Osteoarthritis – a case for personalized health care?

Open ArchivePublished:December 19, 2013DOI:https://doi.org/10.1016/j.joca.2013.10.018

      Summary

      For both economic and ethical reasons, identification of the optimal treatment for each individual patient is a pressing concern, not only for the patients and their physician, but also health care payers and the pharmaceutical industry. In the field of osteoarthritis (OA) this is of particular relevance, due to the heterogeneity of the disease and the very large number of affected individuals. There is a need to pair the right patients with the right therapeutic modes of action. At present, the clinical trial failures in OA may be a consequence of both bona fide treatment failures and trial failures due to clinical design deficiencies. Tools are needed for characterization and segregation of patients with OA. Key lessons may be learned from advances with another form of arthritis, namely rheumatoid arthritis (RA).
      Personalized health care (PHC) may be more advantageous for a number of specific indications which are characterized by costly therapy, low response rates and significant problems associated with trial and error prescription, including the risk of serious side effects. We discuss the use of diagnostic practices guiding RA treatment, which may serve as a source of key insights for diagnostic practices in OA. We discuss the emerging concept of PHC, and outline the opportunities and current successes and failures across the RA field, as the OA field collects further data to support the hypothesis. We attempt to outline a possible path forward to assist patients, physicians, payers and the pharmaceutical industry in assuring the ‘right’ patients are treated with the ‘right drug’ in OA. Finally we highlight methods for possible segregation of OA patients that would allow identification of patient subtypes, such as OA driven by inflammation that may be ideally suited for PHC and for targeted therapies.

      Keywords

      Introduction

      One of the most pressing needs in the health field is to treat the right patients with the right medicine. However, in light of the current complete lack of structure modifying treatments for osteoarthritis (OA), the even more pressing issue is to identify the optimal patient population in which to test a given treatment. For patients with OA, non-surgical treatment consists notably of physical therapy, lifestyle modification, supportive devices, medicinal pain relief, oral and injection drugs (e.g., non-steroidal anti-inflammatory drugs). Once these treatments fail, surgery, such as joint replacement, is the main option for the patients. Presently, structure modification trials are in part hindered by a combination of slow and unpredictable disease progression, quantified by relatively insensitive detection tools such as X-ray criteria. To enable an optimal development program and later ensure that the right patients are treated with the optimal intervention there is a need of a method to segregate patients with different OA subtypes in order to pair them with an optimal mode of action. This may help redeem the OA clinical trial area, and lessen the chances of false negative conclusions.

      Which diseases are optimal candidates for personalized health care (PHC) and what drives the need?

