The need for improved analytical techniques in the study of slow, degenerative diseases such as rheumatoid arthritis and osteoarthritis has driven major efforts aimed at identifying biochemical markers of pathological processes in both diseases. A series of novel biochemical markers has surfaced and their careful validation has become a critical requirement for further use in clinical research.
This report aims at providing a critical review of biochemical markers applied in clinical research of joint diseases, in particular those markers reflecting the turnover of cartilage tissue.
Although pathologically diseases rheumatoid arthritis (RA) and osteoarthritis (OA) are both slow and chronic diseases affecting the joints, and they share the fundamental feature of progressive loss of structural integrity of the joint tissue. The slow progression of the diseases constitutes a significant scientific challenge, not the least from an analytical point of view.
The clinically relevant implications of both diseases are poor joint function and pain, and as no curative treatment is available, it becomes important to identify individuals at risk of developing disease at an early point in time. Such knowledge would allow non-medical, preventive measures to be instituted, for example, weight loss. Lacking strong predictors of pain and loss of joint function, research has focussed on the development of techniques to predict structural damage, in particular loss of cartilage, as this intuitively should be related to clinical symptoms. Loss of articular cartilage over time, however, remains very slow and the annual reduction of cartilage thickness has been reported to be in the range 0.13 ± 0.15 mm . This is similar to the precision error of the technique, and this unfavourable signal-to-noise ratio requests prolonged monitoring periods to recognise any degenerative development in the cartilage compartment of the joint.
We know from associated medical areas, in particular osteoporosis, that biochemical markers are much more dynamic tools than bone mass measurements and the biochemical markers carry significant potential for monitoring disease progression and/or potential effect of medical intervention. Buclin and co-workers observed an almost complete loss of circulating levels of C-telopeptides of type I collagen (CTX-I) a few hours after oral intake of salmon calcitonin, and others have reported similar dose-dependent suppression of bone resorption markers in subjects receiving anti-resorptive therapy. Importantly, the short-term biochemical response to anti-resorptive therapy predicts to some extent the long-term reduction in risk for fractures, and it has been reported from osteoporosis-treatment studies (with alendronate, risedronate and raloxifene) that markers of bone turnover appear even more strongly associated with fracture risk reduction than bone mineral density (BMD) .
Consequently, considerable efforts have been, and continue to be, aimed at the development of sensitive and specific biomarkers reflecting the various stages of joint disease pathology, in general, and cartilage damage, in particular. In this report, we provide a review of recent applications of biochemical markers of cartilage turnover, particularly in research involving patients with RA and OA.
Biomarker classification (BIPED)
Markers of joint diseases can in principle be divided into (a) markers of turnover of the extracellular matrix (ECM), for example, fragments of the constituents of the ECM and the proteolytic enzymes responsible for the breakdown of the ECM (metalloproteinases, cathepsins, aggrecanases, etc.), (b) markers of inflammation and (c) genetic markers.
In the current review, we focus on biochemical markers of cartilage turnover and initially provide examples of their application in clinical research.
Burden of disease, investigative, prognostic, efficacy of intervention and diagnostic (BIPED)
Biochemical markers may have many applications in the study of joint diseases, and recently the Osteoarthritis Biomarkers Network funded by the National Institutes of Health/National Institute of Arthritis, Musculoskeletal, and Skin Disease (NIH/NIAMS) proposed a classification scheme for biomarkers termed BIPED, an acronym for Burden of Disease, Investigative, Prognostic, Efficacy of Intervention and Diagnostic . The objective of the BIPED classification system is to provide specific biomarker definitions for improving development capabilities and analysis of OA biomarkers, and secondly for communicating biomarker progress within a common framework. This scheme is applicable to markers in general, and briefly, the five categories are characterised by the following key features.
Burden of disease (BIPED #1)
Burden-of-disease biochemical markers assess the severity or extent of disease, for example, severity within a single joint and/or number of joints affected. The association of bone and cartilage biochemical markers has been intensively studied in RA in a series of publications around the combinatietherapie bij reumatoïde artritis (COBRA) cohort . Interestingly, it was reported that in particular the urinary excretion of C-telopeptide fragments of type II collagen (CTX-II), a biochemical marker of collagen type II degradation, was associated with various measures of disease activity, for example, number of swollen joints . In patients with OA at multiple sites (knee, hip, hand, vertebral facet joints and spinal disc degeneration), it has been reported that the total radiographic score was significantly associated with the urinary CTX-II concentration . In addition, they found that all the site-specific radiographic OA (ROA) scores, except for spinal disc degeneration, contributed independently to this association. In other studies, CTX-II has been associated with disease severity, that is, Kellgren/Lawrence (K/L) grade, joint space narrowing (JSN) and osteophyte scores , with total radiographic scores in the Genetics Osteoarthritis and Progression (GARP) study , and was reported to be the marker showing the most consistent association with signs and symptoms of disease severity among 10 bone, cartilage and synovium markers . Finally, in a recent comparative study by Kraus and co-workers , the biochemical markers hyaluronic acid (HA), cartilage oligomeric matrix protein (COMP) and CTX-II were quantitative indicators of disease burden in OA as assessed by JSN and osteophyte score.
