Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 Feb 1.
Published in final edited form as: Osteoarthritis Cartilage. 2013 Dec 12;22(2):235–241. doi: 10.1016/j.joca.2013.12.004

TSG-6 Activity as a Novel Biomarker of Progression in Knee Osteoarthritis

Hans-Georg Wisniewski 1, Elisa Colón 1, Viktoriia Liublinska 2, Raj J Karia 3, Thomas V Stabler 4, Mukundan Attur 5, Steven B Abramson 5, Philip A Band 6, Virginia B Kraus 4
PMCID: PMC3939799  NIHMSID: NIHMS549925  PMID: 24333293

Abstract

Objective

To establish whether there is an association between TSG-6 activity and osteoarthritis (OA) progression.

Design

TSG-6 activity was determined in 132 synovial fluids from patients with OA of the knee, using a novel quantitative TSG-6 activity assay. The association between TSG-6 activities at baseline and four distinct disease progression states, determined at three-year follow-up, was analyzed using logistic regression.

Results

There was a statistically significant relationship between TSG-6 activity at baseline and all OA progression states over a three-year period. Patient knees with TSG-6 activities in the top 10th percentile, compared to the median activity, had an odds ratio (OR) of at least 7.86 (confidence interval (CI) [3.2, 20.5]) for total knee arthroplasty (TKA) within three years, and of at least 5.20 (CI [1.8, 13.9]) after adjustment for confounding factors. Receiver operating characteristic (ROC) analysis for knee arthroplasty yielded a cut-off point of 13.3 TSG-6 activity units/ml with the following parameters: area under the curve 0.90 (CI [0.804, 0.996]), sensitivity 0.91 (CI [0.59, 0.99]), specificity 0.82 (CI [0.74, 0.88]) and a negative predictive value (NPV) of 0.99 (CI [0.934, 0.994]).

Conclusion

The TSG-6 activity is a promising independent biomarker for OA progression. Given the high NPV, this assay may be particularly suitable for identifying patients at low risk of rapid disease progression and to assist in the timing of arthroplasty.

Introduction

Biomarkers are objective indicators of normal biologic processes, pathogenic processes, or responses to therapeutic intervention [1]. Radiologic staging, patient symptoms, and OA risk factors such as age, gender and body mass index (BMI) have limited value for predicting the risk of rapid OA progression [2–4]. As a result, progression of OA is highly unpredictable. Therefore there is a need to identify biomarkers of OA progression. Although candidate biomarkers have been investigated, their practical use is still very limited and new and better biomarkers are needed [5, 6]. In OA of the knee, biomarkers of disease progression could be of particular value to assist in decisions regarding the timing of TKA or the implementation of long-term lifestyle changes such as weight loss or activity modification. Biomarkers of OA progression are also of particular interest for aiding development of disease-modifying OA drugs (DMOADs) [5]. Although not currently available, several potential DMOADs have shown efficacy in OA models and are in development for treatment of OA in humans [7]. Potential side effects of such drugs or other constraints may restrict their use to patients with the most serious disease or those at highest risk of rapid progression. Thus, biomarkers of OA progression would be useful for identifying high-risk patients who would benefit most from the use of DMOADs. They could also be used for identifying high-risk patients for inclusion in clinical trials, thereby improving study power over current methods [5, 8].

TNFAIP-6, encoded by TNF-stimulated gene 6 (TSG-6), and commonly known as TSG-6 protein [9], is a hyaluronan (HA) binding protein associated with inflammation [10, 11]. Potent anti-inflammatory and chondroprotective activities of this protein have been demonstrated in experimental arthritis models [12–16]. TSG-6 inhibits both osteoblast differentiation and osteoclast activation [17, 18]. TSG-6 mediates the transfer of heavy chains (HCs) from inter-α-inhibitor (IαI) to HA, resulting in the formation of covalent HA-HC complexes [19–21], and has an essential role in the stabilization of the HA-rich extracellular matrix of cumulus cell-oocyte complexes during ovulation, resulting in infertility of female TSG-6-deficient mice [22].

Expression of TSG-6 is induced by pro-inflammatory cytokines, growth factors and hormones [23–27], resulting in a complex expression pattern. Presence of TSG-6 has been demonstrated in synovial fluids of patients with either OA or rheumatoid arthritis (RA) and also in cartilage of both OA and RA patients [28–30]. However, no quantitative studies of TSG-6 in OA or RA patient specimens have been performed and the potential of TSG-6 as a biomarker in these diseases has not yet been explored.

Because a TSG-6 ELISA sufficiently sensitive and specific to be used with biological fluids is currently not available, we developed an assay that measures the TSG-6 activity in synovial fluid under conditions very close to in vivo conditions. In this study, we analyzed the association of TSG-6 activity, determined retrospectively in synovial fluids collected at baseline, with data from a prospective natural history study of OA progression. Our hypothesis was that the TSG-6 activity at baseline might predict progression of OA over a period of three years.

Method

Reagents

Covalink-NH plates were purchased from Nunc (Thermo Fisher Scientific, Pittsburgh, PA). HA was purchased from Lifecore Biomedical (Chaska, MN). Rabbit anti-human IαI and biotinylated goat anti-rabbit Ig were from Dako (Carpinteria, CA), and streptavidin-alkaline phosphatase conjugate was purchased from Invitrogen (Grand Island, NY). Sulfo-NHS was from Pierce (Thermo Fisher Scientific, Rockford, IL), and EDC and p-nitrophenyl phosphate were purchased from Sigma (St. Lous, MO).

