Article Text

Download PDFPDF

Concise report
Increased cartilage turnover and circulating autoantibodies in different subsets before the clinical onset of rheumatoid arthritis
  1. Carl Turesson1,2,
  2. Ulf Bergström1,2,
  3. Lennart T H Jacobsson1,2,
  4. Lennart Truedsson3,
  5. Göran Berglund4,
  6. Tore Saxne2,5
  1. 1Section of Rheumatology, Department of Clinical Sciences, Lund University, Lund, Sweden
  2. 2Department of Rheumatology, Skåne University Hospital, Lund and Malmö, Sweden
  3. 3Section of Microbiology, Immunology and Glycobiology, Department of Laboratory Medicine, Lund University, Lund, Sweden
  4. 4Department of Clinical Sciences, Lund University, Malmö, Sweden
  5. 5Section of Rheumatology, Department of Clinical Sciences, Lund University, Lund, Sweden
  1. Correspondence to Dr Carl Turesson, Department of Rheumatology, Skåne University Hospital, Södra Förstadsgatan 101, S-205 02 Malmö, Sweden; carl.turesson{at}


Background Previous studies have indicated that autoantibodies may be detected years before the clinical onset of rheumatoid arthritis (RA). Cartilage biomarkers, such as cartilage oligomeric matrix protein (COMP), have not been studied previously in samples collected before the diagnosis of RA.

Methods Between 1991 and 1996, 30 447 subjects were included in the Malmö Diet Cancer Study (MDCS). People who developed RA after inclusion were identified by linking the MDCS database to different Swedish registers. One matched control for each validated case was selected from the MDCS. IgG antibodies against cyclic citrullinated peptide (anti-CCP) and mutated citrullinated vimentin (anti-MCV) and IgM rheumatoid factor (IgM RF) were determined by ELISA. Serum COMP was measured with a sandwich ELISA.

Results 172 incident cases of RA (median time from inclusion to diagnosis 5 years; range 1–13) were identified. Pre-RA cases were significantly more likely than controls to be positive for anti-CCP (21.9% vs 0.6%), anti-MCV (29.6% vs 3.0%) and IgM RF (18.9% vs 2.4%) (all p<0.001). Overall, mean serum COMP levels did not differ between cases and controls. Among pre-RA cases included 1–3 years before diagnosis, raised COMP (>12 U/l) was seen in a greater proportion of anti-CCP-negative than anti-CCP-positive subjects (50% vs 15%; p=0.04).

Conclusions Increased cartilage turnover, measured by COMP, and circulating RA-specific antibodies may be distinct processes in the preclinical phase of RA.

Statistics from


Rheumatoid arthritis (RA) is a heterogeneous disorder with varying long-term prognosis. Antibodies against cyclic citrullinated peptides (anti-CCP) are widely used in the assessment of patients with early arthritis. The presence of anti-CCP is highly specific for RA, can be detected in 60–70% of patients with early disease, and has been associated with increased risk of progressive joint damage.1 Rheumatoid factor (RF) has been found years before the clinical onset of RA,2 and in a survey from Northern Sweden, anti-CCP could also be detected in a subset of patients well before RA diagnosis.3 Similar findings were later reported from other studies of blood donors.4 5 In a recent study, serum proinflammatory cytokine concentrations were raised in particular among anti-CCP-positive pre-RA cases.6

Cartilage oligomeric matrix protein (COMP) is a non-collagenous matrix protein, which is predominantly expressed in cartilage, but also in pressure-loaded parts of tendon and in synovial and dermal fibroblasts.7 Importantly, the bulk of circulating COMP in RA and osteoarthritis most likely originates from cartilage.8 Increased COMP in serum and synovial fluid is a marker of increased cartilage turnover.9 10 In most, but not all, studies of patients with early RA, baseline serum concentrations of COMP have been associated with radiographic progression over time.11,,13 The extent of cartilage involvement before clinical onset of RA has not been studied previously.

The purpose of the present study was to investigate circulating autoantibodies and serum COMP concentrations in samples collected before the onset of RA.

