SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

Objective

Recently, new classification criteria for rheumatoid arthritis (RA) have been devised by methodology that used first a quantitative approach (data from databases), then a qualitative approach (consensus; based on paper patients), and finally a common sense–based approach (evaluation of the former phases). Now the individual items that make up these criteria are being evaluated. This study was undertaken to analyze the item “autoantibodies,” in particular rheumatoid factor (RF) level.

Methods

Three separate cohorts comprising a total of 972 patients with undifferentiated arthritis were studied for RA development (according to the 1987 American College of Rheumatology criteria) and arthritis persistence. The positive predictive value (PPV), negative predictive value (NPV), and likelihood ratios (LRs) were compared between different levels of RF and the presence of anti–citrullinated protein antibody (ACPA). A similar comparison was made in 686 RA patients for the rate of joint destruction and achievement of sustained disease-modifying antirheumatic drug–free remission during 7 years of followup. The variation in RF levels obtained by different measurement methods in the same RF-positive sera was explored.

Results

Compared to high RF levels, presence of ACPA had a better balance between positive LR and negative LR and between PPV and NPV for RA development. The additive value of ACPA assessment after testing for RF level was higher than vice versa. The association between high RF level and RA severity was not as strong as that between ACPA antibodies and RA severity. The RF level obtained by different methods in the same patients' sera varied considerably.

Conclusion

Our findings indicate that determination of RF level is subject to large variation; high RF level has limited additive prognostic value compared to ACPA positivity. Thus, omitting RF level and using RF presence, ACPA presence, and ACPA level may improve the 2010 criteria for RA.

Recently, the American College of Rheumatology (ACR) classification criteria for rheumatoid arthritis (RA), which were developed in 1987 (1), have been subjected to review by a joint task force of the ACR and the European League Against Rheumatism (EULAR). The aim of the review was to enable RA classification at an earlier disease stage compared to the 1987 ACR criteria, and the development of new criteria is an important step forward.

The development of the 2010 ACR/EULAR criteria comprised 3 phases. The first was a data-driven phase using findings in 3,115 patients from Europe and Canada. The second phase incorporated the expertise of 39 rheumatologists, and the third phase was a consensus phase undertaken by the same group (2–4). In coming years, the criteria will be studied in cohorts with different ethnic backgrounds and in dissimilar health care systems, in which the pretest probability for RA in new patients visiting rheumatologists differs.

The 2010 criteria are the first to include anti–citrullinated protein antibodies (ACPA) in addition to rheumatoid factor (RF). Presence of these autoantibodies can contribute substantially to the diagnosis of RA, for which ≥6 points are required; presence of ACPA or RF yields 2 points, and a high level of ACPA or RF yields 3 points. In the data-driven phase of criteria development, using data from several early arthritis cohorts, ACPA and RF were recognized as a theme in a factor analysis. Then, ACPA and RF were summarized as “serology.” Subsequently, the importance of serology, independent of other variables, was determined using a multivariate regression analysis. It was observed that within the group of patients with a positive serology, a level higher than the median received a higher weight than a level lower than the median. After the expert phase and the consensus phase, a high level was redefined as ≥3 times the reference value.

The present study investigated 2 main characteristics of the items defined as “serology,” particularly the RF level criterion, in the 2010 ACR/EULAR criteria for RA. The first characteristic was the discriminative ability of high levels of RF compared to ACPA for identifying early RA. Several studies have demonstrated an increased specificity for RA of a higher RF level compared to RF positivity (5, 6). However, an increased specificity for RA has also been observed for the presence of ACPA compared to the presence of RF (7). Thus far, extensive comparisons of the ability of increased RF levels to predict RA development compared with the ability of the presence of ACPA, notably anti–cyclic citrullinated peptide (anti-CCP) antibodies, to predict RA development have not been made. In 3 separate prospective cohorts of patients with undifferentiated arthritis (UA) of recent onset from 3 different countries, RA development was studied in relation to baseline RF levels and ACPA. RA was diagnosed according to the 1987 ACR criteria (1). To verify that the results were not different when other outcome measures were used, analyses in patients with UA were repeated with arthritis persistence as the outcome measure. Furthermore, the same analyses were performed in RA patients, with the rate of joint destruction and the achievement of sustained disease-modifying antirheumatic drug (DMARD)–free remission as outcomes.

The second characteristic was the capacity of different assays to uniformly define a high RF level. Despite the existence of international units for RF, RF level measurement is not adequately standardized between different methods. Subsequent variations in RF levels may yield differences between laboratories with regard to the classification or diagnosis of RA. Therefore, we determined the degree of variation in RF levels obtained when the same RF-positive serum samples were tested by the methods that are currently most frequently applied (enzyme-linked immunosorbent assay [ELISA], nephelometry, and turbidimetry). Although previous studies have evaluated the correlations between results of the Rose-Waaler method and ELISA (8), data on head-to-head comparisons of currently applied methods are, to the best of our knowledge, not available.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

Patients.

Development of RA in patients with UA.