OA patient population and sample preparation

We analyzed 132 synovial fluid samples from 91 patients enrolled in the NIH-sponsored POP (Strategies to Predict OA Progression) study. This study was conducted at Duke University. All procedures were in accordance with the ethical standards of the Helsinki Declaration of 1975, as revised in 2000, and were approved by the Duke University Institutional Review Board with informed consent being obtained from all subjects. The approval authorizes the use of the specimens for the current study. All human specimens analyzed in the current study were provided in a de-identified form. All synovial fluids included in this study were collected without lavage and cell free supernatants were prepared and frozen (−80°C) within one hour of collection. Blood samples obtained from some OA patients enrolled in this study were centrifuged at 3500 rpm for 10 min and serum stored in 1 ml aliquots and frozen (−80°C) until analyzed. Matched synovial fluids (N=30) and sera were available for 21 individuals (including synovial fluid from both knees of 9 patients). Inclusion and exclusion criteria for this study and details of the radiographic assessment of OA progression have been described [31]. All participants met radiographic criteria for knee OA based on Kellgren-Lawrence (KL) grades 1–3 [32] and American College of Rheumatology criteria for symptomatic OA of at least one knee [33]. In addition, contralateral knees with symptomatic and/or radiographic OA were also surveyed resulting in the availability of knees of all OA severity grades (KL grades 0–4) for this study. 88 of the 91 patients in this study group had bilateral radiographic OA based on a baseline KL grade of ≥1. Demographic information on the study population can be found in Supplemental Table 1 and information regarding baseline KL grades can be found in Supplemental Table 2. Progression of the disease over a 3-year period was determined by comparison of standardized radiographs taken at baseline and at a 3-year follow-up evaluation [31]. In addition to KL grades, individual joints were scored for osteophytes (OST, range 0–12 per knee) and joint space narrowing (JSN, range 0–6 per knee) using the standardized OARSI atlas [34] of individual radiographic features in OA. Details of the scoring and the inter-rater reliability of scoring have been described [35]. An increase in the OST scores between baseline and follow-up was designated progression based on osteophyte formation (OST+), while an increase in the JSN score was designated progression based on joint space narrowing (JSN+). All joints that showed progressive joint space narrowing also showed progression of osteophyte formation. Mutually exclusive outcome groups were used consisting of non-progressors (NP), OST progressors (OST+/JSN−), JSN progressors (OST+/JSN+), and TKA progressors, i.e. patients who underwent TKA for end-stage OA sometime during the 3-year interval from baseline to follow-up. Radiological progression data for all patient knees are shown in Table 1. A small number (N=5) of synovial fluids from humans without known symptoms of joint disease were procured from the National Disease Research Interchange (NDRI). These specimens were collected at autopsy and stored frozen at −80 °C. Their use in de-identified form in the current study was in accordance with the ethical standards of the Helsinki Declaration of 1975, as revised in 2000, and was approved by the institutional review board of New York University.

Table 1.

Synovial fluid TSG-6 activity by knee OA progression status (N=132).

Outcome category N % Mean (median) TSG-6 activity (u/ml) SD
NP 60 45.5 6.94 (6.12) 4.35
OST+/JSN− 49 37.1 9.87 (8.5) 5.72
OST+/JSN+ 12 9.1 12.68 (11.16) 6.05
TKA 11 8.3 22.61 (21.4) 11.47
Open in a new tab

NP=non-progressor, OST+/JSN− = osteophyte only progressor, OST+/JSN+ = osteophyte and joint space narrowing progressor, TKA=total knee arthroplasty progressor.

Recombinant TSG-6 protein

Recombinant TSG-6 protein was expressed in BTI-TN-5B1-4 insect cells and purified as described [19].

TSG-6 activity assay

The TSG-6 activity assay was carried out in a blinded fashion with respect to the clinical progression status of the samples. HA was coupled covalently to Covalink-NH plates (96-well plates) as described [19]. The plates were blocked by overnight incubation at 37 °C with 0.5% casein in Tris-buffered saline (TBS: 20 mM Tris, 500 mM NaCl, pH 8.0) and stored at 4 °C for up to several months. The TSG-6 activity assay was based on an assay for recombinant TSG-6 using exogenous IαI [19]. In contrast to the earlier assay, the assay used here exclusively relies on IαI present in the analyzed synovial fluid or serum sample. Synovial fluid or serum samples were diluted 1:100 in PBS and incubated for 2 h at 37 °C (100 μl per well, 2 wells per sample) in Covalink-NH plates to which HA had been coupled (see above). After washing 3x with 200 μl of TTBS (20 mM Tris, 500 mM NaCl, 0.1% Tween-20, pH 7.5, used for washing between all incubation steps), wells were incubated for 1 h at 37 °C with a polyclonal rabbit anti-IαI specific for HCs diluted 1:1,000 in TTBS, followed by incubations with biotinylated goat anti-rabbit IgG (1:1,000 in TTBS, 1 h at 37 °C), streptavidin-alkaline phosphatase conjugate (1:1,000 in TTBS, 1 h at 37 °C), and finally with p-nitrophenyl phosphate (2 mg/ml in 50 mM Tris, 2 mM MgCl2, pH 9.5). After about 5–30 min at 37 °C the absorbance at 410 nm was measured, using 750 nm as a reference wavelength. The TSG-6 activity was expressed in units/ml (u/ml). 1 u/ml of TSG-6 activity in a biological sample (synovial fluid or serum) was defined as the HC transfer activity equivalent to the activity of a standard of 1 nM recombinant TSG-6 under standardized conditions, i.e., in a volume of 100 μl PBS containing a human plasma standard at 1:1000 (as source of IαI) and incubated at 37 °C for 2h. This TSG-6 standard was included in each assay (6 wells per plate). A single batch of human plasma and recombinant TSG-6 was used throughout this study, and both standard reagents were stored frozen in aliquots at −80 °C. PBS was included as a negative control. The TSG-6 activity was calculated as the ratio between the mean absorbance of a sample and the TSG-6 standard, multiplied by the dilution factor of the sample (i.e., 100).