Investigative (BIPED #2)
This group contains markers with insufficient information to allow inclusion into one of the existing biomarker categories. The investigative category includes markers for which a relationship to various normal and abnormal parameters of cartilage ECM turnover has not yet been established in human subjects. Novel markers of type II collagen and aggrecan, for example, urinary type II collagen neo-epitope (uTIINE) , C2M , aggrecan 374 ARGSV have recently been described, and currently the clinical utility is being investigated.
Prognostic (BIPED #3)
The key features of prognostic markers are the ability to (a) assess the risk for future onset of joint disease among healthy subjects and (b) predict the severity of progression among those who already have the disease.
Reijman and co-workers performed a longitudinal, population-based study in 1235 men and women, and found that elevated urinary excretion of CTX-II was associated with an 8.4-fold increased risk of radiographic progression of OA in the hip . Similar studies have associated other biochemical markers with risk of progression of structural joint damage in OA, for example, COMP and serum hyaluronan . Recently, elevated baseline levels of CTX-II were reported to be significantly associated with loss of knee cartilage volume as determined by magnetic resonance imaging (MRI) .
Studies of the COBRA cohort have demonstrated that baseline measurements of CTX-II were predictive of radiographic progression in patients with early RA over a 4-year follow-up period . In a subsequent report with 11 years of follow-up data, Van Tuyl and co-workers found that baseline CTX-II was still an independent predictor of radiographic progression in these RA patients .
These and other studies suggest that biochemical markers of the ECM potentially could assist in the detection of individuals with elevated risk of accelerated cartilage loss eventually expressed in structural damage as detected by radiographs and/or MRI.
In addition, recent data also suggest that biochemical markers could be associated with the experience of pain in arthritis. In a cross-sectional study of 376 hip OA patients (Evaluation of the CHOndromodulating effect of DIAcerein in osteoarthritis of the Hip (ECHODIAH cohort)), CTX-II and C-reactive protein (CRP) levels were associated with pain by visual analogue scales (VASs) . After adjustment for age, gender and body mass index (BMI), these associations remained significant ( p = 0.0095 and p = 0.046 for CTX-II and CRP, respectively).
Efficacy of intervention (BIPED #4)
An efficacy-of-intervention biomarker provides information about the efficacy of treatment among persons with disease or those at high risk for developing the disease (preventive medication). In the COBRA study referred to above, the association of early biochemical response to medical intervention was investigated in 155 subjects who met the American College of Rheumatology (ACR) criteria for RA. The intervention regimens involved a step-down combination procedure with prednisolone, methotrexate and sulfasalazine. Landewé and co-workers found that the magnitude of decrease in CTX-II induced by the COBRA therapy over 3 months could predict the long-term (5 year) radiographic progression. Those patients, who had normalised CTX-II levels following 3 months of therapy, experienced a 4.5 times higher chance of radiographic stability of their disease compared to those without biochemical response.
As no medical intervention is available, which will arrest the pathological processes underlying the destruction of joint tissue in OA, biomarkers per definition cannot yet be assigned to this BIPED category. Salmon calcitonin is currently being investigated in phase III clinical studies for treatment of OA, and has been reported to have beneficial effects on the OA knee joint by improving Lequesne’s algofunctional index score over 84 days of treatment. Both in this, and in a similar trial of shorter duration , the CTX-II levels dropped significantly in the treated patients compared to those receiving placebo.
Diagnostic (BIPED #5)
Diagnostic markers are defined by the ability to classify individuals as either having or not having a disease. Alternatively, such a marker could assign a certain quantity of risk for having the disease. Subsequent more comprehensive physical and laboratory examinations must confirm or reject the diagnosis. Usually clinical arthritis is localised to a single or a few joints, and in each affected joint the damage can be condensed to small, focal lesions. Therefore the sensitivity of the markers to identify individuals with arthritis has been questioned. Taking the diagnostic requirements for the RA and OA diagnosis into account it would seem unreasonable to expect that a single biochemical marker could capture this complexity, that is, could provide sensitivities and specificities approaching 100%. Indeed, the positive and negative predictive values for the cartilage markers to identify patients with a classic arthritic diagnosis have been moderate, and they might find their best clinical utility in other applications within the BIPED system.
Biochemical markers of cartilage
The ECM builds the structural integrity of the joint and serves to protect the chondrocytes from being destroyed by the mechanical load. In addition, the articular cartilage allows smooth, pain-free movement of the skeletal surfaces. Any changes in the molecular components of the cartilage matrix could affect the pathophysiology of the joint, and therefore the study of cartilage biomarkers has attracted much attention. Additional reviews have been published in this area .