Patients with UA from 3 separate cohorts were studied for RA development, comprising an overall total of 972 patients with UA (Figure 1). UA was defined as not fulfilling any of the existing classification criteria for a rheumatic disease diagnosis 2 weeks after the first presentation when the results of laboratory and radiologic examinations were known (9). Patients were followed up for 1 year, after which the final diagnosis was established. Patients were categorized as having RA (according to the 1987 ACR criteria [1]) or non-RA (all other diagnoses).

thumbnail image

Figure 1. Flow chart showing the cohorts investigated in this study. Patients in the Berlin Early Arthritis Cohort (EAC), Norwegian Very Early Arthritis Clinic (NOR-VEAC), and Leiden EAC who were initially diagnosed with undifferentiated arthritis (UA) were studied for development of rheumatoid arthritis (RA). In the Berlin cohort and the NOR-VEAC, data on rheumatoid factor (RF) and anti–citrullinated protein antibody (ACPA) were available for all UA patients. In the Leiden EAC, data on RF were available for 623 patients, and data on ACPA were available for 624 patients. In the Leiden EAC, a total of 687 patients were diagnosed as having RA after 1 year. Of these patients, radiographic data and/or data on sustained disease-modifying antirheumatic drug–free remission status were available for 686 patients. Data on RF and ACPA were available for 663 and 658 patients, respectively.

Download figure to PowerPoint

The Leiden Early Arthritis Cohort (EAC) is a large prospective cohort that was started in 1993 and has been described previously (10). Patients with confirmed arthritis were included when the duration of symptoms was <2 years. At baseline, blood samples were obtained for routine diagnostic laboratory screening (including testing for IgM-RF) and stored for determining the presence of other autoantibodies later (anti–CCP-2). Followup examinations (which included obtaining radiographs) were performed yearly. Between 1993 and 2006, 625 patients were diagnosed as having UA at baseline. Almost all patients had a followup duration of >1 year. Approximately thirty percent of the patients with UA had developed RA by 1 year of followup, and an additional 4% developed RA after >1 year of followup (11).

The Berlin EAC was started in January 2004, and patients were included if they had synovitis in at least 2 joints and a duration of symptoms of between 4 weeks and 12 months. This Berlin cohort has been described previously (12). At first presentation, 154 patients had UA. Fulfillment of the 1987 ACR criteria for RA (1) was assessed after 1 year of followup.

The third cohort consisted of 193 patients with UA from Oslo, Norway, who were included in the Norwegian Very Early Arthritis Clinic (NOR-VEAC) (13). This cohort included patients who had swelling in at least 1 joint with a symptom duration of <16 weeks. During the first year, patients were seen after 3, 6, and 12 months, and the development of RA was classified after 1 year of followup.

In the first, data-driven phase of the development of the new 2010 ACR/EULAR criteria, findings in patients from the Leiden EAC (n = 213) and from the NOR-VEAC (n = 193) were used (3). All studies were approved by the local ethics committees. All patients provided written informed consent.

Persistence of arthritis in patients with UA.

In order to determine whether the results differed when a different outcome measure was used, analyses were repeated in the Leiden data set with arthritis persistence as the outcome. A generally accepted definition for persistence is lacking, and its frequency depends on the observation period. We defined persistent arthritis as the absence of sustained remission, which was defined as the absence of swollen joints for ≥1 year after cessation of eventual DMARD therapy. When remission was not obtained after 5 years of disease, a patient was classified as having persistent arthritis. According to this definition, 61.3% of patients with UA had persistent arthritis.

Severity of the disease course in RA patients.

Patients who fulfilled the 1987 ACR criteria for RA during the first year and were included in the Leiden EAC between 1993 and 2006 were studied. Of the total of 687 RA patients, 486 had already fulfilled the 1987 ACR criteria for RA at baseline, and 201 developed RA within the first year of followup (Figure 1).

Radiographs of the hands and feet were obtained at baseline and in consecutive years in 672 patients with RA. These radiographs were scored chronologically by an experienced reader (MPMvdL) according to the Sharp/van der Heijde method (14). Intraclass correlation coefficients were 0.91 for all radiographs, 0.84 for baseline radiographs, and 0.97 for the radiographic progression rate. To encompass a reliable sample size, radiographic followup data were restricted to a maximum of 7 years (median 5 years [interquartile range 2–7 years]). Treatment strategies for RA changed over time and became more aggressive in the subsequent inclusion periods (1993–1996, 1996–1998, and 1999–2006) (15).

A second outcome measure for the severity of the disease course was the achievement of sustained DMARD-free remission. Remission was defined in a stringent manner as the persistent absence of synovitis, e. g., no swollen joints, for ≥1 year after cessation of DMARD therapy and the identification of remission by the patient's rheumatologist (16). In this analysis, corticosteroids (both oral and intraarticular) were considered DMARDs; nonsteroidal antiinflammatory drugs were allowed. Most patients in whom remission was achieved had a followup period after cessation of DMARDs of >1 year. The remission status in 641 patients with RA could be reliably ascertained using medical files. The frequency of DMARD-free remission in these RA patients was 12.3% (16).

Autoantibody testing.

In the Leiden EAC, RF was determined by ELISA (IgM-RF; in-house ELISA [17]), using a standard cutoff value of 5 arbitrary units. Anti–CCP-2 autoantibodies (total IgG) were measured by ELISA (Immunoscan RA Mark 2; Euro-Diagnostica). The cutoff level for anti–CCP-2 autoantibody positivity was set at 25 arbitrary units, according to the recommendations of the manufacturer.

In the Berlin cohort, RF was determined by ELISA (Autostat II; Hycor Biomedical), using a reference value of >24 IU/liter for a positive test result. Anti–CCP-2 was determined by ELISA (Immunoscan CCPlus; Euro-Diagnostica), using a reference cutoff of >25 units/liter for autoantibody positivity.

In the NOR-VEAC, sera frozen at the time of enrollment were used to analyze anti–CCP-2 levels by ELISA (Inova) and IgM-RF levels by an in-house ELISA, in one batch. Cutoff levels used to define a positive status were those recommended by the local laboratory: 25 units/ml for anti–CCP-2 and 25 units/ml for IgM-RF.

