Rheumatology Advance Access published online on April 9, 2008
Rheumatology, doi:10.1093/rheumatology/ken052
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Value of serum cartilage oligomeric matrix protein as a prognostic marker of large-joint damage in rheumatoid arthritis—data from the RAPIT study
1Department of Rheumatology, Leiden University Medical Center, Leiden, 2Department of Neurology, University Medical Centre St. Radboud, The Netherlands, 3Institute of Rheumatology, Prague, Czech Republic, 4Department of Clinical Epidemiology and Biostatistics, Academic Medical Center University of Amsterdam, Amsterdam, 5Department of Radiology, Leiden University Medical Center, Leiden, 6Leyenburg Hospital, The Hague, 7Department of Rheumatology, Vrije Universiteit Medical Center, Amsterdam, 8Department of Rheumatology, Erasmus University Medical Center, Rotterdam and 9Division of Biomedical Research, TNO Prevention and Health, Leiden, The Netherlands.
Correspondence to:
Z. de Jong, Department of Rheumatology C4-R, Leiden University Medical Center, Post-box 9600, 2300 RC Leiden, The Netherlands. E-mail: z.de_jong{at}lumc.nl; zdejong{at}worldonline.nl
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
Objective. To investigate the utility of serum COMP level measurements as a predictor of future damage of the weight-bearing (large) joints in RA patients participating in intensive exercise.
Methods. Data of the 281 completers of a 2-yr randomized controlled trial (Rheumatoid Arthritis Patients In Training; RAPIT) comparing the effects of usual care physical therapy with high-intensity weight-bearing exercises were analysed. The primary outcome variable was defined as the change in radiological joint damage (Larsen score) of the large joints. Potential predictors of outcome were defined: baseline and change in serum level of COMP after 3 months, baseline radiological damage of the large and small joints, number of months on glucocorticoids, change in disease activity and in physical capacity (aerobic fitness and muscle strength) after 2 yrs, and participation in the exercise group.
Results. In cross-sectional evaluation of baseline data, we found strong association between the high serum COMP level and current damage of the large joints. Serum COMP level at baseline, however, was not associated with an increased rate of radiological joint damage after 2 yrs of follow-up. Furthermore, neither interaction between baseline COMP level and participation in exercises, nor change in COMP level after 3 months of exercising were associated with future damage of the large joints.
Conclusion. Neither baseline serum COMP level nor its individual change after 3 months from start of intensive exercise predict longitudinal progression of damage of the large joints in this population.
KEY WORDS: Cartilage oligomeric matrix protein, Large-joint damage, Intensive exercise, Rheumatoid arthritis
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
A recent randomized controlled trial [Rheumatoid Arthritis Patients In Training (RAPIT)] demonstrated that the majority of included RA patients were able to increase their functional ability and physical capacity safely by long-term intensive exercise [1]. In a limited number of exercising patients, however, more increase in damage of the large joints was noticed than in those participating in the usual care (UC) physical therapy. It has been hypothesized that this increase in damage to the large joints might be a consequence of cartilage wearing—a potentially negative effect of high-impact exercises. Since the baseline Larsen score appeared to explain only a small proportion of the variance of the change score, we have investigated other predictors of progressive destruction in relation to exercise. Ideally, a non-invasive test should be designed that discriminates, prior to or shortly after the start of exercising, those patients who might profit from long-term high intensity exercises from those who should be discouraged from participation.
Biomarkers may represent such a tool to identify this patient group [2]. Since the increase in joint damage in exercising patients was attributed to the wearing of the cartilage, a marker of cartilage destruction, COMP, was investigated [3]. Its possible role as a marker of damage progression of the large joints in RA was studied and published in three studies [4–6].
The main objective of the current study was to investigate if baseline COMP and/or individual short-term changes in serum level of COMP could be useful in clinical practice as a predictor of future damage of the large joints of the patients participating in long-term weight-bearing exercises.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
Study participants
The patients under investigation are completers of the RAPIT study [1]. The patients who were recruited for the study were RA patients [7] with a stable disease who complied with the inclusion criteria [1]. All gave written informed consent and were subsequently randomized to treatments. The medical ethics committee of each participating centre approved the study protocol.
The patients randomized to the exercise group participated in an intensive weight-bearing group exercise programme. Patients assigned to the UC group were treated by a physical therapist only if regarded necessary by their attending physician. From the 281 participants who completed the RAPIT study, 145 were participants in the exercise group and 136 were participants in the UC group [1].
Outcome variables
Primary outcome variable was defined as the change in Larsen score of the large joints (radiological damage of the shoulders, elbows, hips, knees, ankles and subtalar joints).