Cartilage is a non-vascularised tissue consisting of chondrocytes and ECM. The ECM is a composite network of proteins, primarily collagen type II, and also type IX and XI are present in minor amounts , interacting with polysaccharides and proteoglycans. Aggrecan is the predominant proteoglycan in cartilage. Less abundant proteoglycans include decorin, fibromodulin, perlican and biglycan. Other important molecules in articular cartilage include COMP , link protein and hyaluronan (or HA).
A central hallmark in OA pathogenicity is a gradual destruction of articular cartilage and local denudation leading to loss of joint function . The integrity of articular cartilage is normally maintained by a balance between catabolic and anabolic processes; however, in the case of pathological matrix destruction, the rate of cartilage degradation exceeds the rate of formation, resulting in a net loss of cartilage matrix .
The proteolytic enzymes presently receiving the most attention are the matrix metalloproteinases (MMPs) and aggrecanases (ADAMTS (A Disintegrin And Metalloproteinase with Thrombospondin Motifs)), and some have been evaluated as disease markers. Circulating levels of MMP-3 have been reported to be predictors of radiographic progression in RA patients receiving non-biological disease-modifying OA drugs (DMOADs) , and greater reductions in MMP-3 levels were found in RA patients responding clinically to anti-rheumatic treatment (the anti-interleukin-6 receptor inhibitor tocilizumab plus methotrexate) as compared to the non-responders . In ankylosing spondylitis, circulating levels of MMP-3 were associated with structural progression over 2 years . This association remained significant after adjustment for gender, age, disease duration, CRP and baseline radiographic score.
In particular, the proteolytic fragments generated by these proteases upon their action on the ECM components are attractive molecular target candidates for biomarker discovery, and several of these are referenced in Table 1 , including CTX-II, uTIINE, C2C and aggrecanase fragments of aggrecan.
| Biochemical marker | Key findings in clinical studies |
|---|---|
| CTX-II | CTX-II was increased ( p < 0.001) and correlated positively with disease activity and radiographic progression over 4 years in RA patients . |
| In 84 early RA patients, U-CTX-II level was a predictor of radiographic progression in hand OA over 12 months, but did not predict joint destruction assessed by MRI . | |
| Baseline CTX-II was an independent predictor of annual radiographic progression over 11 years in RA patients. RANK/OPG ratio and CTX-I in the same model improved the prediction (36–39% of the variance explained by model) . | |
| Calcitonin significantly and acutely (0–4 h) suppressed excretion of CTX-II in OA patients . | |
| A combination of U-CTX-II and MR-based imaging markers was superior for prediction of radiographic progression of knee OA in 159 patients over 21 months (AUC = 0.77, Odds Ration 5.8) . From the same cohort it was reported that while elevated baseline levels of CTX-II were only non-significantly associated with radiographic progression by K/L and JSN in OA patients, CTX-II predicted longitudinal cartilage loss by MRI (middle/high tertiles had odds ratios 4.0/3.9, p < 0.01) corresponding to 3.1% increased yearly cartilage loss . | |
| Baseline CTX-II was significantly associated ( r = 0.28, p < 0.05) with JSN Sharp score over 1 year in patients with RA receiving infliximab and methotrexate . The level of CTX-II was not modulated by therapy. | |
| CTX-II was significantly elevated in both early untreated RA ( n = 45) as well as in RA patients with more advanced disease ( n = 50), however, the marker was not associated with radiographic progression over 12 months . | |
| A high baseline, U-CTX-II levels were predictive for the radiographically evident joint damage (modified Sharp method) in DMARD-treated RA patients . | |
| Knee OA patients with elevated CTX-II at baseline (>150 ng/mmol creatinine) and with a biochemical response (CTX-II return to low levels) after 6 months of risedronate therapy, had a lower risk of structural progression over 24 months compared to those subjects who had CTX-II levels remaining high . However, risedronate failed to improve symptoms or slow radiographic progression even though CTX-II was dose-dependently decreased . | |
| CTX-II significantly and independently contributed to explaining variation in radiological damage and progression in 83 patients with ankylosing spondylitis followed for 2 years . | |
| In early RA, baseline levels of CTX-II were associated with radiographic progression (change in Larsen score) ( r = 0.29, p = 0.0033) over 2 years . | |
| CTX-II levels did not respond to infliximab therapy in patients with severe RA . | |
| Knee OA patients with baseline CTX-II levels above the median, had a 3.4-fold ( p = 0.02) increased risk for radiographic progression (JSN ≥ 2 mm) over 5 years . | |
