HR 0.61, 95% CI 0.46–0.81 for age >75 years versus age <40 years). However, this result was not confirmed by other anti-TNF studies (18, 25) and the FIN-RACo study (9).

Late-onset RA, i.e., patients age >65 years at onset, has been shown to be an independent predictor of remission in patients treated with nonbiologic DMARDs at any point within 5 years (OR 2.99, 95% CI 1.77–5.02) (8). However, the influence of age at disease onset on remission was not found in both men and women in the study by Forslind et al (16).

Disease duration.

Forslind et al found that patients who had a longer disease duration were less likely to achieve remission at all points of assessment, both point remission (OR 0.91–0.93, P < 0.02) and persistent remission (OR 0.87–0.91, P ≤ 0.004) (16). Other anti-TNF cohorts failed to demonstrate the influence of disease duration on remission (20, 22, 23, 25).

Smoking status.

The association between smoking and remission was investigated in a non–anti-TNF cohort and 2 anti-TNF cohorts. Current or previous smokers were less likely to achieve remission at 60 months (OR 0.67, 95% CI 0.45–0.99) and at 24 and 60 months of followup (OR 0.62, 95% CI 0.42–0.95) in non–anti-TNF cohorts (16). Current smokers were also less likely to reach remission in patients treated with infliximab, but not etanercept, in the British Society for Rheumatology Biologics Register (BSRBR; OR 0.77, 95% CI 0.6–0.99) (19), whereas the history of smoking (ever versus never) was not a significant predictor of remission in the ReAct trial (HR 1.08, 95% CI 0.987–1.17) (23).

Comorbidity.

Burmester et al (the ReAct trial) showed that patients having more than one comorbidity were less likely to achieve remission (HR 0.85, 95% CI 0.78–0.93) (23), whereas Hyrich et al failed to demonstrate a significant association between comorbidity and remission in both etanercept- and infliximab-treated patients (19).

Disease activity.

Gossec et al (11) found that patients who had baseline DAS of <4 or a Ritchie Articular Index of <17 were more likely to achieve remission at 3 years of followup, with an OR of 5.7 (95% CI 2.3–14.2) and an OR of 2.7 (95% CI 1.1–6.7) for the DAS and Ritchie Articular Index, respectively. Vázquez et al (14) confirmed this finding when using a DAS28 <5.1 in several models (OR 4.1–4.68). Verstappen et al (12) found that patients who had more severe pain at baseline were less likely to achieve remission (β = 0.98, 95% CI 0.97–0.99). Forslind et al (16) found that patients with a high baseline DAS28 were less likely to achieve remission in all models at any time points of assessment (OR 0.67–0.81, P < 0.006). Five anti-TNF cohorts using the DAS or DAS28 showed similar results (19–23). Burmester et al (the ReAct trial) (23) also demonstrated that the higher the baseline DAS28, the lesser chance to achieve remission (HR 0.27, 95% CI 0.26–0.31 for DAS28 >7 versus ≤5.1; and HR 0.5, 95% CI 0.46–0.55 for DAS28 >5.1 versus ≤5.1).

Functional status.

Baseline functional status measured by the Health Assessment Questionnaire (HAQ) has been shown to be inversely associated with remission in several studies. In a nonbiologic DMARD cohort, the HAQ was an independent predictor of remission in all models of both point remission and persistent remission (OR 0.56–0.7, P < 0.03) in one study (16). However, this association was not confirmed in very early RA cohorts in the study by Gossec et al (OR 2.3, 95% CI 0.9–5.7) (11). Five anti-TNF cohorts showed similar results even when it was adjusted for different variables (19, 20, 22–24). Burmester et al (the ReAct trial) (23) demonstrated an inverse relationship between the intensity of baseline functional impairment and remission (HR 0.6, 95% CI 0.53–0.68; HR 0.74, 95% CI 0.66–0.83; and HR 0.87, 95% CI 0.78–0.97 for HAQ score >2, HAQ score 1.5–2, and HAQ score <1.5, respectively).

Treatment.

In the FIN-RACo trial, a nonbiologic DMARD trial, the patients treated with combination therapy were more likely to achieve remission (OR 4.4, 95% CI 1.6–12.2) (13). On the other hand, patients who received single DMARD therapy and delayed treatment of more than 4 months after the onset of symptoms were less likely to achieve remission (35% versus 11% remission; P = 0.021) (9).