Considering the absence of agreement on a uniform definition of high RF level, 2 definitions of high RF level were evaluated. These were 3 times the reference cutoff value (the definition of a high RF level that is used in the 2010 ACR/EULAR criteria), and an RF level of 50 units/ml (RF50) (the definition of high RF levels used in previous studies [5, 6]).

Analysis of variation in RF measurements.

In order to facilitate quality control in laboratories in The Netherlands, the Stichting Kwaliteitsbewaking Medische Laboratoria—Humoral Immunology Section organizes external quality assessment schemes for RF testing twice a year. In each scheme, 6 patient samples are sent to 78 participating laboratories. These 6 patient samples consist of 3 RF-negative samples, 2 RF-positive samples, and 1 standard serum (Reference Laboratory for Rheumatologic Serology [RELARES]). This is a commercially available standard serum, consisting of pooled serum from RF-positive patients, which was previously standardized to correspond with 100 IU using the Rose-Waaler agglutination test (18, 19). In this study, we used the results from the 2 RF-positive patient sera and the standard serum from the spring 2008 scheme. The sera were tested according to local protocols, and results were reported in local units and as a ratio compared to the local cutoff value.

Statistical analysis.

Development of RA in patients with UA.

Different test characteristics (sensitivity, specificity, positive likelihood ratio [LR], and negative LR) were determined. The LR incorporates both the sensitivity and the specificity of the test and provides an estimate of how much a test result will change the odds of having a disease. In addition, absolute posttest changes in RA status after 1 year of followup were determined (positive predictive value [PPV] and negative predictive value [NPV]). Analyses were performed using 2 descriptions of a high RF level (3 times the reference cutoff level and RF50), and the resulting data were compared with the data for ACPA positivity. RA development was analyzed after 1 year of followup, and arthritis persistence was classified after 5 years of followup.

Severity of the disease course in RA patients.

Associations with the rate of joint destruction during 7 years of followup were assessed using a repeated-measures analysis on log-transformed radiologic data, because of skewness. The repeated-measures analysis is performed using a multivariate normal regression model that, on longitudinal data, evaluates the progression rates over time and takes into account the correlation between the measurements within one subject. Adjustments were made for age, sex, and applied treatment strategy as previously described (20).

Analysis of sustained DMARD-free remission was performed by comparing Kaplan-Meier curves and by Cox regression analysis, correcting for age and sex, taking into account the differences in followup times among patients. For patients in whom remission was achieved, the dependent variable was “time-to-event,” indicating the time until remission was reached. For patients in whom remission was not achieved, the time to last followup was used.

Variation in RF measurements.

To test for correlations between the different methods that are used for measurement of the RF level, nonparametric Spearman's correlation coefficients (ρ) were determined.

SPSS software version 17.0 was used. P values less than 0.05 (2-tailed) were considered significant.

RESULTS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

Development of RA in patients with UA.

The baseline characteristics of the UA patients included in the 3 cohorts are presented in Table 1. The percentages of UA patients who developed RA within the first year were 32%, 48%, and 12% in the Leiden EAC, Berlin EAC, and NOR-VEAC, respectively.

Table 1. Baseline characteristics of the patients with early undifferentiated arthritis included in the different cohorts*
 Leiden EAC (n = 625)Berlin EAC (n = 154)NOR-VEAC (n = 193)
  • *

    Except where indicated otherwise, values are the mean ± SD. EAC = early arthritis cohort; NOR-VEAC = Norwegian Very Early Arthritis Clinic; IQR = interquartile range; RF = rheumatoid factor; ACPA = anti–citrullinated protein antibody.

  • 44 joints assessed.

  • 28 joints assessed.

  • §

    66 joints assessed.

  • Abnormal C-reactive protein (CRP) level was defined as ≥10 mg/liter.

  • #

    Abnormal CRP level was defined as >5 mg/liter.

Age at inclusion, years51.0 ± 16.951.2 ± 14.546.1 ± 14.5
Female, no. (%)368 (58.9)110 (71.4)114 (59.1)
Symptom duration at first presentation, days170 ± 181137.4 ± 96.135 ± 30
Swollen joint count5.5 ± 6.02.7 ± 4.53.9 ± 6.8§
Median (IQR) CRP, mg/liter17.0 (7.0-43.0)6.2 (2.0-16.8)#14.0 (5.0-32.0)
RF positive, no. (%)154 (24.7)79 (51.3)18 (9.3)
ACPA positive, no. (%)149 (23.9)44 (28.6)19 (9.8)

First, the predictive values for high RF levels and presence of ACPA antibodies were determined for each cohort separately (Table 2). Increasing the cutoff value for a high RF level yielded an increased PPV and decreased NPV. Similarly, the specificity increased, but the sensitivity decreased. For example, in the Leiden EAC, the PPV increased from 62% to 69% when a cutoff value of 3 times the reference value was used, and from 62% to 72% when a cutoff value of RF50 was used. The NPV decreased from 78% to 75% when a cutoff value of 3 times the reference value was used, and from 78% to 71% when a cutoff value of RF50 was used. Also, the specificity (of RF positivity) increased from 86% to 93% (using a cutoff value of 3 times the reference value) and from 86% to 97% (using a cutoff value of RF50), but the sensitivity obtained using each of these definitions of high RF decreased, from 48% to 33% and from 48% to 14%, respectively. In addition, the positive LR increased at the expense of an increased negative LR. This indicates that the odds of developing RA increased with a high RF level, but that the odds of developing RA increased in the absence of a high RF level as well. The percentage of patients with UA in the Leiden cohort who had a high RF level was 15% (when high RF level was defined as 3 times the reference value) or 6% (when high RF level was defined as RF50), compared to 25% of the UA patients who were RF positive. The observed effects were comparable for all 3 cohorts (Table 2).

