Abstract

Objective. Anti-cyclic citrullinated peptide (anti-CCP) antibodies have been identified as highly specific for rheumatoid arthritis (RA). Studies suggest an association with radiographic outcome. The aims of this study were to assess the diagnostic and prognostic utility of the second-generation anti-CCP2 test in a large cohort of early RA patients compared with connective tissue disease (CTD) controls.

Methods. One hundred and eighty-two patients with RA and 121 patients with CTD were recruited. All RA patients had less than 24 months of symptoms and had CRP, rheumatoid factor (RF), HLA typing (SE) and anti-CCP2 antibodies measured at baseline. Function was assessed using the Health Assessment Questionnaire (HAQ) and X-rays performed at 0, 12 and 24 months.

Results. The anti-CCP2 antibody test demonstrated a specificity of 91% and sensitivity of 81% for RA when compared with controls. In RF-negative patients, specificity was 92% and sensitivity 60%. Baseline demographics of the RA cohort showed mean age 57 yr, mean symptom duration 7 months, 63% RF-positive patients, 72% SE-positive, 81% CCP-positive and 21% erosive. The only predictor of change in Larsen score from 0 to 24 months in the cohort was the presence of the shared epitope (P<0.05) and in the RF-negative subgroup it was CCP2 antibody titre >100 (P<0.05). Baseline HAQ was the only significant predictor of HAQ at 24 months, but in the RF-negative subgroup CCP2 antibody titre >100 predicted a poor functional response at 24 months (P<0.05).

Conclusions. This study confirms the diagnostic utility of anti-CCP2 antibodies in early RA, particularly in seronegative patients, in whom anti-CCP2 positivity also conferred prognostic utility for radiographic and functional outcomes.

The emphasis in the management of rheumatoid arthritis (RA) is early diagnosis and intervention. The hypothesis on which this approach is based is that a window of opportunity exists where therapy has a disproportionate impact on outcome [1, 2]. Although not fully explained scientifically, it would appear logical to introduce therapy prior to irreversible damage [3].

An inherent problem with earlier patient assessment is accurate disease classification. Pathognomonic features of arthropathies such as RA, e.g. deformity and nodules, are related to chronicity and are absent at presentation. The American College of Rheumatology (ACR) classification criteria for RA [4] include such parameters, making them insensitive when applied early [5, 6]. Also, routine laboratory investigations such as acute-phase markers and X-rays may be normal in up to 60 and 70% of patients, respectively [7]. Current best practice for early diagnosis of RA is reliant on the history and examination findings, with supplementary additional investigations.

Rheumatoid factor (RF) has been widely used as a screening test for patients with arthritis. Although RF is prognostically useful, as it correlates with functional [8] and radiographic [9] outcomes in both RA and early inflammatory polyarthritis [10], as a diagnostic test it performs poorly, with low sensitivity and moderate specificity [11]. Used in isolation, RF has little diagnostic utility, but has retained its place in practice because of its prognostic ability and the lack of an alternative test.

More recently a highly specific autoantibody system has been described for RA, in which patients develop antibodies to modified (citrullinated) arginine residues, and this has resulted in the development of the anti-cyclic citrullinated peptide (anti-CCP) antibody test, which has a sensitivity of 68% and a specificity >97% [12, 13]. Further development yielded novel peptides and a second-generation test. The anti-CCP2 test has improved sensitivity [∼80%; comparable to IgM-RF (70–75%)] and equivalent specificity (>98%) [14]. Moreover, 35–40% of RF-negative patients are anti-CCP antibody-positive. Anti-CCP antibodies have also demonstrated prognostic utility with regard to radiographic outcomes [15–17]. In this study we measured anti-CCP2 antibodies in a large cohort of early RA patients and connective tissue disease (CTD) controls. The sensitivity and specificity of the anti-CCP2 test were calculated. In the early RA group we assessed the prognostic utility of the test using functional and radiographic outcomes. The study was approved by the Leeds local research ethics committee.

Methods

Patients

Two hundred and ninety-eight patients were included in the study. One hundred and eighty-two were consecutive early synovitis patients with a confirmed diagnosis of RA at follow-up according to ACR classification criteria [4]. All RA patients were naive to disease-modifying anti-rheumatic drugs (DMARDs) and had less than 24 months of symptoms. One hundred and sixteen were consecutive out-patients with CTD (25 scleroderma and 91 SLE patients). All patients provided written informed consent.