The US case–control study found that the use of anti-TNF is an independent predictor of remission (OR 2.74, 95% CI 1.4–5.34) when adjusted for age, sex, disease duration, nonsteroidal antiinflammatory drug (NSAID) and prednisolone use, and previous DMARD use (25).

In the patients receiving anti-TNF treatment, the previous number of DMARDs or anti-TNF use and the concurrent use of NSAIDs, MTX, and non-MTX DMARDs were predictors of remission. The results from the BSRBR showed that the previous number of DMARDs had an inverse association with remission in etanercept-treated patients (OR 0.83, 95% CI 0.7–0.97) and infliximab-treated patients (OR 0.85, 95% CI 0.75–0.98) (19). A similar association was found in a US case–control study (OR 0.77, 95% CI 0.62–0.9) (25). However, this was not confirmed by the ReAct and TEMPO trials.

The history of previously failed anti-TNF, etanercept, or infliximab, but not the history of previously failed DMARDs, was also inversely associated with remission (HR 0.83, 95% CI 0.71–0.96) in the ReAct trial (23). The concurrent use of NSAIDs has been shown to be a significant predictor in the BSRBR (OR 1.79, 95% CI 1.07–2.99) (19) and the South Swedish Arthritis Treatment Group Register (OR 1.48, P = 0.04) (22).

With regard to the concurrent use of DMARDs in anti-TNF–treated patients, the results of the South Swedish Arthritis Treatment Group Register using 3 anti-TNF agents (etanercept, infliximab, and adalimumab) showed a significant increased chance of remission with the concurrent use of MTX or non-MTX DMARDs (OR 1.97, P < 0.01 and OR 1.96, P = 0.01, respectively) (22). However, the benefit of the concurrent use of MTX was confirmed only in patients receiving etanercept (18, 20), but not infliximab, in other studies (19). The benefit of the concurrent use of non-MTX DMARD therapy could not be demonstrated in etanercept and infliximab in the BSRBR (18).There were no data of MTX analyzed separately from non-MTX DMARDs in the adalimumab cohort of the ReAct trial. This cohort showed that concomitant use of DMARDs (both MTX and non-MTX DMARDs) with adalimumab increased the chance to achieve remission (HR 1.34, 95% CI 1.21–1.48) (23).

Response to treatments.

Moderate response at the first 6 months of the treatments measured by the ACR 50% improvement criteria (OR 3.9, 95% CI 1.14–13.38) and good response (European League Against Rheumatism [EULAR] good response) at the first 6 months (OR 6.23, 95% CI 1.61–24.04) (14) or 1 year of the treatments (β = 4.7, 95% CI 3.2–7.03; P = 0.0001) (12) were the significant predictors of remission.

Laboratory and radiography.

RF.

RF status has been shown to be inversely associated with remission in several studies. Verstappen et al found that the absence of RF was an independent predictor of remission at 4 years of followup (β = 1.63, 95% CI 1.15–2.32; P = 0.061) (12). Pease et al, Forslind et al, and Mancarella et al found that patients who had positive RF at baseline were less likely to achieve remission (OR 0.47, 95% CI 0.28–0.77; OR 0.47–0.58 [in multiple models]; and OR 0.61, 95% CI 0.39–0.96, respectively) (8, 16, 24). However, the predictive value of baseline RF status disappeared in some studies when it was adjusted for anti–citrullinated peptide (anti-CCP) (16), treatment strategy (combination therapy of nonbiologic DMARDs versus single DMARD therapy) (9), treatment with biologic agents (19, 20), and the presence of shared epitopes (10).

Anti-CCP

The predictive value of anti-CCP was studied by Forslind et al. The anti-CCP status at baseline was inversely associated with remission at 24 months (OR 0.63, 95% CI 0.45–0.89) when adjusted for the DAS28, disease duration, the HAQ, and male sex.

Acute-phase reactant

The result of the ReAct trial showed that the patients who had a baseline C-reactive protein level of ≥20 mg/liter were less likely to achieve remission (HR 0.84, 95% CI 0.77–0.91; P < 0.0001) (23). However, other studies failed to demonstrate the predictive value of other acute-phase reactants for remission (8, 11, 14, 22).