      PHC has often been suggested as a solution
      • Karsdal M.A.
      • Henriksen K.
      • Leeming D.J.
      • Mitchell P.
      • Duffin K.
      • Barascuk N.
      • et al.
      Biochemical markers and the FDA critical path: how biomarkers may contribute to the understanding of pathophysiology and provide unique and necessary tools for drug development.
      to this complexity, but the majority of examples of successful PHC are currently drawn from oncology and, as such, reflect a relatively simplistic concept of therapeutic stratification based on the targeting of constitutively activated pathways in an oligogenic model of disease. In contrast, reportable successes for PHC in the field of polygenic disease are few. This poses the question of whether therapeutic areas and specific diseases outside the field of oncology can ultimately benefit from a tailored approach to therapy. It seems likely that certain therapeutic areas may ultimately prove more amenable to the application of PHC than others; these would be characterized by highly heterogeneous patient populations, low response rates and significant disadvantages stemming from stochastic ‘trial and error’ prescribing that increases risk of adverse events and economic burden due to exploratory prescriptions of high cost therapies.
      Contrasting examples of the utility of a PHC paradigm are provided by osteoporosis and non-small-cell lung carcinoma (NSCLC). In osteoporosis, the need for stratification of patients is limited due to the relatively benign side effect profile and high response rates of established therapies
      • Abrahamsen B.
      • Eiken P.
      • Eastell R.
      Cumulative alendronate dose and the long-term absolute risk of subtrochanteric and diaphyseal femur fractures: a register-based national cohort analysis.
      • Cooper C.
      • Emkey R.D.
      • McDonald R.H.
      • Hawker G.
      • Bianchi G.
      • Wilson K.
      • et al.
      Efficacy and safety of oral weekly ibandronate in the treatment of postmenopausal osteoporosis.
      • Cummings S.R.
      • San M.J.
      • McClung M.R.
      • Siris E.S.
      • Eastell R.
      • Reid I.R.
      • et al.
      Denosumab for prevention of fractures in postmenopausal women with osteoporosis.
      , and the limited complexity of the disease involving mainly one tissue, bone. Despite early reports linking the commonly prescribed bisphosphonate therapies with stratified responses in terms of bone mineralization
      • Ravn P.
      • Hosking D.
      • Thompson D.
      • Cizza G.
      • Wasnich R.D.
      • McClung M.
      • et al.
      Monitoring of alendronate treatment and prediction of effect on bone mass by biochemical markers in the early postmenopausal intervention cohort study.
      , novel clinical diagnostics have neither been extensively sought nor established, and therapeutic practice remains solely based on prognostic investigation. In contrast, following the initial characterization of a number of mutations that drive and maintain tumorgenesis
      • Pao W.
      • Hutchinson K.E.
      Chipping away at the lung cancer genome.
      , diagnostic practice for NSCLC has rapidly extended beyond classical pathology; physicians now have an opportunity to target highly effective therapies against signaling pathways activated by these mutations.
      For the current discussion, we identified four major drivers of PHC.
      • Identification of patients that are in greatest need of treatment.
      • Identification of the patients whom may respond optimally, with the highest efficacy and lowest safety concerns, to a given treatment
      • Development strategy for a selected subpopulation of patients.
      • Efficient use of health care resources.
      Although PHC might seem to be a highly attractive ‘magic bullet’ for patients, physicians, and payers, the interaction of several key considerations dictates the ultimate level of attractiveness of a PHC strategy. On the one hand, it is readily appreciated that the non-targeted treatment of a particular group of patients may not provide a cost-benefit-risk assessment that is adequately attractive [Fig. 1(A) ]. In addition, it is equally readily apparent that restricting treatment to a diagnostically selected subpopulation with a high likelihood of a response may significantly and drastically improve this assessment [Fig. 1(B)], whereas another selected treatment more appropriate targets other patients [Fig. 1(C)], in contract to the worst-case scenario in which only non-responders and Serious adverse events (SAEs) are the result of treatment.
      Figure thumbnail gr1
      Fig. 1A schematic representation of cost-benefit assessment scenarios for different treatments and different patient populations. (A) A treatment for a total group of patients that does not provide an attractive cost-benefit assessment. (B) A subpopulation (1 & 2) treated with a targeted therapy may significantly improve the cost-benefit assessment. (C) An alternative therapeutic intervention (different from B) may be more suitable for a separate and distinct patient population (4 & 5). (D) Treatment of the wrong patents with an unsuitable therapy results in adverse events and no efficacy.
      Given the central precept that treatment of non-responders with a therapy associated with considerable side effects and costs will only provide risk without benefit to patients and waste scarce health care resources, it would seem intuitive that treatments delivering low response rates and a high potential for side effects would have the most to gain from PHC. Identification of non-responders will be of particular concern if the adverse effects are considered irreversible. Ethical and economic considerations clearly drive the earlier identification of such non-responders in order to allow for selection of an alternative intervention.