Table 2. Comparison of different cutoff values for high RF level and the reference ACPA for predicting progression from UA to RA in 3 different cohorts*
Cohort and autoantibody test (cutoff value)No. (%) of UA patients with a positive test resultPPV, % (95% CI)NPV, % (95% CI)Positive LR (95% CI)Negative LR (95% CI)Sensitivity,% (95% CI)Specificity, % (95% CI)
  • *

    Three definitions of high rheumatoid factor (RF) level were used: the reference cutoff value (5.0 units/ml in the Leiden Early Arthritis Cohort [EAC], 24.0 units/ml in the Berlin EAC, and 25.0 units/ml in the Norwegian Very Early Arthritis Clinic [NOR-VEAC]), 3 times the reference cutoff value (15.0 units/ml in the Leiden EAC, 72.0 units/ml in the Berlin EAC, and 75.0 units/ml in the NOR-VEAC), and an absolute level of 50 units/ml. The reference cutoff value for anti–citrullinated protein antibody (ACPA) positivity was used in all 3 cohorts. UA = undifferentiated arthritis; RA = rheumatoid arthritis; PPV = positive predictive value; 95% CI = 95% confidence interval; NPV = negative predictive value; LR = likelihood ratio.

Leiden EAC (n = 625)       
 RF (5.0)154 (24.8)61.7 (54.0, 69.4)77.8 (74.1, 81.6)3.45 (2.60, 4.53)0.61 (0.53, 0.70)47.7 (40.8, 54.7)86.1 (82.8, 89.4)
 RF (15.0)96 (15.4)68.8 (59.5, 78.0)74.9 (71.2, 78.6)4.71 (3.17, 7.01)0.72 (0.65, 0.79)33.3 (26.8, 39.9)92.9 (90.5, 95.4)
 RF (50.0)39 (6.3)71.8 (57.7, 85.9)70.8 (67.2, 74.5)5.45 (2.77, 10.72)0.88 (0.83, 0.94)14.1 (9.3, 19.0)97.4 (95.9, 98.9)
 ACPA149 (23.9)67.1 (59.6, 74.7)78.9 (75.3, 82.6)4.33 (3.21, 5.83)0.57 (0.49, 0.65)50.0 (43.1, 56.9)88.4 (85.4, 91.5)
Berlin EAC (n = 154)       
 RF (24.0)54 (35.3)68.4 (58.1, 78.6)73.3 (63.3, 83.3)2.34 (1.64, 3.33)0.39 (0.26, 0.59)73.0 (62.9, 83.1)68.8 (58.6, 78.9)
 RF (50.0)39 (25.3)72.2 (60.3, 84.2)65.0 (55.7, 74.3)2.81 (1.70, 4.66)0.58 (0.45, 0.76)52.7 (41.3, 64.1)87.3 (72.7, 89.8)
 RF (72.0)34 (22.1)79.1 (66.9, 91.2)64.0 (55.0, 72.9)4.08 (2.10, 7.93)0.61 (0.49, 0.76)45.9 (34.6, 57.3)88.8 (81.8, 95.7)
 ACPA41 (26.6)93.2 (85.7, 100.6)70.0 (61.4, 78.6)14.77 (4.78, 45.68)0.46 (0.36, 0.60)55.4 (44.1, 66.7)96.3 (92.1, 100.4)
NOR-VEAC (n = 193)       
 RF (25.0)11 (5.7)61.1 (38.6, 83.6)93.1 (89.4, 96.9)11.61 (5.01, 26.95)0.54 (0.37, 0.81)47.8 (27.4, 68.2)95.9 (92.9, 98.9)
 RF (50.0)9 (4.7)75.0 (50.5, 99.5)92.3 (88.4, 96.2)22.17 (6.47, 76.01)0.62 (0.45, 0.86)39.1 (19.2, 59.1)98.2 (96.3, 100.2)
 RF (75.0)6 (3.1)85.7 (59.8, 111.6)90.9 (86.7, 95.0)44.35 (5.59, 352.06)0.74 (0.59, 0.95)26.1 (8.1, 44.0)99.4 (98.3, 100.6)
 ACPA14 (7.3)73.7 (53.9, 93.5)94.8 (91.5, 98.1)20.70 (8.22, 52.12)0.40 (0.24, 0.67)60.9 (40.9, 80.8)97.1 (94.5, 99.6)

Second, the results for high RF level were compared to those for ACPA positivity. In all 3 cohorts, the 95% confidence intervals (95% CIs) overlapped. Nevertheless, the balance between PPV (preferably high) and NPV (preferably high) tended to be better for ACPA than for high RF level. In addition, the balance between positive LR (preferably high) and negative LR (preferably low) was better for ACPA presence than for high RF level in all 3 cohorts. These effects were less compelling in the NOR-VEAC than in the Berlin EAC and Leiden EAC. However, the findings in the NOR-VEAC are more difficult to interpret because of large CIs. These larger CIs may be related to the low percentage of UA patients with a high RF level in this cohort of patients with very early disease (3% when high RF level was defined as 3 times the reference value and 5% when high RF level was defined as RF50). When arthritis persistence was used as the outcome measure instead of RA development, comparable findings were obtained. (Data are available from the corresponding author upon request.)

