Cartilage Oligomeric Matrix Protein-infl ammation biomarker in knee osteoarthritis

Chondrocytes and synovial cells synthesize Cartilage Oligomeric Matrix Protein (COMP) when activated by proinfl ammatory cytokines. Th e aim of this study was to analyze and compare ultrasound parameters of joint infl ammation, eff usion and synovitis with the levels of COMP in the serum of patients with primary osteoarthritis. Ultrasound was done and the concentration of COMP (ng/mL was examined in  patients.  of patients had eff usion (size .±. mm), . had eff usion in lateral recessus (LR), . (size .±. mm) in suprapatelar (SR), and . (size .±. mm) in medial (MR).  of patients had synovitis size .±. mm in SR, .±. mm in MR; and .±. mm in LR. . of patients had nodular type of synovitis, . had diff usive, and . nodular – diff usive. Th ere was a signifi cant link between the size of synovitis and eff usion in SR (r=., p=.), MR (r=., p=.) and LR (r=., p=.). Th e median of COMP concentration was  (.-) ng/mL in patients without eff usion. In those with eff usion it was  (.-.) ng/mL (p=.). Without synovitis it was  (.-) ng/mL, with synovitis  (-) ng/mL, (p=.), diff usion type synovitis  (-) ng/mL, nodular  (-) ng/mL, nodular-diff usion  (.-.) ng/mL (p=.). With longer osteophytes the median of COMP was  (-) ng/mL, with shorter osteophytes  (.-) ng/mL (p=.). Cartilage oligomeric matrix protein has a moderate signifi cance in the assessment of disturbance of the metabolism of synovial and cartilage tissue in patients with knee osteoarthritis (sensitivity=; specifi city=; cut off =. ng/mL). ©  Association of Basic Medical Sciences of FBIH. All rights reserved


Introduction
Knee osteoarthritis is degenerative disease of the joints with progressive character.It induces intensive pain and restricts knee motion, thus disturbing everyday activities.It is characterized by focal deterioration and abrasion of articular cartilage, sclerosis, cystic formations below the bone surface and formation of osteophytes on the joint surface [].Because of secondary inflammation and prolipheration of synovial membrane, periodical swelling can appear with pain [, ].Chronic synovitis is most often found in the later stage of the disease [].Th e presence of synovitis in the early phase leads to the progression of chondropathy.
Ultrasound technique of joint examination (arthrosonography) is highly advised as a standard in rheumatology, since it is more sensitive in comparison to the clinical examination [] and provides reliable and quick data.Ultrasound is very useful in diagnosing and monitoring of joint effusion and synovitis, especially in early osteoarthritis, with the results that are comparable with MRI [].Biochemical markers (biomarkers) are molecules or fragments of connective tissue matrix released in biological fl uids during the process of tissue metabolism that can be measured using immunoassay method [, ].Th e trend of applying whole-genome analysis techniques has also contributed to a better understanding of physiological and pathological processes involved in homeostasis of bone and cartilage tissues [].Th e key role of infl ammation in osteoarthritis is supported by histological fi nding of synovial joint inflammation and increased biomarker level of synovial tissue metabolism and pro-inflammatory cytokines in chondrocytes [].Today, potentially specific biochemical markers that reflect the quantitative and dynamic degeneration and reparation changes in remodelling of joint tissue in osteoarthritis are being developed [, , , , ].Cartilage Oligomeric Matrix Protein -COMP is a non-collagen protein of articular cartilage matrix [].Its molecule is a pentamer with the molecular mass of  kDa, which contains five identical disulfide related subunits [].It is synthesized by chondrocytes and synovial cells after activation by cytokines.COMP is detected in the tendon, meniscus, ligament and synovial membrane in very small quantities, but it cannot be detected in other organs that are rich in cartilage, such as the lungs and bronchi [].Th is protein enters the composition of collagen type II regulates and stabilizes the collagen network in cartilage tissue [].It is useful as a marker of early cartilage destruction since released during the tearing of collagen network, which results in cartilage deterioration [].Th e growing concentration of COMP reflects disease progression in the early stages [], and it is a very sensitive tool of radiological changes that can be detected only in the later stages of disease [].
T h e l e v e l s o f C O M P i n s e r u m a n d u r i n e a r e i n c r e a s e d i n p a t i e n t s w i t h o s t e o a r t h r itis in comparison to the control group [, ].The objective of the study was to analyze and establish a degree of correlation between the ultrasound parameters of joint infl ammation, defi ned as eff usion and synovitis (hiperthrophy of synovial membrane), with the levels of COMP biomarker in serum in patients with primary osteoarthritis.