Statistics

Logistic regression was used to estimate the association between TSG-6 activity and OA progression based on the above-defined outcome categories while controlling for covariates (for details, see Table 3). The associations were expressed as ORs of progression associated with a one-unit increase of TSG-6 activity. The regression was adapted to correlated data, due to the presence of contralateral knees, using generalized estimating equations (GEE) [36, 37]. Median TSG-6 activities of synovial fluid samples from non-arthritic and OA knees, and median TSG-6 activities of synovial fluid samples from different OA outcome groups (Table 2) were compared using the nonparametric Wilcoxon-Mann-Whitney test. Receiver operating characteristic (ROC) analysis was employed to evaluate the performance of the TSG-6 activity assay [38]. Youden’s index was used to estimate the optimal decision threshold for the differentiation of TKA and non-TKA patient knees [39]. All computations were performed using R software, version 2.15.0.

Table 3.

Logistic regression analysis of the association of the TSG-6 activity in synovial fluid at baseline with OA progression over three years represented by different outcome categories, using generalized estimating equations (GEE).

Odds-ratio (p-value) (CI)
Response TKA (N=11) No Progression (−/−) (N=60) OST+/JSN+ (N=12) TKA or OST+/JSN+ (N=23) Any progression (TKA, OST+/JSN−, OST+/JSN+) (N=72)
No adjustment 1.23 (<0.00001) (1.12,1.35) 0.858 (0.000363) (0.788,0.934) 1.13 (0.0231) (1.02,1.26) 1.20 (<0.0001) (1.1,1.31) 1.18 (<0.0001) (1.09,1.28)
Adjusted for BMI, Age, Gender 1.23 (<0.00001) (1.13,1.35) 0.864 (0.000687) (0.794,0.940) 1.14 (0.0179) (1.02,1.27) 1.20 (<0.0001) (1.11,1.31) 1.18 (<0.0001) (1.09,1.27)
Adjusted for BMI, Age, Gender, Pain 1.19 (<0.0001) (1.1,1.3) 0.865 (0.000726) (0.795,0.941) 1.14 (0.0171) (1.02,1.27) 1.19 (0.00012) (1.09,1.3) 1.18 (<0.0001) (1.09,1.27)
Adjusted for BMI, Age, Gender, Pain, X-ray grade 1.18 (0.00161) (1.06,1.3) 0.896 (0.0126) (0.822,0.978) 1.12 (0.0526) (0.997,1.26) 1.16 (0.00162) (1.06,1.27) 1.13 (0.00125) (1.05,1.22)
Open in a new tab

Abbreviations as per footer of Table 1. CI, 95% confidence interval.

The odds ratio shown is for a 1 u/ml incremental increase of the TSG-6 activity.

All adjustments for covariates refer to baseline data.

Table 2.

Statistical test results for comparisons of median TSG-6 activities between different outcome groups.

Differences between median TSG-6 activities (in u/ml) of outcome groups and (p-values for differences*)
Outcome TKA OST+/JSN+ OST+/JSN−
NP 15.28 (<0.0001) 5.04 (0.002) 2.38 (0.004)
OST+/JSN− 12.9 (0.0001) 2.66 (0.09)
OST+/JSN+ 10.24 (0.02)
Open in a new tab

Abbreviations as per footer of Table 1.

*

Using the nonparametric Wilcoxon-Mann-Whitney test.

In addition, the difference between the median TSG-6 activity of TKA joints and non-TKA joints was 13.8 u/ml and statistically significant (p=0.002), while the difference between the median TSG-6 activity of NP joints and progressor joints (including OST+/JSN−, OST+/JSN+, and TKA joints) was 3.78 u/ml and also statistically significant (p<0.0001).

Results

Quantitative assessment of the TSG-6 activity in synovial fluids of OA patients

We determined the TSG-6 activity in synovial fluid samples from 132 OA patient knees. Figure 1 shows a box plot of the distribution of the TSG-6 activities in the four outcome groups, i.e., NP, OST+/JSN−, OST+/JSN+, and TKA within the three-year follow-up period. Descriptive statistics including sample sizes, percentage of the different outcome groups, mean and median TSG-6 activities, and standard deviations are listed in Table 1. Pairwise comparison of the four outcome groups, using the nonparametric Wilcoxon-Mann-Whitney test, showed that the median TSG-6 activity of the NP group and the TKA group were statistically significantly different from each other and from all other outcome groups (Table 2). The median TSG-6 activity of the group with exclusive osteophyte progression and the group with both osteophyte progression and joint space narrowing did not significantly differ from each other. In addition, a comparison between selected combinations of outcome groups is very instructive. Most notably, the difference between the median TSG-6 activity of TKA joints and non-TKA joints (including NP, OST+/JSN− and OST+/JSN+ joints) was 13.92 u/ml and statistically significant (p=0.002). In addition, the difference between the median TSG-6 activity of NP joints and progressor joints (including OST+/JSN−, OST+/JSN+, and TKA joints) was 5.35 u/ml and it was also statistically significant (p<0.0001). Of note, in patients for which matched synovial fluid (N=30) and serum samples (N=21) were available, TSG-6 activity was higher in synovial fluid than serum in 76.7 % of samples (16.56 ± 13.44 u/ml in synovial fluid of versus 5.62 ± 4.68 u/ml in serum) supporting a joint tissue origin of this marker. Interestingly, the few patients (7 of 30) that had higher TSG-6 activity in serum than in synovial fluids were all non-progressors. Overall there was a non-significant correlation between TSG6 activity in synovial fluid and serum (r=0.11, p=0.58). A comparison of the TSG-6 activity in synovial fluids of OA patients (9.85 ± 7.21 u/ml, median = 7.98 u/ml, N = 132) with that in specimens from humans without known joint disease (3.03 ± 1.70 u/ml, median = 2.99 u/ml, N = 5) suggested elevated TSG-6 activities in OA patients. The difference between the median TSG-6 activities of these two groups was statistically significant (p = 0.004).

