A.R. Poole

Application of biomarkers in the development of drugs intended for the treatment of osteoarthritis

OBJECTIVE Osteoarthritis (OA) is a chronic and slowly progressive disease for which biomarkers may be able to provide a more rapid indication of therapeutic responses to therapy than is currently available; this could accelerate and facilitate OA drug discovery and development programs. The goal of this document is to provide a summary and guide to the application of in vitro (biochemical and other soluble) biomarkers in the development of drugs for OA and to outline and stimulate a research agenda that will further this goal. METHODS The Biomarkers Working Group representing experts in the field of OA biomarker research from both academia and industry developed this consensus document between 2007 and 2009 at the behest of the Osteoarthritis Research Society International Federal Drug Administration initiative (OARSI FDA initiative). RESULTS This document summarizes definitions and classification systems for biomarkers, the current outcome measures used in OA clinical trials, applications and potential utility of biomarkers for development of OA therapeutics, the current state of qualification of OA-related biomarkers, pathways for biomarker qualification, critical needs to advance the use of biomarkers for drug development, recommendations regarding practices and clinical trials, and a research agenda to advance the science of OA-related biomarkers. CONCLUSIONS Although many OA-related biomarkers are currently available they exist in various states of qualification and validation. The biomarkers that are likely to have the earliest beneficial impact on clinical trials fall into two general categories, those that will allow targeting of subjects most likely to either respond and/or progress (prognostic value) within a reasonable and manageable time frame for a clinical study (for instance within 1-2 years for an OA trial), and those that provide early feedback for preclinical decision-making and for trial organizers that a drug is having the desired biochemical effect. As in vitro biomarkers are increasingly investigated in the context of specific drug treatments, advances in the field can be expected that will lead to rapid expansion of the list of available biomarkers with increasing understanding of the molecular processes that they represent.

Evidence to suggest that cathepsin K degrades articular cartilage in naturally occurring equine osteoarthritis

OBJECTIVE The mechanisms leading to degeneration of articular cartilage in osteoarthritis (OA) are complex and not yet fully understood. Cathepsin K (CK) is a cysteine protease which can also cleave the triple helix of type II collagen. This exposes a neoepitope that can now be identified by specific antibodies. The aim of this study was to obtain evidence suggesting a role for CK in naturally occurring equine OA in both lesional and peri-lesional regions. METHODS Articular cartilages (n=12 horses; 5 healthy, 7 OA) were harvested from animals postmortem. A gross macroscopic examination, histologic (Safranin O-Fast Green and Picrosirius red staining) and immunohistochemical evaluation were performed. Samples were divided into normal appearing cartilage, peri-lesional and lesional cartilage. Cartilage degradation in the samples was graded histologically and immunohistochemically. CK and possible CK cleavage were detected immunohistochemically with specific anti-protein and anti-neoepitope antibodies, respectively. A comparison of CK neoepitope (C2K) production with the collagenase-generated neoepitope produced by matrix metalloproteinases (MMP)-1, 8 and 13 (C2C) was also assessed immunohistochemically. RESULTS CK and CK cleavage were significantly more abundant in OA cartilage (both peri-lesional and lesional) when compared to remote cartilage within the sample joint or cartilage from healthy joints. The immunohistochemical pattern observed for CK degradation (C2K) was similar to that of collagenase degradation (C2C). Macroscopic cartilage changes and histologic findings were significantly correlated with immunohistochemistry results. CONCLUSION The data generated suggests that CK may be involved in cartilage collagen degradation in naturally occurring osteoarthritis.

Relationship between pre-radiographic cartilage damage following anterior cruciate ligament injury and biomarkers of cartilage turnover in clinical practice: a cross-sectional observational study.

OBJECTIVE To determine whether differences in synovial fluid (SF) biomarkers of collagen and proteoglycan turnover are associated with pre-radiographic damage to articular cartilage and menisci following anterior cruciate ligament (ACL) injury and are of clinical value. METHOD SF samples from ACL injured knees of 108 patients were obtained when damage to cartilages and menisci was evaluated arthroscopically. Concentrations of SF collagenase-generated cleavage neoepitope of type II collagen (C2C) were determined using ELISA and aggrecan-derived disaccharides of chondroitin-4-sulfate (Δdi-C4S), chondroitin-6-sulfate (Δdi-C6S), and keratan sulfate (KS), were measured in SF by High performance liquid chromatography (HPLC). RESULTS Radiographic examination failed to detect any intra-articular degenerative changes. The number of high-grade cartilage lesions was positively associated with age, duration after injury and the level of C2C, and negatively with the level of KS. There was no association between the number of high-grade cartilage and meniscal lesions. Multivariable logistic regression revealed significant associations of increased C2C (adjusted Odds ratio (OR) of the upper quartile to remainder of 2.49, 95% Confidence interval (CI) = 0.85-7.27) and decreased KS (adjusted OR of the lower quartile to the remainder of 3.32, 95% CI = 1.19-9.24) with the presence of three or more high-grade cartilage lesions, independent of age and duration after injury. The combined impact of increased C2C and decreased KS was 22.8 (95% CI = 1.95-265.9), far exceeding the impact of each independent biomarker. CONCLUSION Combinations of the C2C and KS as described here may offer greater ability to identify patients with early pre-radiographic high-grade cartilage damage compared to single clinical or biomarker parameters.

Influences of menopause, aging, and gender on the cleavage of type II collagen in cartilage in relationship to bone turnover.