| In patients with symptomatic knee OA, chondroitin sulphate did not induce changes in CTX-II. However, there was a significant difference between OARSI non-responders and responders in 24-week changes of CTX-II ( p = 0.014) . | |
| Strontium ranelate induced a 15–20% reduction in CTX-II in postmenopausal women . | |
| Urinary CTX-II was higher in younger subjects (mean age 45 years) with knee pain ( p = 0.04) compared to individuals without knee pain, but this became nonsignificant after adjustment for BMI and osteophytes . | |
| An increase in CTX-II after three months was correlated with 1 year changes in MRI indexes of joint damage in patients with knee OA . | |
| In knee OA patients, oral salmon calcitonin decreased CTX-II and improved joint function (Lequesne’s algofunctional index score) over 84 days . | |
| Adjusted for age and BMI, U-CTX-II was associated with an 8.7-fold increase in risk (Odds Ratio) for rapidly destructive hip OA . | |
| CTX-II was significantly associated with K/L-score and JSN of the knee in OA patients . | |
| CTX-II was longitudinally associated with indexes of joint damage, e.g. DAS28 ( p < 0.001), over 12 months in patients with RA . | |
| Radiographic scores of total, hip, knee, hand, facet, but not spinal disc degeneration, were independently contributing to increasing the urinary CTX-II level in OA patients . | |
| In a study of 62 knee OA subjects, no association was found between urinary CTX-II and changes in cartilage volume or thickness over 1 year . | |
| CTX-II was associated with knee cartilage defect severity in OA subjects . | |
| Oral salmon calcitonin over three months induced a dose-dependant suppression of CTX-II in healthy postmenopausal women . | |
| Increased 25–177% in OA compared to healthy controls . | |
| CTX-II levels were associated with pain (VAS score) and radiographic signs of joint damage (JSW) in patients with hip OA (ECHODIAH cohort) . | |
| Bone marrow abnormality scores by MRI correlated significantly with CTX-II levels ( p < 0.0001) at baseline. Baseline CTX-II levels in the highest tertile had a relative risk of 2.4 (95% confidence interval 1.1–5.0) of worsening bone marrow abnormalities at 3 months compared with patients with levels in the lowest tertile . | |
| Nimesulide, but not ibuprofen, markedly reduced the urinary levels of CTX-II ( p < 0.001) in individuals with flare-up of OA . | |
| Elevated CTX-II levels (highest tertile) in patients with RA were predictive of radiographic progression over 4 years (COBRA cohort) . | |
| uTIINE | Elevated uTIINE was observed in symptomatic OA patients ( n = 38) compared to healthy controls ( p = 0.0012) . |
| Baseline uTIINE was not a consistent predictor of JSN in OA patients, but serial measurements reflect concurrent JSN . | |
| C2C | In a cohort of 135 RA patients with radiographs at baseline and during follow-up for 5–10 years, no association was found between radiographic progression of hand OA and baseline serum levels of C2C . |
| In a cohort of 87 patients with RA, elevated levels of serum C2C were detected in subjects with rapid radiographic progression compared to slow progressors . And C2C levels were predictive of subsequent radiographic progression rate, in particular JSN. | |
| C2C was not associated with progression of JSN in knee OA patients . | |
| Glucosamine sulphate for 6 months failed to induce significant changes in C2C compared to placebo in knee OA patients . | |
| In patients with ankylosing spondylitis treatment with etanercept induced a significant reduction in levels of C2C ( p = 0.005) compared to placebo controls . | |
| Elevated urinary and serum levels were detected in patients with OA . | |
| In knee OA patients, C2C was correlated with cartilage volume (MRI) and with radiographic assessment . | |
| Infliximab and methroxate reduced C2C in RA, and C2C levels reflected symptoms . | |
| PIIANP | In rheumatoid arthritis patients, with an inadequate response to methotrexate, co-administration with tocilizumab induced a marked dose-dependent reductions in PIIANP levels . |
| In RA, circulating PIIANP did not differ from healthy subjects at study entry or any time during follow-up. In patients with longstanding RA, baseline PIIANP was lower among X-ray progressors than in non-progressors ( p = 0.04) . | |
| PIIANP was significantly lower in hand OA compared with controls and non-hand OA . | |
| Knee OA patients with baseline serum PIIANP levels in the highest quartile, had a 3.2-fold ( p = 0.03) increased risk for radiographic progression (JSN ≥ 2 mm) over 5 years . | |
| Serum PIIANP was decreased by 53% ( p < 0.0001) and 35% ( p < 0.001) in patients with knee osteoarthritis and rheumatoid arthritis, respectively . | |
| PIIANP levels were markedly decreased in patients with knee OA (12.0 ± 3.2 vs 25.8 ± 7.5 ng/ml for OA and controls respectively, p < 0.0001) and RA (14.1 ± 2.5 ng/ml vs 21.7 ± 7.6 ng/ml for RA and controls respectively, p < 0.0001) . | |
| Aggrecan 846 | Baseline levels of the 846 epitope were not related to cartilage loss (MRI) over 1 year , or progression of JSN over 16 months . |