Patients and Methods

This nested case–control study used information from the Malmö Diet Cancer Study (MDCS), which is a prospective cohort study in Malmö, a city in the south of Sweden with approximately 300 000 inhabitants. Between 1991 and 1996, 30 447 subjects (12 121 men and 18 326 women) were included in the MDCS. Details of recruitment and the cohort are described elsewhere.14

We identified people who developed RA after inclusion by linking the MDCS register to a community-based RA register and local and national patient administrative registers, as previously described.15 The cases were validated in a structured review of the medical records.15

Four controls for every case, matched for sex, year of birth and year of screening, who were alive and free from RA when the index person was diagnosed with RA, were selected from the MDCS register for studies of lifestyle factor exposures and RA.15 16 For the present study of biomarkers, one control for each case was selected.

IgG anti-CCP antibodies were measured using microtitre plates coated with citrullinated cyclic peptide obtained from Euro-Diagnostica (Malmö, Sweden). IgG antibodies against mutated citrullinated vimentin (anti-MCV) were determined using a commercial kit (Org 548; Orgentec Diagnostika GmbH, Mainz, Germany). IgM RF was determined using a method calibrated against the World Health Organization reference serum. The reference intervals were <20 U/l, <20 kU/l and <20 kIU/l for anti-CCP, anti-MCV and IgM RF, respectively. Serum COMP concentrations were determined using a sandwich ELISA (AnaMar, Lund, Sweden). As our hypothesis was that COMP might be raised as a marker of increased cartilage turnover in particular among subjects screened only a few years before RA onset, the analysis of COMP was restricted to cases with RA diagnosis 1–5 years later and matched controls, and assessed separately for those with 1–3 years to diagnosis and 3–5 years to diagnosis. Based on previous studies of patients with RA, serum COMP >12 U/l was defined as an increased COMP concentration.17

The proportions of pre-RA cases and matched controls with antibody positivity or raised COMP were compared using McNemar's test and conditional logistic regression analysis. Anti-CCP-positive and anti-CCP-negative pre-RA cases were compared using the χ2 test or Fisher exact test, as appropriate. In an additional exploratory analysis, the relation between serum COMP and subsequent development of RA in anti-CCP-negative pre-RA cases included 1–3 years before RA diagnosis and their matched controls was investigated using bivariate conditional logistic regression, with RA development as the independent variable.

The study was approved by the regional ethics review board in Lund, Sweden.


A diagnosis of RA was made for 172 patients a median of 5 years (range 1–13) after their inclusion in the MDCS. Serum was available from 169 cases and 168 matched controls (table 1). Pre-RA cases were significantly more likely to be positive for anti-CCP, anti-MCV and IgM RF (p<0.001 for each), with similar findings in men and women (figure 1). The median anti-CCP concentration for positive cases was 82 U/l (IQR 36–170). Corresponding values for anti-MCV and IgM RF were 130 kU/l (IQR 57–410) and 75 kIU/l (IQR 32–225).

Figure 1

Proportion with positive circulating autoantibodies in pre-rheumatoid arthritis cases and matched controls; by sex. Anti-CCP, antibody against cyclic citrullinated peptide; anti-MCV, antibody against mutated citrullinated vimentin; IgM RF, IgM rheumatoid factor.

Table 1

Baseline characteristics and serum COMP concentrations for pre-RA cases and matched controls

Overall, there was no difference in serum COMP concentrations between cases and controls, and no difference in proportions of cases and controls with raised COMP concentrations (>12 U/l) (table 1). Patients who developed RA 1–3 years from inclusion (n=33) had slightly higher COMP concentrations than the corresponding controls, although the difference did not reach significance (median 11.7 U/l; IQR 9.5–12.9 vs median 10.2 U/l; IQR 8.3–11.8; p=0.40). Raised COMP concentrations (>12 U/l) were more common among cases than controls in this group (OR 3.07; 95% CI 0.76 to 12.41; p=0.12).

In pre-RA cases included 1–3 years before RA diagnosis, COMP concentrations were higher among anti-CCP-negative patients than among those who were anti-CCP positive (figure 2), and raised COMP was also more common in the anti-CCP-negative subset (10/20 vs 2/13; p=0.04). In a bivariate logistic regression model of anti-CCP negative pre-RA cases included 1–3 years before RA diagnosis and their matched controls, higher COMP concentrations were significantly associated with subsequent development of RA (OR 1.42 per U/l; 95% CI 1.05 to 1.92). In this subset, raised COMP concentrations were also significantly more common among pre-RA cases than controls (p=0.01).