As potential determinants of outcome the following were explored: baseline serum level of COMP, change in serum level of COMP (COMP at 3 months minus baseline COMP), baseline radiological joint damage, use of medication [1], change in disease activity and in physical capacity (aerobic fitness and muscle strength) [1] and participation in the exercise group.
COMP serum levels
Sera were obtained from blood samples collected after the visits at baseline and after 3 months in sterile tubes without additives, and stored at –80°C. COMP was analysed by sandwich ELISA with monoclonal antibodies 16-F12 and 17-C10 [8]. Analyses were performed without knowledge of the clinical data.
Radiological joint damage
The radiographs of the large joints were scored independently by two readers by the Larsen method [9]. Larsen score ranges from 0 to 60 units and is presented as a mean of the scores of the two readers. All radiographs, performed at baseline and 24 months, were scored without information on the sequence in time, patient's identity and treatment received. Radiographs of the small joints (hand and feet) were scored by a single observer using the method devised by Larsen [9] and modified by Scott [10].
Statistical analysis
At baseline, differences between the groups were analysed with the Mann–Whitney U-test or 
2 tests where appropriate. Within-group differences (after 3 months and/or after 2 yrs) were analysed by Wilcoxon's paired signed-rank test and are expressed as a mean (S.D.), where appropriate. The mean differences in change in COMP levels after 3 months between the groups were compared by means of ANOVA. Smallest detectable difference (SDD) was used as a threshold for relevant progression in radiological damage of the large joints [11].
The relationships between the dependent variable (change in radiological damage of the large joints) and the independent variables was explored by multivariate analyses and performed as a multiple linear regression model. The risk for the development of radiological damage is expressed as an unstandardized regression coefficient (B), S.E. and a level of significance (P-value). Multiple logistic regression model analyses were also performed with the dependent variable categorized in two groups (<SDD and 
SDD). The independent variable of interest, COMP, was analysed as a continuous variable as well as a categorical variable (low COMP level defined as <1790 ng/ml; high COMP level as 1790 ng/ml).
Since these analyses yielded no difference in results, only results of the analyses with change in damage and COMP as a continuous variable are presented.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
The baseline characteristics of all 281 completers of the RAPIT study are described elsewhere [12].
The completers of the UC group and the exercise group were similar in most characteristics except for duration of RA [median 5 (IQR 7) yrs in the exercise group and 7 (IQR 11) yrs in the UC group; P < 0.05] and radiological damage of the small joints [median 24.0 (54.5) Larsen score units in the exercise group and 37.5 (IQR 54.5) Larsen score units in the UC group; P < 0.005]. Median age of all completers amounted to 54 (IQR 16) yrs, 255 (80%) were female, 197 (70%) tested RF positive and median ESR amounted to 16 (IQR 22) mm.
During the trial, the mean difference in change of Larsen score (95% CI) of the large joints between the exercise group and UC group showed a trend towards more damage in the exercise group [0.2 (0.0, 0.4) Larsen score units after 1 yr and 0.3 (0.0, 0.7) after 2 yrs] [1]. The SDD of the progression of the Larsen score of the large joints amounted to 1.65 units and was found to be exceeded in 35 (12.5%) completers [15 (10.4%) being participants in the UC and 20 (14.7%) in the exercise group]. After 2 yrs, the mean difference in change of Larsen score (95%CI) of the small joints between the exercise group and UC group amounted to –2.0 (–4.2, –0.2), P = 0.045 [12].
Disease activity score (DAS) decreased gradually, but at all time points and over 2 yrs, no significant difference between the exercise and UC group was found [1]. After 2 yrs, 216 (76.9%) completers reported using a DMARD and 34 (12%) a long-term (>6 months) use of oral glucocorticoids with no significant intergroup differences.
Higher COMP levels are associated with more joint damage
To investigate the association of COMP levels and radiological joint damage cross-sectionally, the baseline serum levels of COMP [median (IQR); 1790 (763) ng/ml] were dichotomized. At baseline, a high COMP level was associated with more damage of the large joints (Table 1). The relationship between the age, gender and BMI and level of COMP described by other authors was confirmed [13].
|
Mean level of COMP after 3 months does not change significantly in either group
Due to missing serum samples the change in COMP level could be measured in only 269 completers of the RAPIT trial.
At baseline, the serum COMP levels were similar in both groups and were increased when compared with healthy populations [8]. After 3 months, the mean (S.D.) serum COMP level increased slightly with 26 (451) ng/ml in the exercise group and decreased in the UC group with 39 (362) ng/ml. These changes from baseline and the mean difference in change between the groups were not statistically significant.
