key: cord-0014273-usw7mzmw authors: Malek, Sarah; Weng, Hsin-Yi; Martinson, Shannon A.; Rochat, Mark C.; Béraud, Romain; Riley, Christopher B. title: Evaluation of serum MMP-2 and MMP-3, synovial fluid IL-8, MCP-1, and KC concentrations as biomarkers of stifle osteoarthritis associated with naturally occurring cranial cruciate ligament rupture in dogs date: 2020-11-19 journal: PLoS One DOI: 10.1371/journal.pone.0242614 sha: 7e471d1af048d5054c6aed346aa376343b08237a doc_id: 14273 cord_uid: usw7mzmw The purpose of this study was to evaluate matrix metalloproteinases (MMP) -2 and MMP-3 in serum, and keratinocyte-derived chemoattractant (KC), interleukin 8 (IL-8) and monocyte chemoattractant 1 (MCP-1) in synovial fluid (SF) as stifle osteoarthritis (OA) biomarkers in dogs. Dogs with naturally occurring cranial cruciate ligament (CrCL) rupture (OA group) and healthy controls were recruited. Stifles with CrCL deficiency were surgically stabilized. Serum, SF, and synovial biopsy samples were collected from the OA group preoperatively, whereas samples were collected once from control dogs. A blinded veterinary pathologist graded synovial biopsies. Serum and SF analyses were performed using xMAP technology. General linear regression was used for statistical comparisons of serum biomarkers, and mixed linear regression for SF biomarkers and temporal concentration changes. The overall discriminative ability was quantified using area under curve (AUC). Spearman’s correlation coefficient was used to assess correlations between synovial histology grades and the biomarkers. Samples from 62 dogs in the OA group and 50 controls were included. The MMP-2 and MMP-3 concentrations between the OA and control groups were not significantly different, and both with an AUC indicating a poor discriminative ability. All three SF biomarker concentrations were significantly different between the OA group and controls (P <0.05). The MCP-1 was the only biomarker showing an acceptable discriminative performance with an AUC of 0.91 (95% confidence interval: 0.83–0.98). The sum of the inflammatory infiltrate score was significantly correlated with all three SF biomarkers (P <0.01). Summed synovial stroma, and all scores combined were significantly correlated with IL-8 and MCP-1 concentrations (P <0.003), and the summed synoviocyte scores were significantly correlated with MCP-1 concentrations (P <0.001). Correlations between MCP-1 concentrations and synovial histopathologic grading and its discriminative ability suggest its potential as a synovitis biomarker in canine stifle OA associated with CrCL rupture. a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 The most common cause of stifle (knee) osteoarthritis (OA) development in dogs is degenerative (non-traumatic) cranial cruciate ligament rupture (CrCLR) [1, 2] . The resulting morbidity arising from the joint instability and related OA has an estimated annual treatment cost of $1.32 billion in the United States [3] . Dogs that develop OA associated with degenerative CrCLR have reported incidence of bilateral CrCLR that ranges from 18-61.3% [4, 5] . In cases of unilateral degenerative CrCLR, subsequent CrCLR in the contralateral stifle is often reported approximately a year after the initial diagnosis with the reported risk ranging from 22-54% [4, 6] . Despite the focused and extensive research on stifle OA in dogs, none of the therapeutic and management strategies have proven efficacious beyond alleviating symptoms and do not control the progression of OA in this joint [7] . Factors considered to be contributing to the limited success include the complexity of the etiology and pathophysiology of canine OA, and a lack of robustly validated biomarkers of OA suitable as reliable outcome measures in all stages of the disease [8] . Clinical examination and standard digital radiography are unable to detect early canine pre-clinical OA changes or objectively evaluate responses to interventions. Therefore, the assessment of molecular changes in biological fluids (i.e., serum, synovial fluid and urine) as soluble (wet) biomarkers of the disease is an area of interest in OA-related research [9, 10] . Additional challenges in soluble biomarker research are posed by the complexity of the sources of the molecules associated with the disease, and the presence of cross-talk