Figure 1.

Figure 1

Open in a new tab

TSG-6 activities in synovial fluid by knee OA progression status. Shown is a box plot of TSG-6 activities in four OA outcome categories: NP=non-progressor (N=60); OST=osteophyte only progressor (OST+/JSN−, N=49); JSN=joint space narrowing and osteophyte progressor (OST+/JSN+, N=12); and TKA=total knee replacement progressor (N=11). The boxes indicate 10%, 25%, 75% and 90% quantiles and the median for each outcome group. The line indicates the mean of the whole sample (N=132, mean = 9.85 u/ml).

Logistic regression analysis of TSG-6 activity and clinical progression status

The results of logistic regression analysis, shown in Table 3, estimate the ORs for specific outcomes of OA progression associated with a 1 u/ml increase of the TSG-6 activity in synovial fluid from OA patients collected at baseline. The ORs are greater than 1 for all categories of progression but less than 1 for non-progression. The OR is highest for TKA as outcome, and generally higher for more severe radiographic outcome categories than for less severe ones.

The OR for progression to TKA associated with an increase in TSG-6 activity of 1 u/ml is 1.23, with a range of TSG-6 activities between 0 and 49.6 u/ml for the full dataset. The OR for progression to TKA for patients in the top 10th percentile of TSG-6 activities is particularly informative. Patient knees in the top 10th percentile of TSG-6 activities had TSG-6 values of 17.94 u/ml or more (up to 49.6 u/ml), i.e., at least 9.96 u/ml above the median TSG-6 activity of 7.98 u/ml for the whole sample. Compared to the median TSG-6 activity, this translates into an OR for progression to TKA of at least 7.86 (= 1.239.96, CI [3.2, 20.5]). After adjusting for covariates, including BMI, age, gender, pain severity, and radiographic grade (all baseline values), the OR for progression to TKA for patient-knees in the top 10th percentile as compared to the median TSG-6 activity is at least 5.20 (= 1.189.96, CI [1.8, 13.9]).

Receiver operating characteristics (ROC) analysis of the TSG-6 activity data for the determination of progression to TKA vs. all other outcome groups is shown in Figure 2. The area under the curve is 0.90 (CI [0.804, 0.996]). Application of the Youden index suggested 13.3 u/ml as an approximation for the optimal cut-off point [39]. For this threshold, the sensitivity of the assay is 0.91 (CI [0.59, 0.99]), its specificity is 0.82 (CI [0.74, 0.88]), the positive predictive value (PPV) is 0.31 (CI [0.22, 0.95]), and the NPV is 0.99 (CI [0.934, 0.994]). A contingency table with additional parameters and information about true or false positives or negatives is shown in Table 4.

Figure 2.

Figure 2

Open in a new tab

Receiver operating characteristics (ROC) curve for the TSG-6 activity assay and the outcome total knee arthroplasty (TKA) (vs. non-TKA). The area under the curve is 0.90. A TSG-6 cut-off point of 13.3 u/ml, determined by Youden’s index, yielded a sensitivity of 0.91 and a specificity of 0.82.

Table 4.

Contingency table for the TSG-6 activity assay for the condition TKA vs. non-TKA and a cut-off point of 13.3 u/ml.

TKA positive TKA negative
Test positive TP = 10 FP = 22 PPV = 0.31
Test negative FN = 1 TN = 99 NPV = 0.99
Sensitivity 0.91 Specificity 0.82
Open in a new tab

Abbreviations as per footer of Table 1. PPV, positive predictive value; NPV, negative predictive value; TP, true positive; FP, false positive; TN, true negative; FN, false negative. The positive likelihood ratio (LR+) is 5.06, and the negative likelihood ratio (LR−) is 0.11.

Discussion

The presence of TSG-6 in synovial fluid and cartilage of patients with a spectrum of joint diseases, including OA and RA, has been demonstrated using immunological techniques [28–30]. Using a different approach, Marshall et al. [40] reported decreased expression of TSG-6 in blood leukocytes from patients with mild OA compared to non-arthritic subjects. In contrast, Appleton et al. [41] determined that expression of the TSG-6 gene was upregulated in chondrocytes of rats during early experimental OA. Furthermore, a single nucleotide polymorphism (SNP) in the TSG-6 gene has been linked to the risk of knee OA [42]. A possible linkage between TSG-6 and OA was also found in another genome-wide linkage analysis [43]. None of these studies analyzed the association of TSG-6 gene expression, TSG-6 protein concentration, or TSG-6 activity with disease progression.

