Identifying Potential Osteoarthritis Biomarkers
Identifying Potential Osteoarthritis Biomarkers
Background Research for the use of biomarkers in osteoarthritis (OA) is promising, however, adequate discrimination between patients and controls may be hampered due to innate differences. We set out to identify loci influencing levels of serum cartilage oligomeric protein (sCOMP) and urinary C-telopeptide of type II collagen (uCTX-II).
Methods Meta-analysis of genome-wide association studies was applied to standardised residuals of sCOMP (N=3316) and uCTX-II (N=4654) levels available in 6 and 7 studies, respectively, from TreatOA. Effects were estimated using a fixed-effects model. Six promising signals were followed up by de novo genotyping in the Cohort Hip and Cohort Knee study (N=964). Subsequently, their role in OA susceptibility was investigated in large-scale genome-wide association studies meta-analyses for OA. Differential expression of annotated genes was assessed in cartilage.
Results Genome-wide significant association with sCOMP levels was found for a SNP within MRC1 (rs691461, p=1.7Ă—10) and a SNP within CSMD1 associated with variation in uCTX-II levels with borderline genome-wide significance (rs1983474, p=8.5Ă—10). Indication for association with sCOMP levels was also found for a locus close to the COMP gene itself (rs10038, p=7.1Ă—10). The latter SNP was subsequently found to be associated with hip OA whereas COMP expression appeared responsive to the OA pathophysiology in cartilage.
Conclusions We have identified genetic loci affecting either uCTX-II or sCOMP levels. The genome wide significant association of MRC1 with sCOMP levels was found likely to act independent of OA subtypes. Increased sensitivity of biomarkers with OA may be accomplished by taking genetic variation into account.
Osteoarthritis (OA) is a prevalent, complex, disabling disease affecting articular joints. Age and unfavourable metabolic states, for example high body mass index (BMI), are strong risk factors of OA, however, development of OA is also determined by a considerable genetic component of a polygenic nature. Commonly, OA is diagnosed clinically and radiographically at a stage in which joint damage is already significant, and non-surgical therapies to cure or halt progress of OA are not yet available. Research directed to identify clinical biochemical markers to sensitively monitor OA disease activity in an individual over time or to assess quantitative joint tissue remodelling is currently in progress. So far, these studies have suggested that serum cartilage oligomeric protein (sCOMP), and urinary C-telopeptide of type II collagen (uCTX-II) are promising candidates and they are listed among the biomarkers that were recommended to focus on in future research efforts.
Initially, sCOMP and uCTX-II were considered as markers for cartilage degradation: sCOMP would be representative of cartilage turnover, and uCTX-II is a collagen type II cleavage product representative of hyaline (articular) cartilage degradation. However, some reports indicate that uCTX-II may not merely be a cartilage specific marker but rather a reflection of joint tissue remodelling including calcified cartilage and subchondral bone. Moreover, Meulenbelt et al showed that in patients with generalised OA uCTX-II is a sensitive marker of whole body cartilage degeneration as reflected by radiographic signs of OA but association with a specific joint site could not be demonstrated. This was recently confirmed in a large-scale meta-analyses showing that uCTX-II associated with risk for hand, knee and hip OA. In addition, the analyses showed that sCOMP as well as uCTX-II can be useful to monitor OA disease activity. However, the study also highlighted that the use of sCOMP and uCTX-II as biomarkers in the clinic still remains largely unsatisfactory.
Importantly, investigations have shown that in subjects with familial OA at multiple joint sites besides clustering of sCOMP together with hand OA and with age, sCOMP has a high familial aggregation. At present, little is known about genetic variants affecting basal levels of sCOMP and uCTX-II. Identification of genes that contribute to these levels may provide mechanistic insight into factors that influence cartilage tissue turnover, dependent or independent of the OA pathophysiology, which in the latter case would thus perturb the use of these molecules as biochemical markers of OA. The objective of the current work is, therefore, to identify genetic variants involved in the variation in basal levels of sCOMP and uCTX-II by means of a large-scale hypothesis-free meta-analysis of genome-wide association studies (GWAS) across three English and four Dutch Caucasian studies. Additional replication occurred in the Dutch Cohort Hip and Cohort Knee (CHECK) study and signals were further tested for association with hip and knee OA. Finally, given that functional follow-up studies of susceptibility loci have demonstrated that they frequently result in expression differences of positional genes within the relevant diseased tissue, we explored the behaviour of mRNA expression of the annotated genes in preserved and osteoarthritic articular cartilage.