among joints that has made identification of a single representative biomarker for identifying and evaluating OA unrealistic [11, 12] . The process of validating biomarkers in the clinical setting also remains a challenge due to difficulties in case definition and selection and in the recruitment and associated costs of verifying the repeatability and accuracy of candidate biomarkers [13, 14] . The investigated biomarkers of canine stifle OA include pro-inflammatory mediators (e.g., cytokines), degradative enzymes and their inhibitors (e.g., matrix metalloproteinases), and extracellular matrix proteins (e.g., proteoglycans, collagen type II degradation or synthesis products) and their composites [8, 15] . Evaluations of candidate biomarkers for canine stifle OA have been performed using meniscectomy, cranial cruciate ligament transection (CrCLt), and groove models [16] [17] [18] [19] . However, the naturally-occuring OA secondary to CrCLR model is of particular interest due to its frequent occurrence in the clinical setting, and similarities to human knee OA [10, 20, 21] . A study by Garner et al. (2011) investigated multiple candidate diagnostic OA biomarkers in serum, synovial fluid (SF), and urine based on experimentally-induced (i.e., CrCLt), and clinical cases of stifle OA associated with naturally occurring canine CrCLR [15] . These authors suggested two matrix metalloproteases (MMP-2 and MMP-3) as candidate serum biomarkers, and a panel of three SF biomarkers (keratinocyte-derived chemoattractant (KC), interleukin 8 (IL-8) and monocyte chemoattractant protein 1 (MCP-1)) as diagnostic biomarkers of stifle OA [15] . However, the number of clinical cases in that study was limited (n = 10), and to the authors' knowledge, no further studies evaluating these panels of serum and SF biomarkers have been published. The overarching goal of this study was to evaluate MMP-2 and MMP-3 in serum, and KC, IL-8 and MCP-1 in SF as stifle OA biomarkers in dogs. We hypothesized that MMP-2 and MMP-3 in serum and KC, IL-8 and MCP-1 in SF would have discriminative abilities as diagnostic, monitoring and predictive biomarkers of OA with close associations with concurrent macroscopic and microscopic stifle joint pathologies related to naturally-occuring OA associated with CrCLR in dogs. The first objective of this study was to evaluate the discriminative ability of these candidate biomarkers for canine stifle OA associated with naturally occurring CrCLR. The second was to evaluate temporal changes in these serum and SF biomarker concentrations after surgical stabilization of the CrCL deficient stifles. The third objective was to assess these serum and synovial biomarkers in evaluating changes in the contralateral stifles that were stable at the time of initial enrollment of the OA group in the study to evaluate the possibility of predicting the fate of the CrCL in the contralateral stifle. The fourth objective was to assess the biomarker associations between additional joint pathologies, including the presence of meniscal injury, degree of CrCL tear, and histological grade of synovitis in the OA affected stifles. In accordance with the Guide to the Care and Use of Experimental Animals of the Canadian Council on Animal Care (#11-062), the Animal Care Committee of the University of Prince Edward Island approved this prospective clinical cohort observational study. Owners of clinical cases and control dogs were required to complete and sign a written consent form prior to enrollment of dogs in the study. Sample size calculation for this study was based on a 95% confidence level and power range 95% for an unmatched case-control study with a single control per two OA cases, resulting in a group size estimate of 27 OA cases with an effect size of 1.287 (for MMP-2 and MMP-3) based on an estimated 80% prevalance of OA. Because there are no true prevalence data for stifle OA in dogs, group sizes were increased based on previous studies [22] to allow for potential sample losses and patient variations (~50 OA and 50 control dogs). An additional 10% allowance was made for the exclusion of samples not found to meet inclusion criteria. This increased sample size was also necessary to allow a meaningful number of the OA group dogs (~45-50) to be included in the recheck groups given the propensity for clinical