We developed an activity assay for the purpose of quantifying TSG-6 in biologic fluids. An ELISA for TSG-6 suitable for determining TSG-6 in serum or synovial fluids has not yet been developed, and our own efforts have not been successful. The protocol for the activity assay is very similar to that of a sandwich ELISA; the main differences are that immobilized HA replaces the capture antibody and that HA-bound HC, rather than antibody-bound TSG-6, is then quantified using an anti-HC detection antibody.

Our study has two main findings: (1) there is a general association between the TSG-6 activity at baseline and subsequent OA progression, covering the full range of outcomes defined in this study, and (2) the one outcome group most clearly set apart by high TSG-6 activity is the one comprising patients who underwent TKA within three years, a group that can be considered to have progressed to end-stage OA.

The strong association of a very high TSG-6 activity with rapid progression to end-stage OA was supported by an OR of at least 7.86 for TKA within three years for patient knees within the top 10th percentile of TSG-6 activities, compared to the median TSG-6 activity. 54.5% of patient-knees in the top 10th percentile of TSG-6 activity progressed to TKA. An OR of greater than 5.20 for this same group, after adjustment for covariates (Table 3), identified TSG-6 activity as a promising independent biomarker for OA progression over three years.

Parameters derived from the ROC curve analysis, such as the area under the curve (0.90), sensitivity (0.91) and specificity (0.82) suggest that the TSG-6 activity assay performs very well in discriminating between OA patients at high or low risk of rapid progression to end-stage OA. In contrast to sensitivity and specificity, the predictive values of the assay are also affected by the low prevalence of TKA in our sample (8.3 % over 36 months). Of note, this rate of TKA in our sample was roughly similar to the risk of TKA of 8.8 % over a 30- month period for OA patients with consistent frequent knee pain in the much larger MOST study [44].

The low prevalence of TKA patients works in favor of the NPV and against the PPV. This is most clearly demonstrated by the very high NPV of 0.99 for our test sample, i.e., 99% of all negative test results were true negatives. Therefore, it should be emphasized that the assay may be particularly suitable for identifying patients who are at low risk of rapid progression to end-stage disease. In our patient sample, 46.7% (28 of 60) of NP knees had KL scores of 3 or 4 at baseline, indicating advanced radiologic OA. 85.7% (24 of 28) of these cases were identified to be at low-risk for TKA by virtue of a TSG-6 activity below the threshold of 13.3 u/ml. A low TSG-6 activity might encourage patients and surgeons to consider non-surgical treatment options and lifestyle modifications that require more time.

Although the specificity of the assay is 0.82, the observed PPV of 0.31 indicates a higher rate of false positive test results, i.e., knees with TSG-6 activities above the threshold that did not have TKA. This is, partially, a consequence of the low prevalence of TKA patients and makes the identification of high-risk patients inherently more difficult. This will require a more differentiated approach to identifying true high-risk patients among the test positives, but also efforts to control certain causes of false-positive test results in future studies. While TKA is an attractive endpoint in clinical studies [45], it has peculiarities that may have increased the rate of false positives in our study. Factors unrelated to OA, such as comorbidities, age, patient willingness or procedure affordability, may prevent or delay TKA and are potential causes for false positive test results. The correct classification of such cases in future studies could potentially improve the accuracy of identifying high-risk patients using the TSG-6 activity assay. Importantly, the evaluation of the individual risk of OA progression is not limited to a high vs. low risk prediction. Individual odds of progression can be estimated on a continuous scale, thereby providing much more detailed information for individual patients. A very high TSG-6 activity may support a decision to undergo TKA. When DMOADs become available, adverse side effects and cost may increase the need to identify high-risk patients for whom DMOAD therapy has a favorable risk-benefit ratio. The ability to differentiate low-risk and high-risk patients could also be very valuable for clinical trials of DMOADs. By excluding prospective low-risk patients, who are unlikely to progress and therefore to benefit from DMOAD treatment, the power of the trial could be increased. In addition, because TSG-6 activity reflects a composite response to multiple pro-inflammatory mediators [23, 24], among other factors [25–27], OA patients with high TSG-6 activity may constitute a subset for whom anti-inflammatory treatment strategies may be particularly beneficial.

A limiting factor for the clinical application of the TSG-6 activity is the availability of OA synovial fluid specimens. However, among patients with mid-or late-stage OA, the increasing likelihood of effusions or of intra-articular drug injections provides opportunities to collect synovial fluid. This group represents the patient population with a particular need to estimate the individual risk of OA progression.

Another limitation of this study is the small number of patients with TKA or progressive JSN, a result of their low prevalence. However, despite the low number of TKA patients, the p-value of the OR for progression to TKA is very low (< 0.00001, Table 3).

Exploratory statistical analysis, logistic regression and ROC analysis strongly suggest that the TSG-6 activity is a promising independent biomarker of OA progression, in particular to end-stage disease. Further prospective studies in different patient populations are needed to verify these findings.

Supplementary Material

01
02

Acknowledgments

This work was funded by a research grant from The Vilcek Foundation (H.-G. W.), and in part by the National Institute of Arthritis and Musculoskeletal and Skin Diseases (RO1AR48769 and P01 AR050245) and the National Institute of Aging (5P30 AG028716) at the National Institutes of Health (V.B.K.). Funds from the Rudin Foundation provided further support (to P.B.). We wish to thank Dr. Stavros Garantziotis, NIH/NIEHS, for his generous gift of rabbit anti-IαI antibody and Drs. Charles Weiss, Jan Vilcek and Lidija Klampfer for helpful suggestions regarding the manuscript.