| In RA, serum levels of the 846 epitope 1 year after disease onset were associated with subsequent radiographic progression rate . | |
| In patients with ankylosing spondylitis treatment with etanercept induced a significant increase in the 846 epitope ( p = 0.01) compared to placebo controls . | |
| Patients with ankylosing spondylitis demonstrated significant elevations in serum levels of the Aggrecan 846 epitope . | |
| Aggrecan 846 was unrelated to clinical end points (status or change over 1 year) . | |
| Aggrecan 846 was suppressed in OA subjects compared to controls, i.e. 0.05 vs 0.087 μg/ml, respectively . | |
| Aggrecan fragments | Increased concentrations of aggrecan fragments 374 ARGSV were detected in plasma and synovial fluid from patients with moderate osteoarthritis . |
| Elevated concentrations of the aggrecan 374 ARGS fragments determined by LS-MS/MS have been reported in human urine from OA patients (0.56 ± 0.15 ng/mg creatinine) compared to healthy controls (0.36 ± 0.15 ng/mg creatinine) . In synovial fluid the corresponding levels were 19.3 ± 16.5 ng/ml and 5.5 ± 5.0 ng/ml, respectively. | |
| By immunoassay, the concentration of aggrecan fragments carrying the 374 ARGSV epitope in synovial fluid from patients with acute inflammatory arthritis and acute knee injury (median 88.5 and 53.9 pmol ARGSV/ml, respectively) were elevated compared to subjects with chronic knee injury, knee osteoarthritis and healthy controls (median 0.5, 4.6, and 0.5 pmol ARGSV/ml, respectively) . | |
| Aggrecan fragments carrying the G1 and/or the G2 domain was suppressed 40% in patients with RA compared to healthy controls . | |
| Glc–Gal–PYD | Baseline levels were significantly associated with Sharp score of both erosion ( r = 0.34, p < 0.01) and JSN ( r = 0.36, p < 0.01) . The level of Glc–Gal–PYD was not modulated by therapy over 1 year. |
| Glc–Gal–PYD was significantly associated with K/L-score and JSN of the knee . | |
| Elevated 18% in knee OA and associated with total WOMAC index . | |
| COMP | The decrease in serum COMP after 6 months of etanercept treatment was significant in a small group ( n = 10) of RA patients with remission, but not in the non-remission group ( n = 35) . |
| In a cohort of 135 RA patients with radiographs at baseline and during follow-up for 5–10 years, no association was found between radiographic progression of hand OA and baseline serum levels of COMP . | |
| In 98 RA patients, serum COMP concentration was not different among groups identified with low, moderate, and high disease activity as determined by DAS28-CRP values . | |
| Serum COMP was significantly higher in hand OA compared with controls and non-hand OA . | |
| Change in COMP from baseline to 6 years was associated with development of incident radiographic hip OA, and stabilisation of disease . | |
| In an investigation of 281 RA patients, elevated COMP was associated with more damage in the large joints as determined by Larsen score . However, neither baseline COMP nor 3 months change was associated with change in Larsen score in the large joints over 2 years. | |
| In early RA, baseline levels of serum COMP were associated with radiographic progression (change in Larsen score) ( r = 0.27, p = 0.0038) over 2 years . | |
| Baseline COMP predicted cartilage loss (WORMS) determined by MRI over 30 months . For each SD increase in COMP, the loss of cartilage volume increased 6.1 times, which essentially remained unchanged after adjustment for age, gender and BMI. | |
| Baseline levels were not correlated with 1 year change in cartilage volume or thickness , or 3 years change in JSW or WOMAC in OA patients. | |
| Baseline (and area under the curve per month) of COMP concentrations was significantly higher in patients with tibiofemoral compared to patellofemoral osteoarthritis ( p < 0.01) . | |
| Significant correlation between baseline COMP and JSN in knee OA over 3 years , however no association was detected to Lequesne and WOMAC. | |
| Serum COMP was reported to be unrelated to clinical end points (status or change over 1 year) . | |
| Serum COMP is elevated 16% in knee OA . However, suppressed levels in OA have been reported . | |
| HA | Elevated levels associated with increased risk for progression of radiographic hip OA (ECHODIAH) . |
| Nimesulide but not ibuprofen markedly reduced the serum levels of HA in individuals with flare-up of OA ( p < 0.05) . | |
| HA is unrelated to clinical end points in OA patients (status or change over 1 year) . | |
| Serum levels in OA patients indifferent from controls . |
Finally, in RA, the presence of anti-cyclic citrullinated peptide (CCP) antibodies was associated with a 5.7-fold increase (95% confidence interval (CI) range; 2.6–12.5) in risk of radiographic progression over 10 years of follow-up . Presence of anti-mutated citrullinated vimentin (MCV) antibodies was associated with a 7.3-fold increase (95% CI range; 3.2–16.5) in risk for radiographic progression, and both associations remained significant after adjustment for other predictors.