Objective: It is unclear whether there are changes in cartilage turnover after menopause, although these are well documented in bone. The aim of this study was to explore the possibility that menopause might change cartilage turnover as well as bone turnover. We also examined age and gender to estimate the independent influences of these parameters together with menopause on cartilage and bone turnover. Design: Serum samples were collected from 140 healthy volunteers, 69 men (mean age ± SD, 42.8 ± 13.8 y) and 71 women (44.4 ± 10.5 y) with self-reported pre- or postmenopausal status who had no swelling or pain in their spine and joints. Body mass index was also recorded. The serum concentration of a biomarker of cartilage type II collagen (C2C) degradation and concentrations of serum bone alkaline phosphatase and urine cross-linked type I collagen N-telopeptide, both accepted biomarkers of bone turnover, were measured together. Analyses of covariance were performed to test the mean differences of biomarkers by gender and by menopause status with age adjustment. Results: In men there were no significant correlations between age and the biomarkers. In women C2C concentration decreased with age. Cross-linked type I collagen N-telopeptide and bone alkaline phosphatase levels were both increased after menopause, whereas C2C showed no detectable changes. C2C was not significantly related to body mass index. Conclusions: This study suggests that there are fundamental differences in the degradation of uncalcified C2C and bone type I collagen degradation and bone assembly in response to menopause and aging.

Ability of a Urine Assay of Type II Collagen Cleavage by Collagenases to Detect Early Onset and Progression of Articular Cartilage Degeneration: Results from a Population-based Cohort Study

Objective. To evaluate the association of a sandwich assay for cartilage collagenase-mediated degradation, the C2C human urine sandwich assay (IB-C2C-HUSA), with early and late knee cartilage pathology and with progression of cartilage damage. Methods. A population-based cohort with knee pain, age 40–79 years, was evaluated at baseline (n = 253) and after mean 3.3 years (n = 161). We evaluated the IB-C2C-HUSA and a related competitive inhibition assay (C2C). The C2C assay was applied to serum (sC2C) and urine (uC2C). Based on knee radiographs and magnetic resonance imaging (MRI), 3 subgroups [no cartilage pathology, preradiographic cartilage pathology, and radiographic osteoarthritis (ROA)] were evaluated cross-sectionally for association with biomarker levels. Longitudinally, we evaluated whether baseline assays predict subsequent progression of cartilage degeneration, defined by MRI cartilage loss. Results. Cross-sectionally, statistically significant differences were seen in the 3 subgroups for IB-C2C-HUSA (p < 0.001), with the highest levels seen in ROA, and for sC2C (p = 0.023), while no differences were seen for uC2C (p = 0.501). Baseline IB-C2C-HUSA levels were higher in progressors vs nonprogressors (p = 0.003). In logistic regression analysis, only baseline IB-C2C-HUSA was associated with an increased risk of progression of cartilage damage (OR 1.78, 95% CI 1.03–3.09). Conclusion. The IB-C2C-HUSA degradation assay detects the generation of a pathology-related cartilage collagen peptide(s) that increase(s) with onset of degeneration of knee articular cartilage. The baseline values are associated with progression of cartilage degeneration over 3 subsequent years. This assay may have value in clinical OA trials. Further, it points to collagenase activity as a therapeutic target for controlling degeneration of articular cartilage.

P224 IDENTIFICATION OF EARLY KNEE OSTEOARTHRITIS: RESULTS FROM THE DEVELOPMENT OF A MODEL FOR THE DIAGNOSIS OF EARLY KNEE OSTEOARTHRITIS (MODEKO) STUDY

J. Cibere1, H. Zhang1, A. Thorne1, H. Wong1, J.A. Kopec1, J. Singer1, A. Guermazi2, C. Peterfy2, S. Nicolaou1, P. Munk1, P. Garnero3, A.R. Poole4, T. Lobanok4, V.B. Kraus5, A. Way5, T. Saxne6, J.M. Esdaile1 1University of British Columbia, Vancouver, BC, Canada, 2Synarc Inc, San Francisco, CA, 3INSERM Research Unit 664, Synarc, Lyon, France, 4McGill University, Montreal, PQ, Canada, 5Duke University, Durham, NC, 6Lund University Hospital, Lund, Sweden

113 ANALYSIS OF 31 BIOMARKERS AND BIOMARKER FACTORS IN PRE-RADIOGRAPHIC AND RADIOGRAPHIC KNEE OSTEOARTHRITIS: RESULTS OF A POPULATION-BASED STUDY USING MRI

When biomarker concentrations of all patient samples were plotted in chronological order of assaying, it appeared that there was a sufficient window of variation within a positively skewed distribution. In general there was no evident variation over time and between assays, except for two remarkable observations: For the sCOMP assay there was a clear difference in measured biomarker concentrations between the first 6 kits and last 8 kits. In the first kits measured biomarker concentrations were a 10-fold higher and showed significantly more variation than in the last kits. For the sC1,2C and sCS846 assays, there was a significant negative correlation between the measured biomarker concentration and sample order (chronologic/position) in the assay plates, repeatedly found for each of the assay plates. Conclusions: Reliable biomarker measurement in this large sample set seems possible for the majority of the studied biomarkers. However, despite attempts to minimize variation, there are some striking challenges, presumably technical in nature. The remarkable observations in the sCOMP assay, and the sC1,2C and sCS846 assays need further evaluation (as is presently ongoing) before correlations with the clinical and radiographic data sets can be made.