Role of the funding source

Study sponsors did not have any role in study design, data collection, analysis and interpretation, writing of the manuscript, and the decision to submit the manuscript for publication.

Footnotes

Contributions

Hans-Georg Wisniewski: conception and design, data acquisition, collection and assembly of data, analysis and interpretation of data, drafting of article, final approval of manuscript, obtaining of funding

Elisa Colón: data acquisition, technical and logistic support, collection and assembly of data, analysis of data, critical revision of manuscript, final approval of manuscript

Viktoriia Liublinska: analysis of data, statistical expertise, drafting of article, critical revision of manuscript, final approval of manuscript

Raj J. Karia: analysis of data, collection and assembly of data, critical revision of manuscript, final approval of manuscript

Thomas V. Stabler: administrative and logistic support, provision of study materials, critical revision of manuscript, final approval of manuscript

Mukundan Attur: administrative and logistic support, provision of study materials, critical revision of manuscript, final approval of manuscript

Steven B. Abramson: administrative and logistic support, provision of study materials, critical revision of manuscript, final approval of manuscript

Philip A. Band: conception and design, analysis and interpretation of data, drafting of article, administrative and logistic support, critical revision of manuscript, final approval of manuscript, obtaining of funding

Virginia B. Kraus: conception and design, provision of study materials, analysis and interpretation of data, drafting of article, administrative and logistic support, critical revision of manuscript, final approval of manuscript, obtaining of funding

Competing interest statement

A joint patent application “Quantifying Local Inflammatory Activity And Its Use To Predict Disease Progression And Tailor Treatments” by New York University and Duke University is pending (Inventors P.A.B., H.-G. W., V.B.K.).

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Contributor Information

Hans-Georg Wisniewski, Email: hans-georg.wisniewski@nyumc.org.

Viktoriia Liublinska, Email: vliublin@fas.harvard.edu.

Philip A. Band, Email: philip.band@nyumc.org.

Virginia B. Kraus, Email: vbk@duke.edu.