Biochemical markers of cartilage
The ECM builds the structural integrity of the joint and serves to protect the chondrocytes from being destroyed by the mechanical load. In addition, the articular cartilage allows smooth, pain-free movement of the skeletal surfaces. Any changes in the molecular components of the cartilage matrix could affect the pathophysiology of the joint, and therefore the study of cartilage biomarkers has attracted much attention. Additional reviews have been published in this area .
Cartilage is a non-vascularised tissue consisting of chondrocytes and ECM. The ECM is a composite network of proteins, primarily collagen type II, and also type IX and XI are present in minor amounts , interacting with polysaccharides and proteoglycans. Aggrecan is the predominant proteoglycan in cartilage. Less abundant proteoglycans include decorin, fibromodulin, perlican and biglycan. Other important molecules in articular cartilage include COMP , link protein and hyaluronan (or HA).
A central hallmark in OA pathogenicity is a gradual destruction of articular cartilage and local denudation leading to loss of joint function . The integrity of articular cartilage is normally maintained by a balance between catabolic and anabolic processes; however, in the case of pathological matrix destruction, the rate of cartilage degradation exceeds the rate of formation, resulting in a net loss of cartilage matrix .
The proteolytic enzymes presently receiving the most attention are the matrix metalloproteinases (MMPs) and aggrecanases (ADAMTS (A Disintegrin And Metalloproteinase with Thrombospondin Motifs)), and some have been evaluated as disease markers. Circulating levels of MMP-3 have been reported to be predictors of radiographic progression in RA patients receiving non-biological disease-modifying OA drugs (DMOADs) , and greater reductions in MMP-3 levels were found in RA patients responding clinically to anti-rheumatic treatment (the anti-interleukin-6 receptor inhibitor tocilizumab plus methotrexate) as compared to the non-responders . In ankylosing spondylitis, circulating levels of MMP-3 were associated with structural progression over 2 years . This association remained significant after adjustment for gender, age, disease duration, CRP and baseline radiographic score.
In particular, the proteolytic fragments generated by these proteases upon their action on the ECM components are attractive molecular target candidates for biomarker discovery, and several of these are referenced in Table 1 , including CTX-II, uTIINE, C2C and aggrecanase fragments of aggrecan.
| Biochemical marker | Key findings in clinical studies |
|---|---|
| CTX-II | CTX-II was increased ( p < 0.001) and correlated positively with disease activity and radiographic progression over 4 years in RA patients . |
| In 84 early RA patients, U-CTX-II level was a predictor of radiographic progression in hand OA over 12 months, but did not predict joint destruction assessed by MRI . | |
| Baseline CTX-II was an independent predictor of annual radiographic progression over 11 years in RA patients. RANK/OPG ratio and CTX-I in the same model improved the prediction (36–39% of the variance explained by model) . | |
| Calcitonin significantly and acutely (0–4 h) suppressed excretion of CTX-II in OA patients . | |
| A combination of U-CTX-II and MR-based imaging markers was superior for prediction of radiographic progression of knee OA in 159 patients over 21 months (AUC = 0.77, Odds Ration 5.8) . From the same cohort it was reported that while elevated baseline levels of CTX-II were only non-significantly associated with radiographic progression by K/L and JSN in OA patients, CTX-II predicted longitudinal cartilage loss by MRI (middle/high tertiles had odds ratios 4.0/3.9, p < 0.01) corresponding to 3.1% increased yearly cartilage loss . | |
| Baseline CTX-II was significantly associated ( r = 0.28, p < 0.05) with JSN Sharp score over 1 year in patients with RA receiving infliximab and methotrexate . The level of CTX-II was not modulated by therapy. | |
| CTX-II was significantly elevated in both early untreated RA ( n = 45) as well as in RA patients with more advanced disease ( n = 50), however, the marker was not associated with radiographic progression over 12 months . | |
| A high baseline, U-CTX-II levels were predictive for the radiographically evident joint damage (modified Sharp method) in DMARD-treated RA patients . | |
| Knee OA patients with elevated CTX-II at baseline (>150 ng/mmol creatinine) and with a biochemical response (CTX-II return to low levels) after 6 months of risedronate therapy, had a lower risk of structural progression over 24 months compared to those subjects who had CTX-II levels remaining high . However, risedronate failed to improve symptoms or slow radiographic progression even though CTX-II was dose-dependently decreased . | |
| CTX-II significantly and independently contributed to explaining variation in radiological damage and progression in 83 patients with ankylosing spondylitis followed for 2 years . | |