      The heterogeneity of the patient population – lessons learned from rheumatoid arthritis (RA)

      Experience with PHC in the context of RA provides an instructive example of the potential possibilities for OA. RA is a chronic autoimmune disease characterized by poly-articular inflammation resulting in massive tissue destruction in the affected synovial joints. Due to fast progression of joint structural damage that may lead to complete disability in less than 10 years, some patients need to be treated aggressively
      • Syversen S.W.
      • Goll G.L.
      • van der H.D.
      • Landewe R.
      • Lie B.A.
      • Odegard S.
      • et al.
      Prediction of radiographic progression in rheumatoid arthritis and the role of antibodies against mutated citrullinated vimentin: results from a 10-year prospective study.
      , and precise guidelines from European league against rheumatology (EULAR) are provided for patient care for both synthetic and biological interventions
      • Bijlsma J.W.
      Optimal treatment of rheumatoid arthritis: EULAR recommendations for clinical practice.
      • Knevel R.
      • Schoels M.
      • Huizinga T.W.
      • Aletaha D.
      • Burmester G.R.
      • Combe B.
      • et al.
      Current evidence for a strategic approach to the management of rheumatoid arthritis with disease-modifying antirheumatic drugs: a systematic literature review informing the EULAR recommendations for the management of rheumatoid arthritis.
      • Nam J.L.
      • Winthrop K.L.
      • van Vollenhoven R.F.
      • Pavelka K.
      • Valesini G.
      • Hensor E.M.
      • et al.
      Current evidence for the management of rheumatoid arthritis with biological disease-modifying antirheumatic drugs: a systematic literature review informing the EULAR recommendations for the management of RA.
      • Gaujoux-Viala C.
      • Smolen J.S.
      • Landewe R.
      • Dougados M.
      • Kvien T.K.
      • Mola E.M.
      • et al.
      Current evidence for the management of rheumatoid arthritis with synthetic disease-modifying antirheumatic drugs: a systematic literature review informing the EULAR recommendations for the management of rheumatoid arthritis.
      . Methotrexate therapy (MTX), often in combination with corticosteroids, is the most commonly used Disease Modifying Anti-Rheumatic Drug (DMARD) regimen in RA
      • Upchurch K.S.
      • Kay J.
      Evolution of treatment for rheumatoid arthritis.
      • Kay J.
      • Upchurch K.S.
      ACR/EULAR 2010 rheumatoid arthritis classification criteria.
      . Upon inadequate response or intolerance to therapy, initiation of a biologic therapy is recommended
      • Reynolds A.
      • Koenig A.S.
      • Bananis E.
      • Singh A.
      When is switching warranted among biologic therapies in rheumatoid arthritis?.
      . There are a growing number of approved biologic therapies for RA. Given such a battery of diverse therapeutic options available to rheumatologists, it is not surprising that clinical remission of RA, or at least low disease activity, is often the goal of therapy
      • Smolen J.S.
      • Aletaha D.
      Forget personalised medicine and focus on abating disease activity.
      • Smolen J.S.
      • Aletaha D.
      • Bijlsma J.W.
      • Breedveld F.C.
      • Boumpas D.
      • Burmester G.
      • et al.
      Treating rheumatoid arthritis to target: recommendations of an international task force.
      . However, the absolute performance of biological therapies in randomized controlled studies in patients who have failed initial DMARD therapy is very similar based on the standardized criteria established by the American College of Rheumatology (ACR) for 20% (ACR20), 50% (ACR50) and 70% (ACR75) improvement in disease activity: namely 50–65% for ACR20, 20–45% for ACR50 and 10–25% for ACR70
      • Verweij C.L.
      Pharmacogenetics: anti-TNF therapy in RA – towards personalized medicine?.
      • Bergman G.J.
      • Hochberg M.C.
      • Boers M.
      • Wintfeld N.
      • Kielhorn A.
      • Jansen J.P.
      Indirect comparison of tocilizumab and other biologic agents in patients with rheumatoid arthritis and inadequate response to disease-modifying antirheumatic drugs.
      • Devine E.B.