References

  • 1.De Gruttola VG, Clax P, DeMets DL, Downing GJ, Ellenberg SS, Friedman L, et al. Considerations in the evaluation of surrogate endpoints in clinical trials. summary of a National Institutes of Health workshop. Control Clin Trials. 2001;22:485–502. doi: 10.1016/s0197-2456(01)00153-2. [DOI] [PubMed] [Google Scholar]
  • 2.Eckstein F, Maschek S, Wirth W, Hudelmaier M, Hitzl W, Wyman B, et al. One year change of knee cartilage morphology in the first release of participants from the Osteoarthritis Initiative progression subcohort: association with sex, body mass index, symptoms and radiographic osteoarthritis status. Annals of the rheumatic diseases. 2009;68:674–9. doi: 10.1136/ard.2008.089904. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Hunter DJ. Risk stratification for knee osteoarthritis progression: a narrative review. Osteoarthritis Cartilage. 2009;17:1402–7. doi: 10.1016/j.joca.2009.04.014. [DOI] [PubMed] [Google Scholar]
  • 4.Kraus VB, Feng S, Wang S, White S, Ainslie M, Brett A, et al. Trabecular morphometry by fractal signature analysis is a novel marker of osteoarthritis progression. Arthritis Rheum. 2009;60:3711–22. doi: 10.1002/art.25012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Kraus VB, Burnett B, Coindreau J, Cottrell S, Eyre D, Gendreau M, et al. Application of biomarkers in the development of drugs intended for the treatment of osteoarthritis. Osteoarthritis Cartilage. 2011;19:515–42. doi: 10.1016/j.joca.2010.08.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.van Spil WE, DeGroot J, Lems WF, Oostveen JC, Lafeber FP. Serum and urinary biochemical markers for knee and hip-osteoarthritis: a systematic review applying the consensus BIPED criteria. Osteoarthritis Cartilage. 2010;18:605–12. doi: 10.1016/j.joca.2010.01.012. [DOI] [PubMed] [Google Scholar]
  • 7.Hunter DJ. Pharmacologic therapy for osteoarthritis--the era of disease modification. Nat Rev Rheumatol. 2011;7:13–22. doi: 10.1038/nrrheum.2010.178. [DOI] [PubMed] [Google Scholar]
  • 8.Lohmander LS, Felson D. Can we identify a ‘high risk’ patient profile to determine who will experience rapid progression of osteoarthritis? Osteoarthritis Cartilage. 2004;12 (Suppl A):S49–52. doi: 10.1016/j.joca.2003.09.004. [DOI] [PubMed] [Google Scholar]
  • 9.Lee TH, Wisniewski HG, Vilcek J. A novel secretory tumor necrosis factor- inducible protein (TSG-6) is a member of the family of hyaluronate binding proteins, closely related to the adhesion receptor CD44. J Cell Biol. 1992;116:545–57. doi: 10.1083/jcb.116.2.545. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Milner CM, Day AJ. TSG-6: a multifunctional protein associated with inflammation. J Cell Sci. 2003;116:1863–73. doi: 10.1242/jcs.00407. [DOI] [PubMed] [Google Scholar]
  • 11.Wisniewski HG, Vilcek J. Cytokine-induced gene expression at the crossroads of innate immunity, inflammation and fertility: TSG-6 and PTX3/TSG-14. Cytokine Growth Factor Rev. 2004;15:129–46. doi: 10.1016/j.cytogfr.2004.01.005. [DOI] [PubMed] [Google Scholar]
  • 12.Bardos T, Kamath RV, Mikecz K, Glant TT. Anti-inflammatory and chondroprotective effect of TSG-6 (tumor necrosis factor-alpha-stimulated gene-6) in murine models of experimental arthritis. Am J Pathol. 2001;159:1711–21. doi: 10.1016/s0002-9440(10)63018-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Glant TT, Kamath RV, Bardos T, Gal I, Szanto S, Murad YM, et al. Cartilage-specific constitutive expression of TSG-6 protein (product of tumor necrosis factor alpha-stimulated gene 6) provides a chondroprotective, but not antiinflammatory, effect in antigen-induced arthritis. Arthritis Rheum. 2002;46:2207–18. doi: 10.1002/art.10555. [DOI] [PubMed] [Google Scholar]
  • 14.Mindrescu C, Thorbecke GJ, Klein MJ, Vilcek J, Wisniewski HG. Amelioration of collagen-induced arthritis in DBA/1J mice by recombinant TSG-6, a tumor necrosis factor/interleukin-1-inducible protein. Arthritis Rheum. 2000;43:2668–77. doi: 10.1002/1529-0131(200012)43:12<2668::AID-ANR6>3.0.CO;2-E. [DOI] [PubMed] [Google Scholar]
  • 15.Mindrescu C, Dias AA, Olszewski RJ, Klein MJ, Reis LF, Wisniewski HG. Reduced susceptibility to collagen-induced arthritis in DBA/1J mice expressing the TSG-6 transgene. Arthritis Rheum. 2002;46:2453–64. doi: 10.1002/art.10503. [DOI] [PubMed] [Google Scholar]
  • 16.Szanto S, Bardos T, Gal I, Glant TT, Mikecz K. Enhanced neutrophil extravasation and rapid progression of proteoglycan-induced arthritis in TSG-6-knockout mice. Arthritis Rheum. 2004;50:3012–22. doi: 10.1002/art.20655. [DOI] [PubMed] [Google Scholar]
  • 17.Mahoney DJ, Mikecz K, Ali T, Mabilleau G, Benayahu D, Plaas A, et al. TSG-6 regulates bone remodeling through inhibition of osteoblastogenesis and osteoclast activation. J Biol Chem. 2008;283:25952–62. doi: 10.1074/jbc.M802138200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Mahoney DJ, Swales C, Athanasou NA, Bombardieri M, Pitzalis C, Kliskey K, et al. TSG-6 inhibits osteoclast activity via an autocrine mechanism and is functionally synergistic with osteoprotegerin. Arthritis Rheum. 2011;63:1034–43. doi: 10.1002/art.30201. [DOI] [PubMed] [Google Scholar]
  • 19.Colón E, Shytuhina A, Cowman MK, Band PA, Sanggaard KW, Enghild JJ, et al. Transfer of inter-alpha-inhibitor heavy chains to hyaluronan by surface-linked hyaluronan-TSG-6 complexes. J Biol Chem. 2009;284:2320–31. doi: 10.1074/jbc.M807183200. [DOI] [PubMed] [Google Scholar]
  • 20.Mukhopadhyay D, Asari A, Rugg MS, Day AJ, Fulop C. Specificity of the tumor necrosis factor-induced protein 6-mediated heavy chain transfer from inter-alpha-trypsin inhibitor to hyaluronan: implications for the assembly of the cumulus extracellular matrix. J Biol Chem. 2004;279:11119–28. doi: 10.1074/jbc.M313471200. [DOI] [PubMed] [Google Scholar]