COMP level does not predict the rate of damage of the large joints
The following potential predictors of the rate of damage of the large joints were included in the analyses: baseline and change in COMP level after 3 months, baseline Larsen score of the large and small joints, number of months on glucocorticoids, the area under the curve (AUC) in a plot of DAS against time (AUCDAS), change in aerobic fitness and in muscle strength after 2 yrs and participation in exercise. The interaction term between the participation in exercise and baseline COMP did not significantly predict the outcome [unstandardized regression coefficient (S.E.); 0.138 x 10–3(0.081), P = 0.091] and was not included. Age, a possible confounder of the outcome, also did not significantly predict the outcome [–0.0007 (0.008), P = 0.923] and was not included.
First, the analyses were performed with all completers (Table 2) and confirmed that neither baseline COMP nor change in COMP level after 3 months or the intensive exercise was associated with the change in Larsen score of the large joints after 2 yrs. Multivariate analysis showed that a higher Larsen score of the large joints at baseline and a more frequent use of glucocorticoids were independently associated with the increase in the Larsen score of the large joints. However, despite being statistically significant, the Larsen score at baseline explained only 7% of the change in Larsen score and the combination with baseline COMP level added 0.1% to it.
|
In addition, the analyses were performed with a subgroup of completers (n = 35) with relevant increase in damage of the large joints (>SDD). This subgroup analysis confirmed the lack of association between baseline COMP levels and change in COMP, and change in the damage of the large joints (data not shown).
When all the analyses were performed as a logistic regression model no different results were found (data not shown).
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
This study investigates the value of a cartilage biomarker as a predictor of excessive joint damage by long-term weight-bearing exercises in RA patients.
It demonstrates that neither baseline serum COMP levels nor its individual changes after 3 months from start of intensive exercise are useful as a predictor of the rate of damage of the large joints. Cross-sectionally, a higher serum COMP level was found associated with more radiological joint damage.
These findings seem in contradiction with the results of other investigators [4–6] who reported a positive relationship between baseline serum COMP level and future joint damage in RA patients. There could be several explanations for this disagreement.
The differences in the duration of the observation period and/or the COMP assay used might be of importance. The observation period was shorter in our study in comparison with the cohort described by others [4–6] and consequently possibly less change in damage is found in our study. Indeed, the increase in radiological damage of the large joints in the majority of the participants in our study was relatively small. The population of interest, defined as patients with relevant increase in large-joint damage, amounted to about 13% of the total population, rendering it relatively difficult to find a significant association with serum COMP level.
The differences between the COMP assays could also, theoretically, bias our results. However, Vilim et al. [8] demonstrated that this difference in detection technique should be only of minor importance.
Serum COMP levels might reflect the cartilage pathology of both large and small joints, small joints often more involved in the RA disease process. Although the literature concerning the value of the COMP as a marker of future small joint damage is not unanimous [14–18] we discounted for this possible confounder in our analyses.
Since COMP was also found in synovial tissue its level might partially reflect the inflammation. In this study, however, we only found a very low correlation between the serum COMP levels and inflammation [19].
Finally, since our serum samples were collected at random and without a defined time relationship with respect to exercise we might have missed rapid changes in COMP levels due to loading [20].
No significant changes in COMP level after 3 months were found. Although data from a drug trial [16] demonstrated that 3 months interval may be sufficient to detect a change in serum COMP level induced by drug therapy, it cannot be ruled out that it is too short to detect changes induced by mechanical wearing.
In conclusion, serum COMP level was not found to be a reliable predictor of future large-joint damage in this RA population.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
We gratefully acknowledge the patients for their trust and participation. We thank the assessors, I. Perquin, B. Oud, M. Fluit and M. van Gulijk, for performing the clinical assessments and Prof. A. Cats for reading the radiographs.
Disclosure statement: V.V. has received royalties from Immunotech, Czech Republic and BioVendor, Czech Republic for COMP ELISA. All other authors have declared no conflicts of interest.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionAcknowledgementsReferences |
|---|
- de Jong Z, Munneke M, Zwinderman AH, et al. Is a long-term high-intensity exercise program effective and safe in patients with rheumatoid arthritis? Results of a randomized controlled trial. Arthritis Rheum (2003) 48:2415–24.[CrossRef][Web of Science][Medline]
- Garnero P, Delmas PD. Noninvasive techniques for assessing skeletal changes in inflammatory arthritis: bone biomarkers. Curr Opin Rheumatol (2004) 16:428–34.[CrossRef][Web of Science][Medline]
- Otterness IG, Swindell AC, Zimmerer RO, Poole AR, Ionescu M, Weiner E. An analysis of 14 molecular markers for monitoring osteoarthritis: segregation of the markers into clusters and distinguishing osteoarthritis at baseline. Osteoarthritis Cartilage (2000) 8:180–5.[CrossRef][Web of Science][Medline]
- Forslind K, Eberhardt K, Jonsson A, Saxne T. Increased serum concentrations of cartilage oligomeric matrix protein. A prognostic marker in early rheumatoid arthritis. Br J Rheumatol (1992) 31:593–8.