      • fonso-Cristancho R.
      • Sullivan S.D.
      Effectiveness of biologic therapies for rheumatoid arthritis: an indirect comparisons approach.
      . These data reveal a heterogeneity of drug responses and point to a significant and remaining unmet need for alternative therapies for the substantial number of patients who fall short of attaining remission or low disease activity from current therapy
      • Smolen J.S.
      • Aletaha D.
      Forget personalised medicine and focus on abating disease activity.
      . Given the exposure of these inadequately treated patients to the risk of adverse events, together with a considerable societal economic burden of unremitting disease and cost of therapy, a PHC approach to the treatment of RA is pursued aggressively. The success will require that the current seemingly random stochastic approach to RA therapy needs to be replaced by a mechanistic and predictive understanding of response.
      The heterogeneity of the patient population is heavily debated in the OA; failures in clinical trials, have made it increasingly more evident that OA is not one disease but has different phenotypes
      • Bijlsma J.W.
      • Berenbaum F.
      • Lafeber F.P.
      Osteoarthritis: an update with relevance for clinical practice.
      . Although the OA field is still without efficacious treatments to validate this hypothesis, several subtypes have been identified that may warrant different treatment strategies. Those subtypes receiving the most attention currently include the following: metabolic OA (including obesity)
      • Bay-Jensen A.C.
      • Slagboom E.
      • Chen-An P.
      • Alexandersen P.
      • Qvist P.
      • Christiansen C.
      • et al.
      Role of hormones in cartilage and joint metabolism: understanding an unhealthy metabolic phenotype in osteoarthritis.
      • Bos S.D.
      • Beekman M.
      • Maier A.B.
      • Karsdal M.A.
      • Kwok W.Y.
      • Bay-Jensen A.C.
      • et al.
      Metabolic health in families enriched for longevity is associated with low prevalence of hand osteoarthritis and influences OA biomarker profiles.
      , traumatic OA (including joint malalignment)
      • Karsdal M.A.
      • Madsen S.H.
      • Christiansen C.
      • Henriksen K.
      • Fosang A.J.
      • Sondergaard B.C.
      Cartilage degradation is fully reversible in the presence of aggrecanase but not matrix metalloproteinase activity.
      , inflammation driven OA
      • Attur M.
      • Belitskaya-Levy I.
      • Oh C.
      • Krasnokutsky S.
      • Greenberg J.
      • Samuels J.
      • et al.
      Increased interleukin-1beta gene expression in peripheral blood leukocytes is associated with increased pain and predicts risk for progression of symptomatic knee osteoarthritis.
      and subchondral bone turnover driven OA phenotypes
      • Karsdal M.A.
      • Leeming D.J.
      • Dam E.B.
      • Henriksen K.
      • Alexandersen P.
      • Pastoureau P.
      • et al.
      Should subchondral bone turnover be targeted when treating osteoarthritis?.
      . This may be illustrated schematically as shown in Fig. 2. If a particular therapy works in a subset of patients with a specific disease phenotype, the response rates will be low in the absence of patient selection. Consequently, targeting a particular disease phenotype with the etiologically appropriate therapy would be expected to result in far higher response rates.
      Figure thumbnail gr2
      Fig. 2Schematic overview to illustrate that different subtypes of patient exist that to some extent overlap, but only though targeting of the right subpopulation the most optimal cost-benefit ratio may be achieved. (A) RA (and to an even larger extent OA) may consist of many different molecular and clinical phenotypes that remain to be identified and characterized. It is considered likely that each particular phenotype will respond differently to therapies with differing modes of action. Treatment is not currently targeted by phenotype with the consequence of an unmet medical need for a greater proportion of therapeutic responses. (B) With patient selection through PHC, a given phenotype may be selected with a greater potential to respond to a specific therapy.