  • 21.Sanggaard KW, Sonne-Schmidt CS, Krogager TP, Kristensen T, Wisniewski HG, Thogersen IB, et al. TSG-6 transfers proteins between glycosaminoglycans via a Ser28-mediated covalent catalytic mechanism. J Biol Chem. 2008;283:33919–26. doi: 10.1074/jbc.M804240200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Fülöp C, Szanto S, Mukhopadhyay D, Bardos T, Kamath RV, Rugg MS, et al. Impaired cumulus mucification and female sterility in tumor necrosis factor-induced protein-6 deficient mice. Development. 2003;130:2253–61. doi: 10.1242/dev.00422. [DOI] [PubMed] [Google Scholar]
  • 23.Kehlen A, Pachnio A, Thiele K, Langner J. Gene expression induced by interleukin-17 in fibroblast-like synoviocytes of patients with rheumatoid arthritis: upregulation of hyaluronan-binding protein TSG-6. Arthritis Res Ther. 2003;5:R186–92. doi: 10.1186/ar762. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Lee TH, Lee GW, Ziff EB, Vilcek J. Isolation and characterization of eight tumor necrosis factor-induced gene sequences from human fibroblasts. Mol Cell Biol. 1990;10:1982–8. doi: 10.1128/mcb.10.5.1982. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Maier R, Wisniewski HG, Vilcek J, Lotz M. TSG-6 expression in human articular chondrocytes. Possible implications in joint inflammation and cartilage degradation. Arthritis Rheum. 1996;39:552–9. doi: 10.1002/art.1780390403. [DOI] [PubMed] [Google Scholar]
  • 26.Robinson LJ, Tourkova I, Wang Y, Sharrow AC, Landau MS, Yaroslavskiy BB, et al. FSH-receptor isoforms and FSH-dependent gene transcription in human monocytes and osteoclasts. Biochemical and biophysical research communications. 2010;394:12–7. doi: 10.1016/j.bbrc.2010.02.112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Yoshioka S, Ochsner S, Russell DL, Ujioka T, Fujii S, Richards JS, et al. Expression of tumor necrosis factor-stimulated gene-6 in the rat ovary in response to an ovulatory dose of gonadotropin. Endocrinology. 2000;141:4114–9. doi: 10.1210/endo.141.11.7784. [DOI] [PubMed] [Google Scholar]
  • 28.Bayliss MT, Howat SL, Dudhia J, Murphy JM, Barry FP, Edwards JC, et al. Up-regulation and differential expression of the hyaluronan-binding protein TSG-6 in cartilage and synovium in rheumatoid arthritis and osteoarthritis. Osteoarthritis Cartilage. 2001;9:42–8. doi: 10.1053/joca.2000.0348. [DOI] [PubMed] [Google Scholar]
  • 29.Wisniewski HG, Maier R, Lotz M, Lee S, Klampfer L, Lee TH, et al. TSG-6: a TNF-, IL-1-, and LPS-inducible secreted glycoprotein associated with arthritis. J Immunol. 1993;151:6593–601. [PubMed] [Google Scholar]
  • 30.Wisniewski HG, Vilcek J. TSG-6: an IL-1/TNF-inducible protein with anti-inflammatory activity. Cytokine Growth Factor Rev. 1997;8:143–56. doi: 10.1016/s1359-6101(97)00008-7. [DOI] [PubMed] [Google Scholar]
  • 31.Denoble AE, Huffman KM, Stabler TV, Kelly SJ, Hershfield MS, McDaniel GE, et al. Uric acid is a danger signal of increasing risk for osteoarthritis through inflammasome activation. Proceedings of the National Academy of Sciences of the United States of America. 2011;108:2088–93. doi: 10.1073/pnas.1012743108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Kellgren JH, Lawrence JS. Radiological assessment of osteo-arthrosis. Annals of the rheumatic diseases. 1957;16:494–502. doi: 10.1136/ard.16.4.494. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Altman RD. Criteria for classification of clinical osteoarthritis. The Journal of rheumatology Supplement. 1991;27:10–2. [PubMed] [Google Scholar]
  • 34.Altman RD, Hochberg M, Murphy WA, Jr, Wolfe F, Lequesne M. Atlas of individual radiographic features in osteoarthritis. Osteoarthritis Cartilage. 1995;3 (Suppl A):3–70. [PubMed] [Google Scholar]
  • 35.Kraus VB, McDaniel G, Worrell TW, Feng S, Vail TP, Varju G, et al. Association of bone scintigraphic abnormalities with knee malalignment and pain. Ann Rheum Dis. 2009;68:1673–9. doi: 10.1136/ard.2008.094722. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Prentice RL, Zhao LP. Estimating equations for parameters in means and covariances of multivariate discrete and continuous responses. Biometrics. 1991;47:825–39. [PubMed] [Google Scholar]
  • 37.Zeger SL, Liang KY. Longitudinal data analysis for discrete and continuous outcomes. Biometrics. 1986;42:121–30. [PubMed] [Google Scholar]
  • 38.Zweig MH, Campbell G. Receiver-operating characteristic (ROC) plots: a fundamental evaluation tool in clinical medicine. Clin Chem. 1993;39:561–77. [PubMed] [Google Scholar]
  • 39.Youden WJ. Index for rating diagnostic tests. Cancer. 1950;3:32–5. doi: 10.1002/1097-0142(1950)3:1<32::aid-cncr2820030106>3.0.co;2-3. [DOI] [PubMed] [Google Scholar]
  • 40.Marshall KW, Zhang H, Yager TD, Nossova N, Dempsey A, Zheng R, et al. Blood-based biomarkers for detecting mild osteoarthritis in the human knee. Osteoarthritis Cartilage. 2005;13:861–71. doi: 10.1016/j.joca.2005.06.002. [DOI] [PubMed] [Google Scholar]
  • 41.Appleton CT, Pitelka V, Henry J, Beier F. Global analyses of gene expression in early experimental osteoarthritis. Arthritis Rheum. 2007;56:1854–68. doi: 10.1002/art.22711. [DOI] [PubMed] [Google Scholar]
  • 42.Valdes AM, Doherty M, Spector TD. The additive effect of individual genes in predicting risk of knee osteoarthritis. Annals of the rheumatic diseases. 2008;67:124–7. doi: 10.1136/ard.2007.075838. [DOI] [PubMed] [Google Scholar]
  • 43.Chen HC, Kraus VB, Li YJ, Nelson S, Haynes C, Johnson J, et al. Genome-wide linkage analysis of quantitative biomarker traits of osteoarthritis in a large, multigenerational extended family. Arthritis Rheum. 2010;62:781–90. doi: 10.1002/art.27288. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Wise BL, Felson DT, Clancy M, Niu J, Neogi T, Lane NE, et al. Consistency of knee pain and risk of knee replacement: the Multicenter Osteoarthritis Study. J Rheumatol. 2011;38:1390–5. doi: 10.3899/jrheum.100743. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Altman RD, Abadie E, Avouac B, Bouvenot G, Branco J, Bruyere O, et al. Total joint replacement of hip or knee as an outcome measure for structure modifying trials in osteoarthritis. Osteoarthritis Cartilage. 2005;13:13–9. doi: 10.1016/j.joca.2004.10.012. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

01
NIHMS549925-supplement-01.doc (28.5KB, doc)
02
NIHMS549925-supplement-02.doc (31KB, doc)

RESOURCES