Patients
The analysis included  patients diagnosed with primary knee osteoarthritis according to the criteria of ACR (American College of Rheumatology), where the disease was present at least six months prior to the beginning of the study.The patients who had knee injuries six months prior to the research, total or partial endoprothesis or osteotomy of the knee joint, arthroscopy of knee joint in the previous year, or received corticosteroid or chondroprotective substance intraarticularly over the period of four weeks prior to the research were excluded from the research.
Procedure Th e rheumatologist produced ultrasound of both knees in B mode using the apparatus SDU-, linear probe of .- MHz.Using the front longitudinal approach, the signs of presence or absence of synovial inflammation were determined.Eff usion was defi ned as the size of discharge greater than  mm in suprapatelar, lateral and/or medial recessus of the knee.Synovitis was defi ned as the thickening of synovial membrane greater than mm.Maximum depth of discharge and synovial tissue was measured and expressed in mm.Morphologically, discharge and synovitis (nodular, diff usion or nodular-diff usion type) were marked as present or absent.Blood samples were collected by means of vein puncture using the antecubital vein.Th e blood samples were allowed to clot for  min.at room temperature.Th ey were centrifuged at  g for  min at °C and serum samples were stored at -°C until analyzed.Th e concentration of COMP (ng/mL) was determined using Cartilage Oligomeric Matrix Protein (Wieslab TM hCOMP quantitative kit, Eurodiagnostica, Lund, Sweden).The assay utilized native human articular cartilage COMP coated to -well microtiter plates and a rabbit polyclonal antiserum directed to human COMP.It was in standard ELISA inhibition format including an overnight pre-incubation step with a sample and primary antiserum.After the overnight pre-incubation, the solution was transferred to the COMP coated plate.Bound antibody was detected using an alkaline phosphatase labelled as anti rabbit IgG conjugate.Samples were tested in duplicate, according to the recommendations of the manufacturer.

Statistical analysis
During the procedure, the following descriptive statistics were used: arithmetic mean, standard deviation, median, quartiles.For the examination of normal dispersion, Kolmogorov-Smirnov and Shapiro-Wilk were used.The comparison of middle values of two populations was done using the t test and Mann-Whitney test.The correlation between the category variables was examined using the Hi square test.The correlation between the categories of constant variables was examined using the Spearman coeffi cient of correlation.Th e dependence of constant variables and other variables was examined using linear regression.Backward method was also used with regressions.Confidence level in all applied methods was within the limit of ..For the examination of marker quality, ROC curve was used and suitable cut off was determined.Thus, the sensitivity and specifi city of the obtained test was performed.

Results
Arthrosonography was performed in  patients,  (.) males and  (.) females with primary knee OA.The average age of the subjects was .±.(minimum , maximum ), and duration of the disease was .±.(minimum ., maximum ) years.Using arthrosonography, the effusion was determined in  of patients; . had effusion in suprapatelar (SR), . in medial (MR), and . in lateral (LR) recessus of the knee joint.Prolypheration of the synovial membrane (synovitis) was proved in  of patients; . of the subjects had nodular, . had diffusion and . had nodular-diffusion type of synovitis.The average value of effusion in SR was .±.mm, in MR .±.mm, and in LR .±.mm.Mean values of synovitis size were .±.mm in SR, .±.mm in MR and .±.mm in LR.
A significant link was determined between the average values of prolipheral synovial membrane thickness and the size of effusion in suprapatelar (r=.;p=.), medial (r=.;p=.) and lateral (r=.;p=.) recessus (Figure ).A significant link was also determined between the size of synovitis and the size of effusion in MR (r=.,p=.) and LR (r=.,p=.).During ultrasound examination, a signifi cant diff erence was found between the median of COMP biomarker concentration in patients with and without eff usion, regardless of the localization and size (p=.).Median of COMP concentration in  of patients where eff usion could not be seen by arthrosonograpy was  (.-) ng/mL, while  of patients had eff usion of  (.-.)ng/mL (Table ).A signifi cant diff erence was found in the COMP concentration between patients with present or absent prolipheration of synovial membrane (p=.).Median of COMP concentration in . of patients without prolipheration of synovial membrane seen by ultrasound was  (.-) ng/mL, while  of patients with prolipheration of synovial membrane had  (-) ng/mL (measured by ultrasound -see Table ).A significant correlation could not be proved between the median of COMP concentration and the size of effusion in SR (r=.;p=.),MR (r=.;p=.),