[Abstract/Free Full Text] - Mansson B, Carey D, Alini M, et al. Cartilage and bone metabolism in rheumatoid arthritis. Differences between rapid and slow progression of disease identified by serum markers of cartilage metabolism. J Clin Invest (1995) 95:1071–7.[Web of Science][Medline]
- Wollheim FA, Eberhardt KB, Johnson U, Saxne T. HLA DRB1* typing and cartilage oligomeric matrix protein (COMP) as predictors of joint destruction in recent-onset rheumatoid arthritis. Br J Rheumatol (1997) 36:847–9.
[Abstract/Free Full Text] - Arnett FC, Edworth SM, Bloch DA, et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum (1988) 31:315–24.[Web of Science][Medline]
- Vilim V, Voburka Z, Vytasek R, et al. Monoclonal antibodies to human cartilage oligomeric matrix protein: epitope mapping and characterization of sandwich ELISA. Clin Chim Acta (2003) 328:59–69.[CrossRef][Web of Science][Medline]
- Larsen A, Dale K, Eek M. Radiographic evaluation of rheumatoid arthritis and related conditions by standard reference films. Acta Radiol Diagn (1977) 18:481–91.[Medline]
- Scott DL, Houssien DA, Laasonen L. Proposed modification to Larsen's scoring methods for hand and wrist radiographs. Br J Rheumatol (1995) 34:56.
[Free Full Text] - Lassere M, Boers M, van der HD, et al. Smallest detectable difference in radiological progression. J Rheumatol (1999) 26:731–9.[Web of Science][Medline]
- de Jong Z, Munneke M, Zwinderman AH, et al. Long-term high-intensity exercise and damage of small joints in rheumatoid arthritis. Ann Rheum Dis (2004) 63:1399–404.
[Abstract/Free Full Text] - Jordan JM, Luta G, Stabler T, et al. Ethnic and sex differences in serum levels of cartilage oligomeric matrix protein. Arthritis Rheum (2003) 48:675–81.[CrossRef][Web of Science][Medline]
- Fex E, Eberhardt K, Saxne T. Tissue-derived macromolecules and markers of inflammation in serum in early rheumatoid arthritis: relationship to development of joint destruction in hands and feet. Br J Rheumatol (1997) 36:1161–5.
[Abstract/Free Full Text] - Roux-Lombard P, Eberhardt K, Saxne T, Dayer JM, Wollheim FA. Cytokines, metalloproteinases, their inhibitors and cartilage oligomeric matrix protein: relationship to radiological progression and inflammation in early rheumatoid arthritis. A prospective 5-year study. Rheumatology (2001) 40:544–51.
[Abstract/Free Full Text] - den Broeder AA, Joosten LA, Saxne T, et al. Long term anti-tumour necrosis alpha monotherapy in rheumatoid arthritis: effect on radiological course and prognostic value of markers of cartilage turnover and endothelial activation. Ann Rheum Dis (2002) 61:311–8.
[Abstract/Free Full Text] - Skoumal M, Kolarz G, Klingler A. Serum levels of cartilage oligomeric matrix protein. A predicting factor and a valuable parameter for disease management in rheumatoid arthritis. Scand J Rheumatol (2003) 32:156–61.[CrossRef][Web of Science][Medline]
- Lindquist E, Eberhardt K, Bendtzen K, Heinegard D, Saxne T. Prognostic laboratory markers of joint damage in rheumatoid arthritis. Ann Rheum Dis (2005) 64:196–201.
[Abstract/Free Full Text] - DiCesare PE, Carlson CS, Stollerman ES, Chen FS, Leslie M, Perris R. Expression of cartilage oligomeric protein by human synovium. FEBS Lett (1997) 412:249–52.[CrossRef][Web of Science][Medline]
- Neidhart M, Muller-Ladner U, Frey W, et al. Increased serum levels of non-collagenous matrix proteins (cartilage oligomeric matrix protein and melanoma inhibitory activity) in marathon runners. Osteoarthritis Cartilage (2000) 8:222–9.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||