Next, the additive value of performing a second autoantibody test for predicting RA development was investigated. The additive value of performing an ACPA test in UA patients without a high RF level was determined, as well as the additive value of testing RF levels in ACPA-negative UA patients. As shown in Table 3, the PPVs of performing an ACPA test in patients without a high RF level were approximately twice as high as the PPVs of RF level testing in ACPA-negative patients. This analysis was performed using different definitions of high RF level and in the different cohorts. In the Leiden and Berlin EACs, the positive LR for additional ACPA testing in patients without a high RF level ranged from 3.6 to 12.4, and the negative LR ranged from 0.63 to 0.77. RF level testing in ACPA-negative patients resulted in marginal positive LR and negative LR values (∼1) in these cohorts. This contrast was less evident in the NOR-VEAC, but in this cohort the number of ACPA-negative UA patients with a high RF level who developed RA was very low (n = 1). Overall, for the prediction of RA development in patients with early UA, performing an ACPA test in addition to RF level testing seems more valuable than determining the RF level after determining the presence of ACPA antibodies.

Table 3. Additional value of testing for high RF level or ACPA in predicting development of RA in patients with UA, when the test result for the other autoantibody is negative*
Cohort and primary test resultAdditional testPPV, % (95% CI)NPV, % (95% CI)Positive LR (95% CI)Negative LR (95% CI)Sensitivity, % (95% CI)Specificity, % (95% CI)Additional no. (%) of patients with a positive test result
  • *

    Two different definitions of high RF level were used for the primary test: a cutoff value of 3 times the reference cutoff value (RF15) and a cutoff value of 50 units/ml (RF50). The reference cutoff value was used to determine ACPA positivity. See Table 2 for other definitions.

  • The additional number of RA patients with a positive test result was the number of RA patients identified by the second test that was performed. The percentage was calculated out of the total number of patients in each cohort.

Leiden EAC (n = 625)        
 RF15ACPA54.3 (42.6, 66.0)79.6 (75.9, 83.3)3.57 (2.33, 5.47)0.77 (0.69, 0.87)29.0 (21.2, 36.8)91.9 (89.2, 94.6)38 (6.1)
 RF50ACPA63.5 (54.7, 72.3)79.4 (75.8, 83.1)4.25 (3.04, 5.94)0.63 (0.55, 0.72)43.2 (35.7, 50.7)89.8 (86.9, 92.7)73 (11.7)
 ACPA−RF1523.5 (3.4, 43.7)79.6 (75.9, 83.3)1.19 (0.40, 3.57)0.99 (0.95, 1.04)4.1 (0.2, 8.1)96.5 (94.7, 98.4)4 (0.6)
 ACPA−RF5020.0 (−15.1, 55.1)79.4 (75.8, 83.1)0.97 (0.11, 8.55)1.00 (0.98, 1.02)1.0 (−1.0, 3.0)98.9 (79.9, 100.0)1 (0.2)
Berlin EAC (n = 154)        
 RF50ACPA84.6 (65.0, 104.2)72.4 (63.0, 81.8)10.21 (2.40, 43.52)0.71 (0.56, 0.89)31.4 (16.0, 46.8)96.6 (92.7, 101.1)11 (7.1)
 RF72ACPA87.5 (71.3, 103.7)72.6 (63.7, 81.6)12.43 (2.97, 51.92)0.67 (0.53, 0.84)35.0 (20.2, 49.8)97.2 (93.3, 101.0)14 (9.1)
 ACPA−RF5039.1 (19.2, 59.1)72.4 (63.0, 81.8)1.50 (0.72, 3.12)0.89 (0.70, 1.12)27.3 (12.1, 42.5)81.8 (73.2, 90.4)9 (5.8)
 ACPA−RF7246.7 (21.4, 71.9)72.6 (63.7, 81.6)2.04 (0.81, 5.17)0.88 (0.73, 1.07)21.2 (7.3, 35.2)89.6 (82.8, 96.4)7 (4.5)
NOR-VEAC (n = 193)        
 RF50ACPA54.5 (25.1, 84.0)95.3 (92.1, 98.5)14.31 (4.99, 41.07)0.59 (0.37, 0.93)42.9 (16.9, 68.8)97.0 (94.4, 99.6)6 (3.1)
 RF75ACPA64.3 (39.2, 89.4)95.3 (92.2, 98.5)17.89 (6.76, 47.34)0.48 (0.29, 0.80)52.9 (29.2, 76.7)97.0 (94.5, 99.6)9 (4.7)
 ACPA−RF5025.0 (−17.4, 67.4)95.3 (92.1, 98.5)6.11 (0.70, 53.07)0.91 (0.72, 1.14)11.1 (−9.4, 31.6)98.2 (96.1, 100.2)1 (0.5)
 ACPA−RF7550.0 (−19.3, 119.3)95.3 (92.2, 98.5)18.33 (1.25, 269.92)0.89 (0.71, 1.13)11.1 (−9.4, 31.6)99.4 (98.2, 100.6)1 (0.5)

Severity of the disease course in RA patients.