Serum interleukin-2 (IL-2)

The predictive value of serum IL-2 was investigated in the FIN-RACo trial, a randomized controlled trial comparing combination therapy of MTX plus hydroxychloroquine plus sulfasalazine (SSZ) plus prednisolone and single DMARD. A serum IL-2 level of <442 units/ml at baseline was an independent predictor of remission at 6 months (OR 4.7, 95% CI 1.4–15.2) when adjusted for treatment regimens, sex, RF status, erythrocyte sedimentation rate, and the number of tender and swollen joints.

Serum RANKL (sRANKL)

A Spanish group investigated the predictive value of RANKL and osteoprotegerin serum levels for the therapeutic response to 2 anti-TNFs: infliximab and adalimumab (data from the ReAct trial were used for adalimumab) (21). They found that the sRANKL levels and RANKL/osteoprotegerin ratio in the patients who achieved remission were significantly lower at baseline than in the remaining patients at both 3 and 7 months of followup. Both lower DAS28 values (OR 0.314, 95% CI 0.109–0.904; P = 0.032) and sRANKL levels (OR 0.993, 95% CI 0.987–0.999; P = 0.037) were significantly and independently associated with remission. However, the upper end of the 95% CI was close to 1 for the sRANKL levels.

Genetic studies

Two studies concerning genetic markers related to the folate pathway, a site of action of MTX and SSZ, found that MTHFR 677T alleles and 1298C alleles were significantly associated with an increased rate of remission in RA patients treated with MTX (OR 5.21, 95% CI 1.95–13.93 and OR 2.91, 95% CI 1.21–7.02, respectively) (15). The MTR 2756A allele ± either the SLC 19A180A allele or the TYMS 3R-del6 haplotype was also associated with remission in RA patients treated with the combination therapy of MTX plus SSZ (OR 3.4, 95% CI 1.1–10) (17), whereas Molenaar et al (10) found that the shared epitopes, both predisposing (HLA–DQB1/DQA1) and protecting (HLA–DRB1) alleles for RA, were not associated with RA remission when adjusted for RF and DMARD use.

Radiography

The results of the early RA cohort showed that an initial Sharp score of <4 was an independent predictor of remission at 3 years when adjusted for the DAS, morning stiffness, a HAQ score <1.25, and Ritchie's score (OR 2.9, 95% CI 1.2–7) (11). The results from the TEMPO trial also showed that the chance to achieve remission increases with every 10 units decreasing of the baseline total Sharp score (OR 1.08, 95% CI 1.03–1.11; P < 0.05) (20).

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES
  9. Supporting Information

We systematically summarized the clinical and laboratory predictors of RA remission. We included only studies using multivariate analysis to identify predictors to ensure that these factors independently predict remission. The predictive value of most predictors is plausible and consistent among the studies. The association is also explainable by and consistent with the natural history of RA. Most predictors represent the disease severity of RA at baseline, e.g., disease activity, functional status, history of failed DMARDs, the presence of RF and anti-CCP, acute-phase reactant, or evidence of radiographic damage. Genetic markers are also associated with the response to MTX treatment. Therefore, these factors should be associated with RA outcome. Recently, smoking has been shown to be associated with the presence and titer of anti-CCP (26, 27). The predictive value of smoking status for remission from 2 studies in this review is controversial and needs to be confirmed in future research. Although several studies consistently found that elderly and female patients were less likely to achieve remission, this should be cautiously interpreted due to the limited validity of RA disease activity measures and remission criteria in these populations. RA disease activity measures and remission criteria comprise patient-reported outcomes, including pain score or patient global assessment of disease activity, which sometimes may unreliably reflect their disease activity. For example, it has been shown that women with RA were more likely than men to report their symptoms and functional status as more severe despite the equivalent number of painful and swollen joint counts (28, 29) or disease activity (DAS28) (30). Additionally, in a cohort of the Central Finland RA database, the majority of the non-RA community population at age >50 years who should be identified as being in remission by the ACR remission criteria did not meet the criteria for ACR remission in RA (31). These RA disease activity measures and remission criteria may not accurately identify RA remission in these populations. Furthermore, the remission criteria of the composite index themselves are problematic regarding the face validity. For example, only 70% of patients classified as being in remission by the DAS28 truly have no swollen joint count (32). These criteria may not truly represent remission in the rheumatologist's perspective.