Abstract and Introduction
Abstract
Background Research for the use of biomarkers in osteoarthritis (OA) is promising, however, adequate discrimination between patients and controls may be hampered due to innate differences. We set out to identify loci influencing levels of serum cartilage oligomeric protein (sCOMP) and urinary C-telopeptide of type II collagen (uCTX-II).
Methods Meta-analysis of genome-wide association studies was applied to standardised residuals of sCOMP (N=3316) and uCTX-II (N=4654) levels available in 6 and 7 studies, respectively, from TreatOA. Effects were estimated using a fixed-effects model. Six promising signals were followed up by de novo genotyping in the Cohort Hip and Cohort Knee study (N=964). Subsequently, their role in OA susceptibility was investigated in large-scale genome-wide association studies meta-analyses for OA. Differential expression of annotated genes was assessed in cartilage.
Results Genome-wide significant association with sCOMP levels was found for a SNP within MRC1 (rs691461, p=1.7Ă—10) and a SNP within CSMD1 associated with variation in uCTX-II levels with borderline genome-wide significance (rs1983474, p=8.5Ă—10). Indication for association with sCOMP levels was also found for a locus close to the COMP gene itself (rs10038, p=7.1Ă—10). The latter SNP was subsequently found to be associated with hip OA whereas COMP expression appeared responsive to the OA pathophysiology in cartilage.
Conclusions We have identified genetic loci affecting either uCTX-II or sCOMP levels. The genome wide significant association of MRC1 with sCOMP levels was found likely to act independent of OA subtypes. Increased sensitivity of biomarkers with OA may be accomplished by taking genetic variation into account.
Introduction
Osteoarthritis (OA) is a prevalent, complex, disabling disease affecting articular joints. Age and unfavourable metabolic states, for example high body mass index (BMI), are strong risk factors of OA, however, development of OA is also determined by a considerable genetic component of a polygenic nature. Commonly, OA is diagnosed clinically and radiographically at a stage in which joint damage is already significant, and non-surgical therapies to cure or halt progress of OA are not yet available. Research directed to identify clinical biochemical markers to sensitively monitor OA disease activity in an individual over time or to assess quantitative joint tissue remodelling is currently in progress. So far, these studies have suggested that serum cartilage oligomeric protein (sCOMP), and urinary C-telopeptide of type II collagen (uCTX-II) are promising candidates and they are listed among the biomarkers that were recommended to focus on in future research efforts.
Initially, sCOMP and uCTX-II were considered as markers for cartilage degradation: sCOMP would be representative of cartilage turnover, and uCTX-II is a collagen type II cleavage product representative of hyaline (articular) cartilage degradation. However, some reports indicate that uCTX-II may not merely be a cartilage specific marker but rather a reflection of joint tissue remodelling including calcified cartilage and subchondral bone. Moreover, Meulenbelt et al showed that in patients with generalised OA uCTX-II is a sensitive marker of whole body cartilage degeneration as reflected by radiographic signs of OA but association with a specific joint site could not be demonstrated. This was recently confirmed in a large-scale meta-analyses showing that uCTX-II associated with risk for hand, knee and hip OA. In addition, the analyses showed that sCOMP as well as uCTX-II can be useful to monitor OA disease activity. However, the study also highlighted that the use of sCOMP and uCTX-II as biomarkers in the clinic still remains largely unsatisfactory.
Importantly, investigations have shown that in subjects with familial OA at multiple joint sites besides clustering of sCOMP together with hand OA and with age, sCOMP has a high familial aggregation. At present, little is known about genetic variants affecting basal levels of sCOMP and uCTX-II. Identification of genes that contribute to these levels may provide mechanistic insight into factors that influence cartilage tissue turnover, dependent or independent of the OA pathophysiology, which in the latter case would thus perturb the use of these molecules as biochemical markers of OA. The objective of the current work is, therefore, to identify genetic variants involved in the variation in basal levels of sCOMP and uCTX-II by means of a large-scale hypothesis-free meta-analysis of genome-wide association studies (GWAS) across three English and four Dutch Caucasian studies. Additional replication occurred in the Dutch Cohort Hip and Cohort Knee (CHECK) study and signals were further tested for association with hip and knee OA. Finally, given that functional follow-up studies of susceptibility loci have demonstrated that they frequently result in expression differences of positional genes within the relevant diseased tissue, we explored the behaviour of mRNA expression of the annotated genes in preserved and osteoarthritic articular cartilage.
Source...