cases to be lost to follow up. Adult, medium to large breed (>15kg bodyweight) client-owned dogs with a clinical diagnosis of naturally occurring CrCLR in one or both stifles were recruited in the OA group. Additional inclusion criteria for the OA group included being free of systemic disease based on complete physical, neurologic and orthopedic examinations, and lack of significant abnormalities on a complete blood count and serum biochemistry profile. Exclusion criteria were additional orthopedic abnormalities (e.g., patella luxation), history of surgery or use of systemic corticosteroids within 4 weeks before recruitment, history of traumatic CrCL tear, or a history of intra-articular corticosteroid injections. Pre-surgical diagnosis of CrCLR was based on observation of clinical lameness in the affected hind limb and the presence of one or more of the following criteria: pain on hyperextension of the stifle, palpable joint effusion, positive cranial drawer test, or positive tibial thrust test [6] . A diagnosis of stifle OA was confirmed based on orthogonal radiographs of the stifle joint and intraoperative observation of CrCLR with gross evidence of OA (e.g., osteophytes, synovitis, joint effusion, cartilage lesions) [23] . The unmatched control group included adult, medium to large breed (>15kg bodyweight) dogs euthanized for reasons unrelated to this project. These dogs were free of orthopedic or systemic abnormalities based on physical and orthopedic examinations immediately prior to euthanasia. Postmortem examination of both stifles was performed to confirm the lack of gross abnormalities in all compartments of both joints. Dogs with concurrent systemic illness, or positive for dirofilariasis, ehrlichiosis, anaplasmosis, or borreliosis based on a commercial ELISA-based test (SNAP 1 4D 1 Test, IDEXX Laboratories, Westbrook, ME.) were excluded. All OA group dogs with CrCLR underwent either tibial plateau leveling osteotomy or lateral fabellotibial suture techniques on the CrCL deficient joints to stabilize the stifle [24, 25] . For bilaterally affected dogs, single-stage bilateral surgeries or staged procedures were performed based upon surgeon preference or the owner's financial constraints. In the case of the staged procedures, the second stifle was not operated until after the conclusion of the study; these dogs were not recruited a second time within the study. At the time of surgery during exploration of the joint via arthroscopy or arthrotomy, the CrCL of the operated stifle was evaluated for the presence of a partial or a complete tear of the ligament. The medial and lateral menisci in the operated stifle were also evaluated for the presence or absence of any tear or damage at the time of surgery. In all operated stifles, damaged or diseased components of the CrCL and menisci were debrided with no attempts at reconstruction of the torn ligaments or menisci. The postoperative pain management regimen included; hydromorphone (Hydromorphone hydrochloride, 2mg/ml, Sandoz Canada Inc., Quebec) at 0.05-0.1mg/kg IV or SC every 4-6 hours for 48 hours, meloxicam (Metacam 1 , 1.5mg/ml, Boehringer-Ingelheim, Burlington, ON) at 0.1mg/kg, orally, every 24 hours for 7-10 days, and tramadol (Tramadol, Chiron, Compounding Pharmacy Inc., Guelph, ON) at 4-8mg/kg, orally, every 8-12 hours for 5-7 days. Serum samples. At the initial visit (T1), a 4 ml venous blood sample was collected preoperatively and allowed to clot before being centrifuged at 5000 rpm (3400 g) for 5 minutes. The serum was separated and stored in cryovials (Nalgene Cryogenic tubes, VWR International, Batavia, IL, USA) at -80˚C for later batch analysis. A subset of the OA dogs were selected for re-evaluation and sampling at 4 (T2), and 12 (T3) weeks after surgery. At each revisit (T2 and T3), physical and orthopedic examinations were performed prior to radiographic imaging of stifles under sedation. Venous blood collection and processing of samples were performed using the same methodology described for the initial visit. Exclusion criteria for this subset of OA dogs included clinical or radiographic evidence of infection at the level of the joint or implant site, catastrophic implant failure, or diagnosis of any other systemic illness. Blood samples from control dogs were obtained once before euthanasia and similarly