The value of evaluating the treatment regimens or strategies in this review is that the results of this review confirm that early aggressive treatment with the combination of nonbiologic DMARDs or anti-TNF agents aiming to achieve at least moderate or good response at the first 6–12 months of the treatment improves RA outcomes regardless of having the other poor prognostic factors. The benefits of NSAIDs may reflect the relative absence of comorbidities in patients who can tolerate these drugs or the continuing presence of reversible inflammatory symptoms.

There are a number of real and artifactual causes of between-study variation in included studies hampering statistically pooling and interpreting the data. Previously, there was no consensus on the outcome measures that should be used to assess the disease activity and the response to the treatments in RA trials; therefore, the remission criteria and outcome measures for prognostic factors used are diverse and may cause inconsistent results among the studies. Recently, the Outcome Measures in Rheumatology Clinical Trials group (33) and the ACR and EULAR (34) provided recommendations on how to report disease activity in clinical trials of RA. These recommendations will harmonize the reporting of results from clinical trials and facilitate the comparison of outcomes across different trials and pooling of trial results.

The time of outcome assessment and factors used to adjust for in the model also varied. The new studies may adjust for variables that are recently reported to be associated with RA outcome. For example, smoking status has not been used to adjust for in the model of the studies done before 2006. Anti-CCP was used to adjust for in the model in only one study in 2007. All of these variations can cause inconsistent results, both in the directions and magnitudes. Limited sample size is another factor causing inconsistent results. Some studies may be underpowered to demonstrate the predictive value of some significant variables.

The treatment is an important factor that modifies the clinical course and outcomes. The predictive value of a variable may be changed when the treatment is taken into account in the model. The presence of RF is one of variables that were independently associated with remission only in the trials that were not adjusted for the treatments.

Several factors were recently identified and investigated, such as anti-CCP, smoking, genetic markers, IL-2, and sRANKL, so their predictive values were evaluated in a few studies. Although the results were impressive, they were studied in a specific population, which may limit their generalizability. These findings need to be confirmed in other populations and settings. In addition, we should cautiously interpret the results of the genetic studies included in this review. Although the multivariate analysis was used, it was adjusted for only other genetic markers. Other important clinical and laboratory characteristics that may be associated with these genetic markers were not taken into account in the model.

Although all of the studies included in this review investigated an independent association using multivariate analysis, the hypotheses of these studies were usually based on the unreal assumption that the association between the prognostic factors and RA remission is direct and isolated. This model may be inadequate to explicitly describe the complex relationship between prognostic factors and remission for multifactorial and unclear mechanisms of a disease condition like RA (35).

We did not include the abstracts from the ACR and EULAR Annual Scientific Meetings because we were aware of the limited data available in the abstracts that may hinder our ability to assess the study quality and the quality of data, as well as to abstract complete information.

In conclusion, we identified a number of clinical, laboratory, and genetic markers that are independent predictors of clinical remission in RA patients. However, when treatment strategy is taken into account, early aggressive treatment improves RA outcomes regardless of having the other poor prognostic factors. To better understand the complex prognostic pathways or processes of RA, these relationships need to be further investigated in phase 3 explanatory prognostic studies.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES
  9. Supporting Information

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be submitted for publication. Dr. Katchamart 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 conception and design. Katchamart, Johnson, Lin, Phumethum, Salliot, Bombardier.

Acquisition of data. Katchamart, Johnson, Lin, Phumethum, Salliot, Bombardier.