The abilities of high RF level and the presence of ACPA to predict the severity of RA were assessed and compared. The rates of joint destruction among patients with high RF levels (defined as either RF50 or an RF level of 3 times the reference value) and among ACPA-positive RA patients are depicted in Figure 2A. To compare the effect sizes of the 3 groups, the estimates obtained from the repeated-measures analyses performed on log-transformed data were back-transformed to the original scale. This yielded a 1.13, 1.05, and 1.04 times greater progression rate per year for the presence of ACPA, RF level ≥3 times the reference value, and RF50, respectively, compared to the absence of either ACPA or high RF level. Over a total followup period of 7 years, this resulted in 2.41 (95% CI 2.06, 2.83) (P < 0.001), 1.45 (95% CI 1.24, 1.70) (P < 0.001), and 1.29 (95% CI 1.05, 1.59) (P = 0.015) times greater progression rates for ACPA, RF levels ≥3 times the reference value, and RF50.

thumbnail image

Figure 2. Comparison of high RF level and ACPA as predictors of disease severity in RA patients. The association of the outcome measure (radiographic progression or achievement of remission) with positive (versus negative) test results for 2 different definitions of high RF level and for ACPA was determined. A, Sharp/van der Heijde scores for radiographic progression over 7 years of followup in patients with a high RF level defined as ≥50 units/ml (RF50) (n = 123) versus those without a high RF level (n = 526), patients with a high RF level defined as ≥3 times the standard cutoff value of 5 units/ml (RF15) (n = 378) versus those without a high RF level (n = 271), and patients with ACPA (n = 342) versus those without ACPA (n = 289). Values are the mean ± SEM. B, Achievement of disease-modifying antirheumatic drug–free remission over 7 years of followup (FU) in patients with RF50 (n = 122) versus those without RF50 (n = 500), patients with RF15 (n = 370) versus those without RF15 (n = 252), and patients with ACPA (n = 336) versus those without ACPA (n = 270). The cutoff value for the presence of ACPA was 25 arbitrary units. See Figure 1 for other definitions.

Download figure to PowerPoint

To further substantiate the findings with regard to RA severity, the analyses were performed with achievement of sustained DMARD-free remission as the outcome measure (Figure 2B). Presence of ACPA or high RF level was associated with a worse disease outcome, reflected by an increased hazard ratio (HR) for not achieving DMARD-free remission. The observed HRs for not achieving DMARD-free remission were 11.3 (95% CI 5.6, 22.7) (P < 0.001), 5.7 (95% CI 2.9, 11.4) (P < 0.001), and 3.1 (95% CI 1.2, 7.6) (P = 0.016) for ACPA, RF level ≥3 times the reference value, and RF50, respectively. Similar to joint destruction, the effect sizes for high RF level (defined as either RF50 or an RF level of 3 times the reference value) were lower than that for the presence of ACPA antibodies.

Variation in RF measurements.

In order to evaluate whether and to what extent the method of measuring the RF level influences the test outcomes, RF levels in the same serum samples were determined by different methods. The results are shown in Figure 3A. Large variation in absolute levels was observed. In general, nephelometry yielded the highest measurements, followed by turbidimetry. ELISA yielded the lowest measurements. The correlation coefficients between the absolute levels were 0.47 for nephelometry and ELISA (P = 0.007), 0.531 for nephelometry and turbidimetry (P = 0.002), and 0.402 for ELISA and turbidimetry (P = 0.022). Since the 2 RF-positive sera used contained high RF levels, all of the measurements obtained using nephelometry and turbidimetry had an absolute RF level of >50 units. With ELISA, a measurement of <50 units was found once. Figure 3A illustrates the large variation in measurements that is observed when local units are used.

thumbnail image

Figure 3. Comparison of the results obtained using different rheumatoid factor (RF) measurement methods and test facilities. Two RF-positive samples were measured. A, Measurements were obtained using enzyme-linked immunosorbent assay (ELISA) (units/ml), nephelometry (kU/liter), and turbidimetry (IU/liter). The dashed line at 50 units represents the cutoff value of RF50, the definition of a high RF level that is used in the literature. B, The number of units determined by each method of measurement was divided by the corresponding cutoff value. The dashed line at a ratio of 3 represents 3 times the reference cutoff value, the definition of a high RF level that is used in the 2010 American College of Rheumatology/European League Against Rheumatism criteria. C, The number of units determined for each method of measurement was divided by the level obtained for the standard serum (Reference Laboratory for Rheumatologic Serology) in the corresponding test facility. Each symbol represents a single measurement obtained in a separate test facility. Horizontal bars show the median.

Download figure to PowerPoint

Expressing the data as a ratio in relation to the local cutoff value did not improve the variation within and between methods (Figure 3B). The correlation coefficients between these ratios were 0.288 for nephelometry and ELISA (P = 0.11), 0.443 for nephelometry and turbidimetry (P = 0.011), and 0.302 for ELISA and turbidimetry (P = 0.093).

To investigate whether expression of RF level in relation to a standard reference serum would increase the reproducibility of results between laboratories and between methods, the absolute levels of the 2 patient sera were divided by the RF levels obtained for the standard serum (RELARES). Although the variance within the methods decreased, the variability between methods was still considerable (Figure 3C). The correlation coefficients were 0.469 for nephelometry and ELISA (P = 0.008), 0.452 for nephelometry and turbidimetry (P = 0.012), and 0.537 for ELISA and turbidimetry (P = 0.002). As is shown, this effort did not lead to harmonization and reflects the difficulty with using standard sera to homogenize RF level measurements.

DISCUSSION

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

Detailed knowledge of the individual items in the 2010 ACR/EULAR classification criteria for RA is necessary to optimally use these criteria in daily clinical practice. The inclusion of the item “low-positive RF” versus “high-positive RF” seems to hamper uniform application of the 2010 ACR/EULAR criteria.