All early synovitis patients underwent standard assessment, including laboratory tests, metrology, disease activity, functional assessment and serum storage. Function was measured using the Health Assessment Questionnaire (HAQ) [18]. Patients had RF measurement and HLA-DR typing using oligonucleotide probes and restriction fragment length polymorphism analysis, as previously described [19]. Patients possessing DR1-0101, 0102; DR4 dw4 0401, dw14 0404, 0408, dw15 0405; DR10 1001 were classified as positive for the shared epitope (SE). Anti-CCP2 antibodies were measured using a commercially available enzyme-linked immuno sorbent assay (ELISA) supplied by Axis-Shield, UK.

X-rays of hands and feet were taken at baseline, 12 and 24 months. X-rays were reported to show the presence or absence of erosions by two experienced readers. A subgroup of patients (n = 118) had X-rays scored using the modified Larsen method in a blinded fashion. There were no statistical differences between the baseline demographics of the patients with full Larsen scoring and those without.

Statistics

The sensitivity and specificity and negative and positive predictive values of the anti-CCP2 test were calculated using the RA and control patient cohorts. Forward, stepwise likelihood ratio, binary logistic regression was used to assess predictors of erosive disease at 12 months in the RA cohort. Ordinal regression was used to assess predictors of HAQ at 12 and 24 months using a 0–3 scale based on quartiles for severity of functional outcome and change in HAQ from baseline, and also in a subgroup of 118 patients with full Larsen scores assessing progression (0–12 and 0–24 months).

The RF-negative patients (n = 67) were also analysed separately to assess the prognostic capacity of anti-CCP2 antibodies in this group.

Results

The baseline demographics of the early RA cohort are shown in Table 1. Sensitivities and specificities of the anti-CCP2 antibody ELISA and RF are shown in Table 2.

Table 1.

Baseline demographic data of the early RA cohort

Baseline demographics n = 182 
Mean age 58 yr (s.d. 13.7) 
Mean symptom duration 7 months (s.d. 4.9) 
Female:male ratio 1.8:1 
RF-positive 63% 
SE-positive 72%, 1 allele (25%, 2 alleles) 
Anti-CCP2-positive All RA = 81% (72% titre >100) 
 RF-negative RA = 60% 
Mean CRP 36.6 mg/l 
Erosive 21% 
Median HAQ 1.375 (IQR = 1.375) 
Baseline demographics n = 182 
Mean age 58 yr (s.d. 13.7) 
Mean symptom duration 7 months (s.d. 4.9) 
Female:male ratio 1.8:1 
RF-positive 63% 
SE-positive 72%, 1 allele (25%, 2 alleles) 
Anti-CCP2-positive All RA = 81% (72% titre >100) 
 RF-negative RA = 60% 
Mean CRP 36.6 mg/l 
Erosive 21% 
Median HAQ 1.375 (IQR = 1.375) 
Table 2.

Utility of anti-CCP2 antibody and RF as diagnostic tests for RA in all patients and anti-CCP2 antibodies in RF-negative patients

 All patients
 
 RF-negative patients 
 CCP2 RF CCP2 
Sensitivity 81% 63% 60% 
Specificity 91% 54% 92% 
PPV 94% 68% 89% 
NPV 75% 48% 68% 
PLR 9.4 1.38 7.5 
NLR 0.2 0.68 0.4 
 All patients
 
 RF-negative patients 
 CCP2 RF CCP2 
Sensitivity 81% 63% 60% 
Specificity 91% 54% 92% 
PPV 94% 68% 89% 
NPV 75% 48% 68% 
PLR 9.4 1.38 7.5 
NLR 0.2 0.68 0.4 

PPV, positive predictive value; NPV, negative predictive value; PLR, positive likelihood ratio; NLR, negative likelihood ratio.

Predicting erosive disease

Twenty-one per cent of patients were erosive at baseline and 51% at 1 yr. Using logistic regression only RF and CRP >13.5 mg/l were significant predictors of erosive disease at baseline (P<0.05), the model predicting 64% of erosive patients correctly in a random sample (63% overall). At 12 months, RF, baseline CRP 13.5–43.5 mg/l and SE emerged as significant (P<0.05) for erosive disease, correctly predicting 70% of erosive patients (65% overall).

Predicting radiographic progression

Using ordinal regression to predict change in Larsen scores from 0 to 12 months, only a trend towards significance for CRP (P = 0.06) and SE (P = 0.08) emerged and for 0–24 months only SE was significant (P<0.05).

The RF-negative patients were analysed separately (n = 67). For change in Larsen score from 0 to 24 months, the only significant predictive factor was CCP antibody titre >100 (P<0.05).