processed. Synovial fluid samples. Samples were collected from both stifles irrespective of whether one or both stifles were affected with CrCLR at the T1 time point. On the day of surgery, the dogs in the OA group were placed under general anesthesia, and stifles were clipped free of hair and prepared for aseptic arthrocentesis. Sterile 6 ml syringes with 1.5", 22-gauge hypodermic needles were used to aseptically aspirate SF from each joint. The SF samples were placed in labeled cryovials and frozen immediately at -80˚C for later batch analysis. Surgical stabilization of the CrCL deficient stifle(s) was performed after the SF sample was obtained. At the subsequent visits (T2 and T3), bilateral stifle SF sample collections were performed using aseptic technique and under sedation following radiographs of the joint. Samples obtained from the CrCL deficient stifle with OA were labeled as OA stifle (index stifle), and the samples collected from the opposite stifle that had an intact CrCLR (stable stifle), based on clinical examination, at the time of enrollment and throughout the study were labeled as contralateral. Therefore, a dog that had bilateral CrCL deficient stifles, did not have a contralateral sample. In the control group of dogs, the SF samples from both stifle joints were aseptically collected immediately after euthanasia using the same method and similarly stored. Both SF samples obtained from each control dog were labeled as control samples. A flow diagram in Fig 1 summarises the sampling order from the OA and control group dogs. Multiplex bead assay. The serum and SF samples frozen at -80˚C were shipped overnight on dry ice to the University of Missouri's Comparative Orthopedic Laboratory and stored at -80˚C. At the time of analyses, samples were thawed, and an aliquot from each SF (100μl), and serum (50μl) sample was processed for testing. The SF samples were centrifuged at 14,000 rpm for 10 minutes to pellet debris, and the supernatant removed. The SF was then incubated with hyaluronidase (MPBiomedicals, LLC, Solon, Ohio) at 37˚C for 90 minutes to decrease viscosity. After digestion, the SF samples were analyzed using a custom luminex canine cytokine Prevalence of and risk factors for hip dysplasia and cranial cruciate ligament deficiency in dogs Epidemiology of cruciate ligament rupture Estimate of the annual economic impact of treatment of cranial cruciate ligament injury in dogs in the United States Contralateral cruciate survival in dogs with unilateral non-contact cranial cruciate ligament rupture Comparison of tibial plateau angles in dogs with unilateral versus bilateral cranial cruciate ligament rupture: 150 cases History and Clinical Signs of Cruciate Ligament Rupture Systematic review of the management of canine osteoarthritis Canine synovial fluid biomarkers for early detection and monitoring of osteoarthritis Osteophytosis, subchondral bone sclerosis, joint effusion and soft tissue thickening in canine experimental stifle osteoarthritis: comparison between 1.5 T magnetic resonance imaging and computed radiography Models of osteoarthritis: the good, the bad and the promising. Osteoarthritis and Cartilage Call for standardized definitions of osteoarthritis and risk stratification for clinical trials and clinical use Association of biomarkers with pre-radiographically defined and radiographically defined knee osteoarthritis in a population-based study Biomarkers for osteoarthritis: current position and steps towards further validation Use of biomarkers and surrogate endpoints in drug development and regulatory decision making: Criteria, validation, strategies Using animal models in osteoarthritis biomarker research Serum and synovial fluid concentrations of keratan sulfate and hyaluronan in dogs with induced stifle joint osteoarthritis following cranial cruciate ligament transection Treatment with licofelone prevents abnormal subchondral bone cell metabolism in experimental dog osteoarthritis The canine 'groove' model, compared with the ACLT model of osteoarthritis Longitudinal characterization of synovial fluid biomarkers in the canine meniscectomy model of osteoarthritis Spontaneous and experimental osteoarthritis in dog: similarities and differences in proteoglycan levels Animal models of osteoarthritis: challenges of model selection and analysis Synovial