).
A significant difference was found in the COMP concentrations between patients with nodular, diffusive or nodular-diffusive synovitis (p=.)(Figure ).Median of the COMP concentration in patients with longer osteophytes was  (-) ng/mL (Table ).Sensitivity of COMP biomarker to the presence of eff usion was , while its specifi city was  (cut off =. ng/mL; Area .; p=.; confi dence interval .-.)(Figure ).

Discussion
Arthrosonography is very useful in diagnosing and monitoring the size and localization of eff usion and synovitis [].
The link between synovial inflammation and progression of structural damages is shown in many studies.One-yearlongitudinal study of Ayral et al., showed that greater volume of aspirated effusion from the painful knee could predict the progression of arthrosis, defined by narrowing of the joint area [].We harmonized arthroscopic measurements in our study with the recommendations of European multcentric studies -EULAR report, part  [] and part  [] that investigated the importance of synovial infl ammation (defi ned as hypertrophy and eff usion) in the genesis of pain worsening and structural progression in patients with osteoarthritis (OA) of the knee joint.In our research,  of patients with knee OA had eff usion.Similar results were also found in the literature, showing the presence of effusion in  - of joints with osteoarthritis [, ].Th e largest number of our patients had eff usion in lateral recessus (LR) -., while . of patients had eff usion in suprapatelar recesus (SR), and . in medial recesus (MR) of the knee.
In the study published in  [], synovitis was detected using arthrosonography in . of patients, thus confi rming it as an informative diagnostic method in discovering synovitis, including the subclinical form.In our research, similar results were also obtained and  of patients had proliferation of synovial membrane (synovitis) out of whom . had nodular type, . diff usive, and . nodular-diff usive type.In our patients, the average value of effusion in SR was .±.mm, in MR .±.mm, and in LR .±.mm.The average value of proliferated synovial membrane was in SR .±.mm, in MR .±.mm and in LR .±.mm.The greatest number of patients had effusion in LR.The size of effusion and synovitis was the greatest in LR recessus in patients with OA.We found that there was a signifi cant correlation between the average values of synovitis size and eff usion in SR (r=.,p=.),MR (r=.,p=.) and LR (r=.,p=.), which proved that higher prolipheration of synovial membrane in individual recesus led to more distinct eff usion in it.It was also found that signifi cant correlation existed between the average value of eff usion in MR and prolipheration of synovial membrane in SR (r=.,p=.), as well as the average value of eff usion in LR and prolipheration of synovial membrane in SR (r=.,p=.).Th is indicated that high prolipheration of synovial membrane in SR led to the presence of signifi cant eff usion in all three recessus of the knee joint and maintained more intensive infl ammation of the knee joint.
Infl ammation of the joint in osteoarthritis is usually gentle and does not lead to disturbance of parameters of the acute phase of inflammation, but can be proved using synovial markers, which are the indicators of synovial activity.Increased values of COMP and hyaluron acid (HA) may suggest that synovial inflammation has a central role in the patogenesis of osteoarthritis.Th e study of Bruyer et al. [] showed that serum levels of HA and COMP correlated with the values of joint damage obtained by magnetic resonance.Physical activity can infl uence the serum level of COMP, but daily values are constant.Recently published study about daily variations of COMP levels in the patients with knee osteoarthritis and with RA pointed out that there were no signifi cant variations in the level of COMP, monitored during the day [].Th is suggests that in clinical practice, further standardization of sample taking time is not necessary.The increasing COMP concentration reveals the progression of the disease especially in the early stage [], and is a lot more sensitive than radiological changes that can be discovered in the late stage of the disease [].Our research showed a signifi cant diff erence between the median of COMP biomarker concentration in patients with and without effusion during ultrasound examination, regardless of the localization and size (p=.).Th e median of COMP concentration in  of patients where eff usion could not be seen by arthrosonograpy was  (.