Predicting functional impairment

Using ordinal regression, patients with high baseline HAQ were more likely to have high HAQ scores at 12 and 24 months (P<0.001). If baseline HAQ scores are taken out of the model, baseline CRP is the only predictor of a high HAQ (P = 0.001). An analysis looking at improvement in HAQ from baseline demonstrated that patients with high baseline HAQ and RF positivity were less likely to demonstrate functional improvement at 12 and 24 months (P<0.05), with a trend for CCP-positive patients (P = 0.07). In the RF-negative subgroup, CCP antibody titre >100 was significantly associated with absence of functional improvement from baseline at 24 months (P<0.05).

Discussion

The results of this study demonstrate the anti-CCP2 antibody test to be both specific (91%) and sensitive (81%) for early RA patients when compared with CTD controls. This is in agreement with previously published studies [12, 13]. The specificity found in our study is lower than reported elsewhere. The most likely explanation is the use of CTD controls, which is a weakness of this study. A low prevalence of anti-citrullinated peptide antibodies has been reported in SLE, although their role in determining arthritis within this group is uncertain [20]. An ideal control population would have been an age-matched normal control population. An alternative explanation may be the inclusion of patients with short disease duration; however, evidence suggests that anti-CCP antibodies precede RA by many years [21].

The results from the seronegative RA patients were perhaps of greatest interest; in these patients the anti-CCP2 antibody test demonstrated a specificity of 92% and sensitivity of 60% compared with controls. These values suggest important diagnostic utility where previously serology had been unhelpful. The high prevalence of anti-CCP2 positivity (60%) in the RF-negative RA patients is noted in this study. This is higher than previously published, where prevalence is reported to be between 20% [12] and 43% [17]. Explanations for this could be the use of the second-generation assay or, perhaps of greater significance, the relationship to the SE, as the demographic features of the cohorts are similar. Published data would support a correlation of citrullination with the SE [22] and also a cumulative increased susceptibility to development of RA [23].

In our study there was a high prevalence of SE in the cohort, 73% carrying at least one allele. This is higher than has been reported for other cohorts. Unlike van Gaalen et al. [22], there was not a dose effect in our cohort, 89% of patients having either one or two alleles positive for anti-CCP. The high prevalence of SE in the cohort may have resulted in the prognostic associations found.

The functional and radiological prognostic capacity of anti-CCP2 antibodies found in this cohort is of interest. Unlike previous studies by Meyer et al. [15], van Jaarsveld et al. [16] and Kroot et al. [17], we were unable to find an association for all RA patients with radiographic outcome. This may simply be a type 2 error. But when the RF-negative patients were analysed separately, significant associations with both radiographic and functional outcome were found. Such findings are of practical importance for clinical decision-making in early disease. Prognostic assessment at the time of first presentation is crucial, particularly given the new early aggressive therapeutic approaches and the availability of expensive biological agents. In this cohort, using anti-CCP antibodies would appear to select seronegative RA patients with poor prognosis and so may have important implications for patient management. In conjunction with the referenced studies, these findings may have a significant impact on the management of early arthritis patients.

We would like to acknowledge Professor Alan Tennant of the Academic Unit of Musculoskeletal and Rehabilitation Medicine in Leeds for his advice regarding statistical analysis for this paper, and Karen Henshaw and Anne English for their assistance with sample processing.

The authors have declared no conflicts of interest.

References

1
Boers M. Understanding the window of opportunity concept in early rheumatoid arthritis.
Arthritis Rheum
 
2003
;
48
:
1771
–4.
2
Quinn MA, Emery P. Window of opportunity in early rheumatoid arthritis: possibility of altering the disease process with early intervention.
Clin Exp Rheumatol
 
2003
;
21(Suppl. 31)
:
S154
–7.
3
Quinn MA, Conaghan PG, Emery P. The therapeutic approach of early intervention for rheumatoid arthritis: what is the evidence?
Rheumatology
 
2001
;
40
:
1211
–20.
4
Arnett FC, Edworthy SM, Bloch DA et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis.
Arthritis Rheum
 
1988
;
3
:
315
–24.
5
Green MJ M-OH, McGonagle et al. Persistence of mild early inflammatory arthritis: the importance of disease duration, rheumatoid factor and the shared epitope.
Arthritis Rheum
 
1999
;
42
:
2184
–8.
6
Harrison BJ, Symmons DPM, Barrett EM, Silman AJ. The performance of the 1987 ARA classification criteria for rheumatoid arthritis in a population based cohort of patients with early inflammatory arthritis.
J Rheumatol
 