Analysis and interpretation of data. Katchamart, Johnson, Bombardier.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES
  9. Supporting Information
  • 1
    Bykerk V, Keystone E. What are the goals and principles of management in the early treatment of rheumatoid arthritis? Best Pract Res Clin Rheumatol 2005; 19: 14761.
  • 2
    Fransen J, Moens HB, Speyer I, van Riel PL. Effectiveness of systematic monitoring of rheumatoid arthritis disease activity in daily practice: a multicentre, cluster randomised controlled trial. Ann Rheum Dis 2005; 64: 12948.
  • 3
    Verstappen SM, Jacobs JW, van der Veen MJ, Heurkens AH, Schenk Y, ter Borg EJ, et al. Intensive treatment with methotrexate in early rheumatoid arthritis: aiming for remission. Computer Assisted Management in Early Rheumatoid Arthritis (CAMERA, an open-label strategy trial). Ann Rheum Dis 2007; 66: 14439.
  • 4
    Grigor C, Capell H, Stirling A, McMahon AD, Lock P, Vallance R, et al. Effect of a treatment strategy of tight control for rheumatoid arthritis (the TICORA study): a single-blind randomised controlled trial. Lancet 2004; 264: 2639.
  • 5
    Alonso-Ruiz A, Pijoan J, Ansuategui E, Urkaregi A, Calabozo M, Quintana A. Tumor necrosis factor α drugs in rheumatoid arthritis: systematic review and metaanalysis of efficacy and safety. BMC Musculoskelet Disord 2008; 9: 52.
  • 6
    Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF, Cooper NS, et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988; 31: 31524.
  • 7
    Hayden JA, Cote P, Bombardier C. Evaluation of the quality of prognosis studies in systematic reviews. Ann Intern Med 2006; 144: 42737.
  • 8
    Pease CT, Bhakta BB, Devlin J, Emery P. Does the age of onset of rheumatoid arthritis influence phenotype? A prospective study of outcome and prognostic factors. Rheumatology (Oxford) 1999; 38: 22834.
  • 9
    Mottonen T, Hannonen P, Korpela M, Nissila M, Kautiainen H, Ilonen J, et al, for the FIN-RACo Trial Group. Delay to institution of therapy and induction of remission using single-drug or combination–disease-modifying antirheumatic drug therapy in early rheumatoid arthritis. Arthritis Rheum 2002; 46: 8948.
  • 10
    Molenaar ET, Voskuyl AE, van der Horst-Bruinsma IE, Schreuder GM, Zanelli E, Dijkmans BA. Influence of HLA polymorphism on persistent remission in rheumatoid arthritis. Ann Rheum Dis 2002; 61: 3513.
  • 11
    Gossec L, Dougados M, Goupille P, Cantagrel A, Sibilia J, Meyer O, et al. Prognostic factors for remission in early rheumatoid arthritis: a multiparameter prospective study. Ann Rheum Dis 2004; 63: 67580.
  • 12
    Verstappen SM, van Albada-Kuipers GA, Bijlsma JW, Blaauw AA, Schenk Y, Haanen HC, et al. A good response to early DMARD treatment of patients with rheumatoid arthritis in the first year predicts remission during follow up. Ann Rheum Dis 2005; 64: 3843.
  • 13
    Kuuliala A, Leirisalo-Repo M, Mottonen T, Hannonen P, Nissila M, Kautiainen H, et al. Serum soluble interleukin-2 receptor predicts early remission in patients with recent-onset rheumatoid arthritis treated with a single disease-modifying antirheumatic drug. Clin Exp Rheumatol 2005; 23: 2436.
  • 14
    Vazquez I, Graell E, Gratacos J, Canete JD, Vinas O, Ercilla MG, et al. Prognostic markers of clinical remission in early rheumatoid arthritis after two years of DMARDs in a clinical setting. Clin Exp Rheumatol 2007; 25: 2318.
  • 15
    Kurzawski M, Pawlik A, Safranow K, Herczynska M, Drozdzik M. 677C>T and 1298A>C MTHFR polymorphisms affect methotrexate treatment outcome in rheumatoid arthritis. Pharmacogenomics 2007; 8: 15519.
  • 16
    Forslind K, Hafstrom I, Ahlmen M, Svensson B, and the BARFOT Study Group. Sex: a major predictor of remission in early rheumatoid arthritis? Ann Rheum Dis 2007; 66: 4652.
  • 17
    James HM, Gillis D, Hissaria P, Lester S, Somogyi AA, Cleland LG, et al. Common polymorphisms in the folate pathway predict efficacy of combination regimens containing methotrexate and sulfasalazine in early rheumatoid arthritis. J Rheumatol 2008; 35: 56271.
  • 18
    Hyrich KL, Symmons DP, Watson KD, Silman AJ, on behalf of the British Society for Rheumatology Biologics Register. Comparison of the response to infliximab or etanercept monotherapy with the response to cotherapy with methotrexate or another disease-modifying antirheumatic drug in patients with rheumatoid arthritis: results from the British Society for Rheumatology Biologics Register. Arthritis Rheum 2006; 54: 178694.