-) ng/ mL, while  of patients had eff usion of  (.-.)ng/mL.This shows that COMP concentrations in serum were higher in patients with knee joint infl ammation, which was indicated by the increased quantity of synovial fluid.It was also found that there was a significant difference in the COMP concentrations between patients with and without prolipheration of synovial membrane (p=.).Th e median of COMP concentration in . of patients without prolipheration of synovial membrane seen by ultrasound was  (.-) ng/mL, while in  of patients with prolipheration of synovial membrane seen by ultrasound it was  (-) ng/mL.This result points to the fact that the values of this biomarker also increase in the serum when there is an inflammation in the knee joint, presented by hyperthrophy of the synovial membrane.The connection was not proved between the average values of COMP concentration with the size of effusion in SR (r=.;p=.), in MR (r=.;p=.), in LR (r=.;p=.) and synovitis in SR (r=.;p=.), in MR (r=.;p=.) and in LR (r=.;p=.).Th e results of our research also indicate a signifi cant diff erence in the mean values of COMP concentration between patients with synovitis of nodular, diff usive or nodular diffusive type (p=.).The median of COMP biomarker concentration in the patients without prolypheration of synovial membrane was  (.-) ng/mL; in those with prolypheration of synovial membrane it was  (-) ng/mL for nodular type,  (-) ng/mL for diffusive type and  (.-.)ng/mL for nodular-diff usive type.Th e highest COMP concentration in serum was found in patients with synovial membrane of diffusive type, indicating the presence of the most intensive infl ammation in the knee joint.Our research shows that there is a signifi cant diff erence in the concentration of COMP biomarkers between patients with shorter and longer osteophytes on kondyles of tibia and femur (p=.).Th e median of COMP concentration in  patients with shorter osteophytes was  (.-) ng/ mL.In  patients with longer osteophytes, it was  (-) ng/mL.Jung et al., [] obtained similar results, showing that patients with longer medial osteophytes and with great capsular distension had higher levels of HA and COMP in serum compared to patients shorter osteophytes.The results published by Garnero et al., [] and Sharif et al., [] confi rm that the levels of COMP, PIIINP and HA serum were higher in patients with osteoarthritis, thus indicating the increased synovial and cartilage metabolism.Th e study of Vilim et al., [] recommended COMP as the marker for OA progression after three-year monitoring of the COMP in serum of the patient with osteoarthritis of the knee.It was concluded that high level of COMP was found in patients with radiographic changes, thus being a prognoses marker of illness progression.The results of our research point to the fact that COMP biomarker can be an indicator of the appearance of eff usion (sensitiv-ity=, specifi city=).Th e cut off was found to be .ng/mL, which means that all patients with osteoarthritis and COMP biomarker concentration below . ng do not have knee inflammation, while the value of COMP biomarker in serum above .ng/mL shows a severe degree of osteoarthritis and the presence of inflammation.

Conclusion
Based on the provided results, it can be concluded that the size of the effusion in the knee is related to the size of synovitis in patients with OA.The patients with effusion, synovitis and longer osteophytes in the knee have higher concentration of COMP biomarker in serum than those without OA inflammatory indicators.Cartilage oligomeric matrix protein has a moderate significance in the assessment of disturbance of the metabolism of synovial and cartilage tissue in patients with knee osteoarthritis.

FIGURE 1 .
FIGURE 1. Link between the average values of prolipheral synovial membrane thickness and the size of eff usion in suprapatelar, medial and lateral recessus

TABLE 1 .
Comparison of the median of the concentration of COMP biomarker between patients with present or absent eff usion and prolipheration of synovial membrane