| In early RA, baseline levels of CTX-II were associated with radiographic progression (change in Larsen score) ( r = 0.29, p = 0.0033) over 2 years . | |
| CTX-II levels did not respond to infliximab therapy in patients with severe RA . | |
| Knee OA patients with baseline CTX-II levels above the median, had a 3.4-fold ( p = 0.02) increased risk for radiographic progression (JSN ≥ 2 mm) over 5 years . | |
| In patients with symptomatic knee OA, chondroitin sulphate did not induce changes in CTX-II. However, there was a significant difference between OARSI non-responders and responders in 24-week changes of CTX-II ( p = 0.014) . | |
| Strontium ranelate induced a 15–20% reduction in CTX-II in postmenopausal women . | |
| Urinary CTX-II was higher in younger subjects (mean age 45 years) with knee pain ( p = 0.04) compared to individuals without knee pain, but this became nonsignificant after adjustment for BMI and osteophytes . | |
| An increase in CTX-II after three months was correlated with 1 year changes in MRI indexes of joint damage in patients with knee OA . | |
| In knee OA patients, oral salmon calcitonin decreased CTX-II and improved joint function (Lequesne’s algofunctional index score) over 84 days . | |
| Adjusted for age and BMI, U-CTX-II was associated with an 8.7-fold increase in risk (Odds Ratio) for rapidly destructive hip OA . | |
| CTX-II was significantly associated with K/L-score and JSN of the knee in OA patients . | |
| CTX-II was longitudinally associated with indexes of joint damage, e.g. DAS28 ( p < 0.001), over 12 months in patients with RA . | |
| Radiographic scores of total, hip, knee, hand, facet, but not spinal disc degeneration, were independently contributing to increasing the urinary CTX-II level in OA patients . | |
| In a study of 62 knee OA subjects, no association was found between urinary CTX-II and changes in cartilage volume or thickness over 1 year . | |
| CTX-II was associated with knee cartilage defect severity in OA subjects . | |
| Oral salmon calcitonin over three months induced a dose-dependant suppression of CTX-II in healthy postmenopausal women . | |
| Increased 25–177% in OA compared to healthy controls . | |
| CTX-II levels were associated with pain (VAS score) and radiographic signs of joint damage (JSW) in patients with hip OA (ECHODIAH cohort) . | |
| Bone marrow abnormality scores by MRI correlated significantly with CTX-II levels ( p < 0.0001) at baseline. Baseline CTX-II levels in the highest tertile had a relative risk of 2.4 (95% confidence interval 1.1–5.0) of worsening bone marrow abnormalities at 3 months compared with patients with levels in the lowest tertile . | |
| Nimesulide, but not ibuprofen, markedly reduced the urinary levels of CTX-II ( p < 0.001) in individuals with flare-up of OA . | |
| Elevated CTX-II levels (highest tertile) in patients with RA were predictive of radiographic progression over 4 years (COBRA cohort) . | |
| uTIINE | Elevated uTIINE was observed in symptomatic OA patients ( n = 38) compared to healthy controls ( p = 0.0012) . |
| Baseline uTIINE was not a consistent predictor of JSN in OA patients, but serial measurements reflect concurrent JSN . | |
| C2C | In a cohort of 135 RA patients with radiographs at baseline and during follow-up for 5–10 years, no association was found between radiographic progression of hand OA and baseline serum levels of C2C . |
| In a cohort of 87 patients with RA, elevated levels of serum C2C were detected in subjects with rapid radiographic progression compared to slow progressors . And C2C levels were predictive of subsequent radiographic progression rate, in particular JSN. | |
| C2C was not associated with progression of JSN in knee OA patients . | |
| Glucosamine sulphate for 6 months failed to induce significant changes in C2C compared to placebo in knee OA patients . | |
| In patients with ankylosing spondylitis treatment with etanercept induced a significant reduction in levels of C2C ( p = 0.005) compared to placebo controls . | |
| Elevated urinary and serum levels were detected in patients with OA . | |
| In knee OA patients, C2C was correlated with cartilage volume (MRI) and with radiographic assessment . | |
| Infliximab and methroxate reduced C2C in RA, and C2C levels reflected symptoms . | |
| PIIANP | In rheumatoid arthritis patients, with an inadequate response to methotrexate, co-administration with tocilizumab induced a marked dose-dependent reductions in PIIANP levels . |
| In RA, circulating PIIANP did not differ from healthy subjects at study entry or any time during follow-up. In patients with longstanding RA, baseline PIIANP was lower among X-ray progressors than in non-progressors ( p = 0.04) . | |
| PIIANP was significantly lower in hand OA compared with controls and non-hand OA . | |
| Knee OA patients with baseline serum PIIANP levels in the highest quartile, had a 3.2-fold ( p = 0.03) increased risk for radiographic progression (JSN ≥ 2 mm) over 5 years . | |
| Serum PIIANP was decreased by 53% ( p < 0.0001) and 35% ( p < 0.001) in patients with knee osteoarthritis and rheumatoid arthritis, respectively . | |