  • 19
    Hyrich KL, Watson KD, Silman AJ, Symmons DP, for the British Society for Rheumatology Biologics Register. Predictors of response to anti-TNF-α therapy among patients with rheumatoid arthritis: results from the British Society for Rheumatology Biologics Register. Rheumatology (Oxford) 2006; 45: 155865.
  • 20
    Van der Heijde D, Klareskog L, Landewe R, Bruyn GA, Cantagrel A, Durez P, et al, for the TEMPO Study Investigators. Disease remission and sustained halting of radiographic progression with combination etanercept and methotrexate in patients with rheumatoid arthritis. Arthritis Rheum 2007; 56: 392839.
  • 21
    Gonzalez-Alvaro I, Ortiz AM, Tomero EG, Balsa A, Orte J, Laffon A, et al. Baseline serum RANKL levels may serve to predict remission in rheumatoid arthritis patients treated with TNF antagonists. Ann Rheum Dis 2007; 66: 16758.
  • 22
    Kristensen LE, Kapetanovic MC, Gulfe A, Soderlin M, Saxne T, Geborek P. Predictors of response to anti-TNF therapy according to ACR and EULAR criteria in patients with established RA: results from the South Swedish Arthritis Treatment Group Register. Rheumatology (Oxford) 2008; 47: 4959.
  • 23
    Burmester GR, Ferraccioli G, Flipo RM, Monteagudo-Saez I, Unnebrink K, Kary S, et al. Clinical remission and/or minimal disease activity in patients receiving adalimumab treatment in a multinational, open-label, twelve-week study. Arthritis Rheum 2008; 59: 3241.
  • 24
    Mancarella L, Bobbio-Pallavicini F, Ceccarelli F, Falappone PC, Ferrante A, Malesci D, et al. Good clinical response, remission, and predictors of remission in rheumatoid arthritis patients treated with tumor necrosis factor-α blockers: the GISEA study. J Rheumatol 2007; 34: 16703.
  • 25
    Liang GC, Cordero M, Dyer A, Chang RW. Current tumor necrosis factor-α inhibitor use is associated with a higher probability of remissions in patients with rheumatoid arthritis. J Rheumatol 2005; 32: 16625.
  • 26
    Lee DM, Phillips R, Hagan EM, Chibnik LB, Costenbader KH, Schur PH. Quantifying anti-cyclic citrullinated peptide titres: clinical utility and association with tobacco exposure in patients with rheumatoid arthritis. Ann Rheum Dis 2009; 68: 2018.
  • 27
    Linn-Rasker SP, van der Helm-van Mil AH, van Gaalen FA, Kloppenburg M, de Vries RR, le Cessie S, et al. Smoking is a risk factor for anti-CCP antibodies only in rheumatoid arthritis patients who carry HLA-DRB1 shared epitope alleles. Ann Rheum Dis 2006; 65: 36671.
  • 28
    Katz P, Criswell L. Differences in symptom reports between men and women with rheumatoid arthritis. Arthritis Care Res 1996; 9: 4418.
  • 29
    Sokka T, Toloza S, Cutolo M, Kautiainen H, Makinen H, Gogus F, et al, and the QUEST-RA Group. Women, men, and rheumatoid arthritis: analyses of disease activity, disease characteristics, and treatments in the QUEST-RA study. Arthritis Res Ther 2009; 11: R7.
  • 30
    Deighton CM, Surtees D, Walker DJ. Influence of the severity of rheumatoid arthritis on sex differences in Health Assessment Questionnaire scores. Ann Rheum Dis 1992; 51: 4735.
  • 31
    Sokka T, Makinen H, Hannonen P, Pincus T. Most people over age 50 in the general population do not meet ACR remission criteria or OMERACT minimal disease activity criteria for rheumatoid arthritis. Rheumatology (Oxford) 2007; 46: 10203.
  • 32
    Aletaha D, Smolen J. Remission of rheumatoid arthritis: should we care about definitions? Clin Exp Rheumatol 2006; 24 Suppl 43 : S4551.
  • 33
    Outcome measures in rheumatoid arthritis clinical trials. J Rheumatol 1993; 20: 52691.
  • 34
    Aletaha D, Landewe R, Karonitsch T, Bathon J, Boers M, Bombardier C, et al. Reporting disease activity in clinical trials of patients with rheumatoid arthritis: EULAR/ACR collaborative recommendations. Arthritis Rheum 2008; 59: 13717.
  • 35
    Hayden JA, Cote P, Steenstra IA, Bombardier C. Identifying phases of investigation helps planning, appraising, and applying the results of explanatory prognosis studies. J Clin Epidemiol 2008; 61: 55260.

Supporting Information

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES
  9. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
ACR_20188_sm_appendix.doc85KSupplementary Data

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.