1998
;
25
:
2324
–30.
7
Emery P. The optimal management of early rheumatoid disease: the key to preventing disability.
Br J Rheumatol
 
1994
;
33
:
765
–8.
8
van der Heide A, Jacobs JW, Haanen HC, Bijlsma JW. Is it possible to predict the first year extent of pain and disability for patients with rheumatoid arthritis?
J Rheumatol
 
1995
;
22
:
1466
–70.
9
Möttönnen T, Paimela L, Leirisalo-Repo M, Kautiainen H, Ilonen J, Hannonen P. Only high disease activity and positive rheumatoid factor indicate poor prognosis in patients with rheumatoid arthritis treated with ‘saw-tooth’ strategy.
Ann Rheum Dis
 
1998
;
57
:
533
–9.
10
Harrison B, Thomson W, Symmons D et al. The influence of HLA-DRB1 alleles and rheumatoid factor on disease outcome in an inception cohort of patients with early inflammatory arthritis.
Arthritis Rheum
 
1999
;
42
:
2174
–83.
11
van Zeben D, Hazes JM, Zwinderman AH, Cats A, van der Voort EA, Breedveld FC. Clinical significance of rheumatoid factors in early rheumatoid arthritis: results of a follow up study.
Ann Rheum Dis
 
1992
;
51
:
1029
–35.
12
Schellekens GA, Visser H, de Jong BAW et al. The diagnostic properties of rheumatoid arthritis antibodies recognising a cyclic citrullinated peptide.
Arthritis Rheum
 
2000
;
43
:
155
–63.
13
Bizzaro N, Mazzanti G, Tonutti E, Villalta D, Tozzoli R. Diagnostic accuracy of the anti-citrulline antibody assay for rheumatoid arthritis.
Clin Chem
 
2001
;
47
:
1089
–93.
14
van Venrooij WJ, van Boekel MAM, van den Hoogen FHJ, Drijfhourt JW. De 2de generatie anti-CCP test voor de vroege detectie van reumatoide arthritis.
Ned T Rheum
 
2002
;
2
:
6
–10.
15
Meyer O, Labarre C, Dougados M et al. Anticitrullinated protein/peptide antibody assays in early rheumatoid arthritis for predicting five year radiographic damage.
Ann Rheum Dis
 
2003
;
62
:
120
–6.
16
van Jaarsveld CH, ter Borg EJ, Jacobs JW et al. The prognostic value of the antiperinuclear factor, anti-citrullinated peptide antibodies and rheumatoid factor in early rheumatoid arthritis.
Clin Exp Rheum
 
1999
;
17
:
689
–97.
17
Kroot EJ, de Jong BAW, van Leeuwen MA et al. The prognostic value of anti-cyclic citrullinated peptide antibody in patients with recent-onset rheumatoid arthritis.
Arthritis Rheum
 
2000
;
43
:
1831
–5.
18
Pincus T, Summey JA, Soraci SA Jr et al. Assessment of patient satisfaction in activities of daily living using a modified Stanford Health Assessment Questionnaire.
Arthritis Rheum
 
1983
;
26
:
1346
–53.
19
Gough A, Faint J, Salmon M et al. Genetic typing of patients with inflammatory arthritis at presentation can be used to predict outcome.
Arthritis Rheum
 
1994
;
37
:
1166
–70.
20
Hoffman IEA, Peene I, Cebecauer L et al. Presence of rheumatoid factor and antibodies to citrullinated peptides in systemic lupus erythematosus.
Ann Rheum Dis
 
2005
;
64
:
330
–2.
21
Rantapaa-Dahlqvist S, de Jong BA, Berglin E et al. Antibodies against cyclic citrullinated peptide and IgA rheumatoid factor predict the development of rheumatoid arthritis.
Arthritis Rheum
 
2003
;
48
:
2741
–9.
22
van Gaalen FA, van Aken J, Huizinga TWJ et al. Association between HLA class II genes and autoantibodies to cyclic citrullinated peptides (CCPs) influence the severity of rheumatoid arthritis.
Arthritis Rheum
 
2004
;
50
:
2113
–21.
23
Berglin E, Padyukov L, Sundin U et al. A combination of autoantibodies to cyclic citrullinated peptide (CCP) and HLA DRB-1 locus antigens is strongly associated with future onset of rheumatoid arthritis.
Arthritis Res Ther
 
2004
;
6
:
R303
–8.

Comments

0 Comments