In the present study, the test characteristics and prognostic ability of high RF levels were compared with those of the presence of ACPA in patients with early UA. The data, originating from 3 cohorts, revealed that the balance between positive LR and negative LR as well as between PPV and NPV was more favorable for ACPA positivity than for high RF level. These findings held both for the diagnosis of RA and for arthritis persistence. The same results were obtained when the severity of the course of RA was studied, which substantiated the findings.

The main outcome measure used in the current study was the development of RA according to the 1987 ACR criteria. An advantage of these criteria is that they could be uniformly applied in the different cohorts in Germany, Norway, and The Netherlands. In light of the new 2010 ACR/EULAR criteria, however, this outcome measure may seem to be an outdated definition of RA. Obviously, the 2010 ACR/EULAR criteria could not be used for the purpose of the present study because of circularity; both the presence of ACPA and RF level are part of these criteria. Using methotrexate (MTX) treatment as the outcome measure, as was done when deriving the 2010 ACR/EULAR criteria for RA, has limitations as well. The Leiden cohort began including UA patients in 1993, and at that time DMARDs were infrequently prescribed in early UA. Hence, there are differences in MTX prescription depending on the inclusion year, which impairs fair comparisons. In addition, MTX is prescribed for other diagnoses, such as psoriatic arthritis. An alternative outcome is expert opinion with regard to the presence of RA. However, expert opinion is likely not independent of the 1987 ACR criteria for RA. Having worked with the 1987 ACR criteria for ∼20 years, clinicians may, consciously or unconsciously, refer to these criteria in their judgments. In the present study, comparable findings were obtained using RA development, arthritis persistence, or RA severity as the outcome measure, suggesting that the findings were not dependent on the use of one particular outcome measure.

Two definitions of high RF level were studied in 3 cohorts. The definitions were RF50 (the definition of high RF level used in previous publications), and 3 times the reference value (the definition of high RF level used in the 2010 ACR/EULAR classification criteria for RA). It was observed that the posttest probabilities (PPV and NPV) varied between the cohorts. For example, the NPV was highest in the NOR-VEAC and lowest in the Berlin EAC. These values are influenced by the different percentages of UA patients who developed RA during the observation period (the pretest probability). Despite this difference, the same differences between the predictive ability of RF level and the predictive ability of ACPA were observed in all 3 cohorts, strengthening the findings. The sensitivities and specificities for high RF levels differed between the cohorts as well. This may be due partly to the different cutoff levels used to define RF positivity. RF50 may be a 2-fold increase compared to the cutoff value in some cohorts (as was the case in the Berlin EAC and the NOR-VEAC), but it may be a 10-fold increase when other methods are applied (as was the case in the Leiden EAC). Although this argument may apply to a lesser extent when the definition of high RF level of 3 times the reference value is used, in this case the stringency with which the reference value is chosen (according to manufacturer instructions or to in-house reference groups) may also affect the test characteristics. The differences in test characteristics of the presence of ACPA were smaller than for RF level.

Another factor that may contribute to differences in measured RF levels and differences in resulting test characteristics are the different techniques that can be used to measure RF. ELISAs were used to measure RF in all cohorts investigated in this study. Generally, there are several variants of each technique, including both in-house and commercially available kits. The manufacturers of these commercially available tests have not provided a 100% standardization of these kits to a reference kit with regard to detection and quantification of RF. Previously, IU/ml have been established, but this method only yields standardized results when the Boehringer nephelometer is used. The prevalent methods also differ with regard to the origin of the antibodies that are directed against RF (human or rabbit) and the isotypes of the antibodies that are tested. Nephelometry usually measures complexes of IgM, IgG, and IgA RFs, whereas ELISAs are specifically directed against one isotype, for instance, IgM-RF.

Appropriate and uniform application of the RF level criterion of the 2010 criteria for RA requires harmonization of all available RF tests. Efforts to harmonize RF determinations have been undertaken by Dutch and European task forces. In The Netherlands, a standard serum consisting of pooled serum from RF-positive patients (RELARES) was developed. However, as shown in the present study (Figure 3C), this did not result in better reproducibility between laboratories. Considerable variability was still observed, not only between various methods for determining RF (such as ELISA, nephelometry, and turbidimetry), but also between different laboratories using the same method. Considering the present difficulties, it is not feasible that worldwide standardization of RF measurement will be achieved in the short term. This study did not address the possibility of standardizing anti-CCP level measurements. In our experience, harmonizing ACPA measurements may be less complicated (data not shown). Therefore, assuming that a modification of the 2010 ACR/EULAR criteria will be undertaken in the future, we propose omitting the RF level and using only ACPA, with different weighted scores for ACPA positivity and ACPA level.

In conclusion, defining a high RF level is complicated due to the variation in RF levels obtained when different methods are applied. This problem hampers uniform application of the 2010 ACR/EULAR criteria for RA. The results of the present study revealed that the overall prognostic ability of ACPA positivity outweighs that of high RF level in patients with UA. For this reason, we suggest that a future modification of the classification criteria for RA should include ACPA determination but not RF level.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

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 published. Dr. van der Linden 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. van der Linden, Batstra, Bakker-Jonges, Burmester, Huizinga, van der Helm-van Mil.

Acquisition of data. van der Linden, Batstra, Bakker-Jonges, Detert, Bastian, Scherer, Burmester, Mjaavatten, Kvien, Huizinga, van der Helm-van Mil.

Analysis and interpretation of data. van der Linden, Batstra, Bakker-Jonges, Toes, Huizinga, van der Helm-van Mil.