Figure 2

Serum cartilage oligomeric matrix protein (COMP) concentrations (U/l) by anti- cyclic citrullinated peptide (CCP) status among patients with rheumatoid arthritis diagnosis 1–3 years later (n=33). Horizontal bars indicate medians.

Overall sensitivity and specificity for raised COMP as a marker of later RA diagnosis was 23% and 78%, respectively. In the subset included 1–3 years before RA diagnosis, sensitivity was 36% and specificity 81%.

For anti-CCP, sensitivity was 22% overall and 39% for those included 1–3 years before diagnosis. The corresponding specificities were 99% and 100%. For IgM RF, sensitivity and specificity were 21% and 98%, respectively, overall, and 33% and 100%, respectively, among those included 1–3 years before RA diagnosis. For anti-MCV, the corresponding sensitivities and specificities were 30% and 97% and 48% and 97%.


In this nested case–control study, anti-CCP and anti-MCV and IgM RF predicted RA. We also found an association between raised COMP concentrations and future RA among anti-CCP-negative subjects.

This study adds to the growing body of evidence indicating that preclinical biomarkers will contribute to our understanding of very early disease mechanisms in RA. In the study of pre-RA cases from Northern Sweden, the presence of anti-CCP before disease onset was associated with progressive radiographic damage after RA diagnosis.18 This indicates that pathways that lead to destructive arthritis may already be active before the development of clinical symptoms. Increased cartilage turnover may also be a very early marker of events in the cartilage depicted as radiographic damage. The present study suggests that early joint damage may occur in different subsets of pre-RA cases—those with pre-disease anti-CCP and those with raised serum COMP.

At the present state of technology, the sandwich ELISA used to analyse serum concentrations of COMP does not allow distinction between different fragments of COMP. Thus, it can be speculated that the somewhat surprising finding of higher COMP concentrations in anti-CCP-negative patients may also reflect a different pattern of COMP fragments, where a repair process may be depicted by the assay used at these early stages of preclinical disease.

Although our results on COMP should be interpreted with caution because they are derived from a subanalysis of pre-RA cases, they are compatible with recent data from studies on early RA. In a Japanese study, bone erosions observed on MRI were associated with higher COMP concentrations, but serum COMP tended to be lower among anti-CCP-positive than anti-CCP-negative patients.19 Similar patterns were observed in preliminary data from a Swedish early-RA sample20 and the Danish CIMESTRA early-RA clinical trial.21 The relation between COMP and anti-CCP should be further studied in samples obtained before the onset of RA.

Strengths of this study include the community-based approach, the well-defined catchment area, and the comprehensive effort to identify incident RA cases using multiple sources. Limitations are mainly due to the relatively small number of cases.

In conclusion, we have demonstrated that anti-CCP, anti-MCV and IgM RF can be detected up to 10 years before RA diagnosis. Among anti-CCP-negative pre-RA cases included 1–3 years before diagnosis and their matched controls, raised COMP was a predictor of RA. Increased cartilage turnover and circulating RA-specific antibodies may be distinct processes in the preclinical phase of RA.


Skilful technical assistance by Mrs Mette Lindell is appreciated. AnaMar Medical provided the kit for COMP assay. We thank Anders Dahlin for excellent help with data extraction from the MDCS database, Jan-Åke Nilsson for advice on the statistical methods, and Professor Dick Heinegård for helpful comments on the manuscript.



  • Funding This study was funded by the Swedish Research Council, the Swedish Cancer Foundation, Lund University, the Crafoord Foundation and the Swedish Rheumatism Association. This research was also supported by the European Community's FP6 funding (‘Autocure’). This publication reflects only the authors' views; the European Community is not liable for any use that may be made of the information herein.

  • Competing interests TS is cofounder and minor shareholder of AnaMar. There are no other competing interests.

  • Ethics approval This study was conducted with the approval of the regional ethics review board in Lund, Sweden.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.