A new sandwich biomarker assay for cartilage type ii collagen can discriminate between no osteoarthritis (oa), pre-radiographic and radiographic knee oa in human urine

s / Osteoarthritis and Cartilage 20 (2012) S54–S296 S80 Conclusions: The current study provides a comprehensive profile of cytokines, chemokines and growth factors present in the OA and the healthy joint environment. Increased soluble mediators, such as IL-6, IP-10, MDC and RANTES, indicate the involvement of a catabolic process in OA. This biomarker profile might be useful to distinguish OA patients from healthy donors, however needs to be validated in future experiments. 148 A NEW SANDWICH BIOMARKER ASSAY FOR CARTILAGE TYPE II COLLAGEN CAN DISCRIMINATE BETWEEN NO OSTEOARTHRITIS (OA), PRE-RADIOGRAPHIC AND RADIOGRAPHIC KNEE OA IN HUMAN URINE. N. Ha , S. Bourdon , E.C. Sayre , A.R. Poole , A. Thorne , H. Wong , J. Singer , J. Kopec , J.M. Esdaile , A. Guermazi , N. Savvakis , J. Cibere . 1 IBEX Pharmaceuticals Inc., Montreal, QC, CANADA; Univ. of British Columbia, Vancouver, BC, CANADA; Arthritis Res. Ctr. of Canada, Vancouver, BC, CANADA; McGill Univ., Montreal, QC, CANADA; Canadian HIV Trials Network, Vancouver, BC, CANADA; Ctr. for Hlth.Evaluation and Outcome Sci., Vancouver, BC, CANADA; Boston Univ. Med. Ctr., Boston, MA; Vancouver Gen. Hosp., Vancouver, BC, CANADA Purpose: Molecular biomarkers, especially those that detect cleavage of cartilage type II collagen, offer the potential to detect early pre-radiographic osteoarthritis (PROA) and radiographic OA (ROA). Increased type II collagen cleavage by collagenases is a well-established feature in OA knee articular cartilage. Yet existing assays have exhibited limited ability to discriminate between these patient populations and those without OA. By focusing on a 45 mer collagenase-generated peptide of type II collagen found in increased amounts in urine of OA patients (Nemirovskiy et al, 2007), a new assay has been developed that can discriminate knee PROA and knee ROA from subjects with no detectable OA (NoOA) on magnetic resonance imaging (MRI). Initial encouraging results are presented here. Methods: The newly developed sandwich assay for urine (IB-C2CHUSATM) detects the type II collagen specific C-terminus of the /4 piece generated by collagenase and an intrachain epitope within 45 residues of this cleavage site. This was compared to an existing competitive inhibition assay (C2C ELISA) that also detects this cleavage site neoepitope in type II collagen alpha-chains and any other alpha-chain fragments containing this neoepitope. Assay results were expressed as pg per mmol creatinine. Urine samples from the population-based Model for the Diagnosis of Early Knee Osteoarthritis (MoDEKO) cohort were analyzed. The unique MoDEKO cohort consists of subjects with knee pain, age 40 to 79, classified based on knee MRI and x-ray (34 NoOA; 120 PROA and 98 ROA_sample-weighted counts). Data were analyzed for differences between OA subgroups using descriptive statistics. Multinomial logistic regression analysis was performed to evaluate the associations between biomarkers and radiographically/MRI defined OA subgroups vs. NoOA, reported as odds ratios (ORs) and 95% confidence intervals (CI) after log-transformation and adjustment for age, gender and BMI. All analyses were sample-weighted. Results: Results for the C2C ELISA inhibition assay revealed no ability to discriminate between the different groups in urine (Fig.1). In contrast, the IB-C2C-HUSATM sandwich assay revealed the presence of progressively higher concentrations of the cleavage neoepitope in PROA and ROA groups compared to the NoOA group (Fig.1). Multinomial logistic regression results revealed that the new urine assay could significantly predict the three groups (overall p <0.001): OR 2.90 (95% CI 1.64, 5.13) for ROA vs. NoOA, OR 1.55 (95% CI 0.95, 2.50) for PROA vs. NoOA and OR 1.88 (95% CI 1.27, 2.77) for ROA vs. PROA. Interestingly, this discrimination was often stronger when urine data were not corrected for creatinine: OR 3.97 (2.03, 7.78) for ROA vs. NoOA ; OR 2.30 (1.22, 4.32) for PROA vs. NoOA and OR 1.73 (1.23, 2.43) for ROA vs. PROA. Conclusions: Our results demonstrate the ability of this new urine-specific sandwich assay to discriminate between NoOA, PROA and ROA in amanner not previously possible with the original C2C inhibition assay. It also provides further evidence to suggest that the 45mer alpha-chain fragment of type II collagen, that contains the collagenase cleavage site, is generated in increased amounts in knee OA and can be selectively detected in urine. This new biomarker assay may be of value in studying disease onset, progression and treatment in OA. Ă 149 MEASUREMENTS OF URINARY COLL2-1NO2 IS PREDICTIF OF JOINT SPACE NARROWING IN KNEE OSTEOARTHRITIS Y. Henrotin , V. Kraus , K.D. Brandt , S.A. Mazzuca , J. Huebner , B. Boulanger , M. Deberg . Bone and Cartilage Res. Unit, Liege, Belgium; Duke Univ. Med. Ctr., Durham, NC; Univ. Sch. of Med., Indianapolis, IN; Arlenda SA, Liege, Belgium; Artialis SA, Liege, Belgium Purpose: To evaluate the predictive value of Coll2-1NO2, a biochemical marker of inflammation-related cartilage degradation, for radiographic knee osteoarthritis (OA) progression. Coll2-1NO2 [HRGY(NO2)PGLDG], is the nitrated from of Coll2-1 (HRGYPGLDG), a denaturation epitope of the type II collagen molecule. Methods: Two OA populations were studied: population 1 was composed by 135 obese men andwomen aged 45 to 64 years with mild andmoderate knee OA (presence of osteophytes and a minimum joint space width 2 mm in themedial compartment); population 2 included 182 obesewomen with unilateral radiographic knee OA (K&L II or III) aged 45 to 64 years from the placebo arm of the doxycycline study (Brandt et al., 2004). For both populations, quantitative estimates of changes in joint space width (JSW) in the medial tibiofemoral compartment were obtained from fluoroscopically assisted semi-flexed AP radiographs performed at baseline,16 and 30 months. The creatinine-adjusted urinary Coll2-1NO2 (uColl2-1NO2) concentrationwas measured on samples obtained at baseline, after 16 and 30 months for the patients from population 1 and at baseline and after 6, 12, 18, 24 and 30 months for the patients from population 2. Patients showing a decrease of JSW 0.5 mm over 30 months follow-up were designated radiographic progressors. The comparison of progressor and nonprogressor patients at inclusion was performed by an unpaired t-test. The results were expressed as medians (minimum and maximum). The ability of biomarker to predict radiographic progression of knee osteoarthritis was evaluated by logistic regression analysis. The covariates in the model are uColl2-1NO2 at baseline and the change over 16 months of uColl2-1NO2.