| PIIANP levels were markedly decreased in patients with knee OA (12.0 ± 3.2 vs 25.8 ± 7.5 ng/ml for OA and controls respectively, p < 0.0001) and RA (14.1 ± 2.5 ng/ml vs 21.7 ± 7.6 ng/ml for RA and controls respectively, p < 0.0001) . | |
| Aggrecan 846 | Baseline levels of the 846 epitope were not related to cartilage loss (MRI) over 1 year , or progression of JSN over 16 months . |
| In RA, serum levels of the 846 epitope 1 year after disease onset were associated with subsequent radiographic progression rate . | |
| In patients with ankylosing spondylitis treatment with etanercept induced a significant increase in the 846 epitope ( p = 0.01) compared to placebo controls . | |
| Patients with ankylosing spondylitis demonstrated significant elevations in serum levels of the Aggrecan 846 epitope . | |
| Aggrecan 846 was unrelated to clinical end points (status or change over 1 year) . | |
| Aggrecan 846 was suppressed in OA subjects compared to controls, i.e. 0.05 vs 0.087 μg/ml, respectively . | |
| Aggrecan fragments | Increased concentrations of aggrecan fragments 374 ARGSV were detected in plasma and synovial fluid from patients with moderate osteoarthritis . |
| Elevated concentrations of the aggrecan 374 ARGS fragments determined by LS-MS/MS have been reported in human urine from OA patients (0.56 ± 0.15 ng/mg creatinine) compared to healthy controls (0.36 ± 0.15 ng/mg creatinine) . In synovial fluid the corresponding levels were 19.3 ± 16.5 ng/ml and 5.5 ± 5.0 ng/ml, respectively. | |
| By immunoassay, the concentration of aggrecan fragments carrying the 374 ARGSV epitope in synovial fluid from patients with acute inflammatory arthritis and acute knee injury (median 88.5 and 53.9 pmol ARGSV/ml, respectively) were elevated compared to subjects with chronic knee injury, knee osteoarthritis and healthy controls (median 0.5, 4.6, and 0.5 pmol ARGSV/ml, respectively) . | |
| Aggrecan fragments carrying the G1 and/or the G2 domain was suppressed 40% in patients with RA compared to healthy controls . | |
| Glc–Gal–PYD | Baseline levels were significantly associated with Sharp score of both erosion ( r = 0.34, p < 0.01) and JSN ( r = 0.36, p < 0.01) . The level of Glc–Gal–PYD was not modulated by therapy over 1 year. |
| Glc–Gal–PYD was significantly associated with K/L-score and JSN of the knee . | |
| Elevated 18% in knee OA and associated with total WOMAC index . | |
| COMP | The decrease in serum COMP after 6 months of etanercept treatment was significant in a small group ( n = 10) of RA patients with remission, but not in the non-remission group ( n = 35) . |
| In a cohort of 135 RA patients with radiographs at baseline and during follow-up for 5–10 years, no association was found between radiographic progression of hand OA and baseline serum levels of COMP . | |
| In 98 RA patients, serum COMP concentration was not different among groups identified with low, moderate, and high disease activity as determined by DAS28-CRP values . | |
| Serum COMP was significantly higher in hand OA compared with controls and non-hand OA . | |
| Change in COMP from baseline to 6 years was associated with development of incident radiographic hip OA, and stabilisation of disease . | |
| In an investigation of 281 RA patients, elevated COMP was associated with more damage in the large joints as determined by Larsen score . However, neither baseline COMP nor 3 months change was associated with change in Larsen score in the large joints over 2 years. | |
| In early RA, baseline levels of serum COMP were associated with radiographic progression (change in Larsen score) ( r = 0.27, p = 0.0038) over 2 years . | |
| Baseline COMP predicted cartilage loss (WORMS) determined by MRI over 30 months . For each SD increase in COMP, the loss of cartilage volume increased 6.1 times, which essentially remained unchanged after adjustment for age, gender and BMI. | |
| Baseline levels were not correlated with 1 year change in cartilage volume or thickness , or 3 years change in JSW or WOMAC in OA patients. | |
| Baseline (and area under the curve per month) of COMP concentrations was significantly higher in patients with tibiofemoral compared to patellofemoral osteoarthritis ( p < 0.01) . | |
| Significant correlation between baseline COMP and JSN in knee OA over 3 years , however no association was detected to Lequesne and WOMAC. | |
| Serum COMP was reported to be unrelated to clinical end points (status or change over 1 year) . | |
| Serum COMP is elevated 16% in knee OA . However, suppressed levels in OA have been reported . | |
| HA | Elevated levels associated with increased risk for progression of radiographic hip OA (ECHODIAH) . |
| Nimesulide but not ibuprofen markedly reduced the serum levels of HA in individuals with flare-up of OA ( p < 0.05) . | |
| HA is unrelated to clinical end points in OA patients (status or change over 1 year) . | |
| Serum levels in OA patients indifferent from controls . |
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