Acknowledgements

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES

We are grateful to Dr. A. Roos for discussions on RF measurements.

REFERENCES

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. Acknowledgements
  8. REFERENCES
  • 1
    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.
  • 2
    Funovits J, Aletaha D, Bykerk V, Combe B, Dougados M, Emery P, et al. The 2010 American College of Rheumatology/European League Against Rheumatism classification criteria for rheumatoid arthritis: methodological report Phase I. Ann Rheum Dis 2010; 69: 158995.
  • 3
    Neogi T, Aletaha D, Silman AJ, Naden RL, Felson DT, Aggarwal R, et al. The 2010 American College of Rheumatology/European League Against Rheumatism classification criteria for rheumatoid arthritis: Phase 2 methodological report. Arthritis Rheum 2010; 62: 258291.
  • 4
    Aletaha D, Neogi T, Silman AJ, Funovits J, Felson DT, Bingham CO III, et al. 2010 Rheumatoid arthritis classification criteria: an American College of Rheumatology/European League Against Rheumatism collaborative initiative. Arthritis Rheum 2010; 62: 256981.
  • 5
    Jansen AL, van der Horst-Bruinsma I, van Schaardenburg D, van de Stadt RJ, de Koning MH, Dijkmans BA. Rheumatoid factor and antibodies to cyclic citrullinated peptide differentiate rheumatoid arthritis from undifferentiated polyarthritis in patients with early arthritis. J Rheumatol 2002; 29: 20746.
  • 6
    Nell VP, Machold KP, Stamm TA, Eberl G, Heinzl H, Uffmann M, et al. Autoantibody profiling as early diagnostic and prognostic tool for rheumatoid arthritis. Ann Rheum Dis 2005; 64: 17316.
  • 7
    Van der Linden MP, van der Woude D, Ioan-Facsinay A, Levarht EW, Stoeken-Rijsbergen G, Huizinga TW, et al. Value of anti–modified citrullinated vimentin and third-generation anti–cyclic citrullinated peptide compared with second-generation anti–cyclic citrullinated peptide and rheumatoid factor in predicting disease outcome in undifferentiated arthritis and rheumatoid arthritis. Arthritis Rheum 2009; 60: 223241.
  • 8
    Stone R, Coppock JS, Dawes PT, Bacon PA, Scott DL. Clinical value of ELISA assays for IgM and IgG rheumatoid factors. J Clin Pathol 1987; 40: 10711.
  • 9
    Van der Helm-van Mil AH, le Cessie S, van Dongen H, Breedveld FC, Toes RE, Huizinga TW. A prediction rule for disease outcome in patients with recent-onset undifferentiated arthritis: how to guide individual treatment decisions. Arthritis Rheum 2007; 56: 43340.
  • 10
    Van Aken J, van Bilsen JH, Allaart CF, Huizinga TW, Breedveld FC. The Leiden Early Arthritis Clinic. Clin Exp Rheumatol 2003; 21: S1005.
  • 11
    Thabet MM, Huizinga TW, van der Heijde DM, van der Helm-van Mil AH. The prognostic value of baseline erosions in undifferentiated arthritis. Arthritis Res Ther 2009; 11: R155.
  • 12
    Detert J, Bastian H, Burmester GR. Update of early arthritis and early rheumatoid arthritis. Dtsch Med Wochenschr 2005; 130: 18916. In German.
  • 13
    Mjaavatten MD, Haugen AJ, Helgetveit K, Nygaard H, Sidenvall G, Uhlig T, et al. Pattern of joint involvement and other disease characteristics in 634 patients with arthritis of less than 16 weeks' duration. J Rheumatol 2009; 36: 14016.
  • 14
    Van der Heijde D. How to read radiographs according to the Sharp/van der Heijde method [corrected and republished in J Rheumatol 2000;27:261–3]. J Rheumatol 1999; 26: 7435.
  • 15
    De Rooy DP, van der Linden MP, Knevel R, Huizinga TW, van der Helm-van Mil AH. Predicting arthritis outcomes—what can be learned from the Leiden Early Arthritis Clinic? Rheumatology (Oxford) 2010; 50: 93100.
  • 16
    Van der Woude D, Young A, Jayakumar K, Mertens BJ, Toes RE, van der Heijde D, et al. Prevalence of and predictive factors for sustained disease-modifying antirheumatic drug–free remission in rheumatoid arthritis: results from two large early arthritis cohorts. Arthritis Rheum 2009; 60: 226271.
  • 17
    Otten HG, Daha MR, de Rooij HH, Breedveld FC. Quantitative detection of class-specific rheumatoid factors using mouse monoclonal antibodies and the biotin/streptavidin enhancement system. Br J Rheumatol 1989; 28: 3106.
  • 18
    Stichting Kwaliteitsbewaking Medische Laboratoriumdiagnostiek (Foundation for Quality Medical Laboratory Diagnostics). Nederlands Referentiepreparaat voor de bepaling van reumafactoren en de anitperinucleaire factor. URL: http://www.skml.nl/referentiematerialen/rheuma-factor-en-anti-perinucleaire-factor2.
  • 19
    Klein F, Janssens MB. Standardisation of serological tests for rheumatoid factor measurement. Ann Rheum Dis 1987; 46: 67480.
  • 20
    Van der Linden MP, Feitsma AL, le Cessie S, Kern M, Olsson LM, Raychaudhuri S, et al. Association of a single-nucleotide polymorphism in CD40 with the rate of joint destruction in rheumatoid arthritis. Arthritis Rheum 2009; 60: 22427.