fluid interleukin 6, tumor necrosis factor, and nitric oxide values in dogs with osteoarthritis secondary to cranial cruciate ligament rupture Correlation between osteoarthritic changes in the stifle joint in dogs and the results of orthopedic, radiographic, ultrasonographic and arthroscopic examinations Advances in the canine cranial cruciate ligament 2ed Extracapsular stabilization Detection and evaluation of matrix metalloproteinases involved in cruciate ligament disease in dogs using multiplex bead technology Piermattei's Atlas of Surgical Approaches to the Bones and Joints of Dog and Cat The OARSI histopathology initiative-recommendations for histological assessments of osteoarthritis in the rabbit The OARSI histopathology initiativerecommendations for histological assessments of osteoarthritis in the dog Recent findings on the role of gelatinases (matrix metalloproteinase-2 and -9) in osteoarthritis High MMP-1, MMP-2, and MMP-9 protein levels in osteoarthritis MMP-2 as an early synovial biomarker for cranial cruciate ligament disease in dogs Matrix metalloproteinase activity in stifle synovial fluid of cranial cruciate ligament deficient dogs and effect of postoperative doxycycline treatment The role of stromelysin in the cartilage destruction that accompanies inflammatory arthritis Significant increases in serum and plasma concentrations of matrix metalloproteinases 3 and 9 in patients with rapidly destructive osteoarthritis of the hip Synovial MMP-3 and TIMP-1 levels and their correlation with cytokine expression in canine rheumatoid arthritis Collagenolytic protease expression in cranial cruciate ligament and stifle synovial fluid in dogs with cranial cruciate ligament rupture Isolation of novel GRO genes and a phylogenetic analysis of the CXC chemokine subfamily in mammals Flow cytometric analysis of intracellular chemokines in chondrocytes in vivo: constitutive expression and enhancement in osteoarthritis and rheumatoid arthritis Synovitis in dogs with stable stifle joints and incipient cranial cruciate ligament rupture: a cross-sectional study Journal of interferon & cytokine research: the official journal of the International Society for Interferon and Cytokine Research IL-8/CXCL8 and growth-related oncogene alpha/CXCL1 induce chondrocyte hypertrophic differentiation Tenidap decreases IL-8 and monocyte chemotactic peptide-1 (MCP-1) mRNA expression in the synovial tissue of rabbits with antigen arthritis and in cultured synovial cells The development of clinical signs of rheumatoid synovial inflammation is associated with increased synthesis of the chemokine CXCL8 (interleukin-8) Borrelia burgdorferi migrates into joint capsules and causes an up-regulation of interleukin-8 in synovial membranes of dogs experimentally infected with ticks Serum concentrations of monocyte chemoattractant protein-1 in healthy and critically ill dogs Use of Serum Concentrations of Interleukin-18 and Monocyte Chemoattractant Protein-1 as Prognostic Indicators in Primary Immune-Mediated Hemolytic Anemia in Dogs Radiographic risk factors for contralateral rupture in dogs with unilateral cranial cruciate ligament rupture Radiographic and magnetic resonance imaging predicts severity of cruciate ligament fiber damage and synovitis in dogs with cranial cruciate ligament rupture Effect of transection of the caudal menisco-tibial ligament on medial femorotibial contact mechanics Are bacterial load and synovitis related in dogs with inflammatory stifle arthritis? Carprofen simultaneously reduces progression of morphological changes in cartilage and subchondral bone in experimental dog osteoarthritis Effects of non-steroidal antiinflammatory drugs and dexamethasone on the activity and expression of matrix metalloproteinase-1, matrix metalloproteinase-3 and tissue inhibitor of metalloproteinases-1 by bovine articular chondrocytes Suppression of matrix metalloproteinase production from synovial fibroblasts by meloxicam in-vitro The effect of non-steroidal anti-inflammatory drugs on matrix metalloproteinases levels in patients with osteoarthritis The authors would like to thank Dr. Trina Bailey for contributing OA group samples, Dr. Aaron Stoker and Dr. Jimmy Cook for facilitating the multiplex bead assay analysis of the study samples at the University of Missouri's Comparative Orthopedic Laboratory.