      PHC in RA: current understanding of possible technological advances

      Presently there are few positive examples of PHC for RA. Seropositivity, the presence of autoantibodies to rheumatoid factor (RF) or anticyclic citrullinated peptide (anti-CCP) observed in around 80% of patients, is an example of one type of biomarker with a modest ability to predict response to Rituximab
      • Isaacs J.D.
      • Cohen S.B.
      • Emery P.
      • Tak P.P.
      • Wang J.
      • Lei G.
      • et al.
      Effect of baseline rheumatoid factor and anticitrullinated peptide antibody serotype on rituximab clinical response: a meta-analysis.
      . Another approach to the identification of subgroups of therapeutic responders has been genetics based. RA has a strong, but complex genetic etiology illustrated by the identification of 31 risk loci in seropositive RA, a very strong genetic risk factor HLA-DRB1, and the group of alleles referred to as the shared epitope (SE). To date, candidate and genome wide approaches to the discovery of response genes have been taken for anti-Tumor necrosis factor (TNF)
      • Verweij C.L.
      Pharmacogenetics: anti-TNF therapy in RA – towards personalized medicine?.
      and tocilizumab interventions. While associations have been established, the predictive capability has so far been insufficient to have meaningful clinical application. Wang et al. reported that polymorphisms identified in a genome wide association scan (GWAS) each accounted for less than 2% of the variance observed with change in DAS28 response to tocilizumab. Thirdly, the production of antibodies against biologic therapies, such as to anti-TNF antagonists, has also been associated with waning of drug response. Lastly, a multi-marker panel approach using 12 different markers is being pursued by Crescendo
      • van der Helm-van Mil A.H.
      • Knevel R.
      • Cavet G.
      • Huizinga T.W.
      • Haney D.J.
      An evaluation of molecular and clinical remission in rheumatoid arthritis by assessing radiographic progression.
      • Li W.
      • Sasso E.H.
      • Emerling D.
      • Cavet G.
      • Ford K.
      Impact of a multi-biomarker disease activity test on rheumatoid arthritis treatment decisions and therapy use.
      • Eastman P.S.
      • Manning W.C.
      • Qureshi F.
      • Haney D.
      • Cavet G.
      • Alexander C.
      • et al.
      Characterization of a multiplex, 12-biomarker test for rheumatoid arthritis.
      . While this may be predictive at the group levels additional tailoring of the panel of markers and algorithm may be needed to provide a meaningful result in the clinical practice at the individual patient level. Models predicting low disease activity (DAS28 < 3.2) at 52 weeks, utilizing clinical data collected at 12 weeks, have been produced for etanercept, etanercept with MTX, and certolizumab pegol
      • Keystone E.C.
      • Combe B.
      • Smolen J.
      • Strand V.
      • Goel N.
      • van V.R.
      • et al.
      Sustained efficacy of certolizumab pegol added to methotrexate in the treatment of rheumatoid arthritis: 2-year results from the RAPID 1 trial.
      . These models require validation in an independent dataset and analysis against a non-anti-TNF therapy to determine if these models are predictive of response to a specific therapy
      • Curtis J.R.
      • Yang S.
      • Chen L.
      • Park G.S.
      • Bitman B.
      • Wang B.
      • et al.
      Predicting low disease activity and remission using early treatment response to antitumour necrosis factor therapy in patients with rheumatoid arthritis: exploratory analyses from the TEMPO trial.
      . Even for RA it is too optimistic to suggest that the optimal marker combination and final selection of the right patient is present, exemplified by the discussion in the RA field “Forget personalized medicine and focus on abating disease activity”
      • Smolen J.S.
      • Aletaha D.
      Forget personalised medicine and focus on abating disease activity.
      .
      For biomarkers to be utilized in PHC they need to be readily measurable with an acceptable level of sensitivity and specificity, accessible and reportable to physicians within a clinically actionable timeframe. As suggested above, current standard practice for RA at best provides an ACR50 response rate of 20–45% in a randomized controlled trial [Fig. 3(A) ]. By excluding patients unlikely to respond to the treatment the response rate may be increased and a subset of patients will not be exposed unnecessarily [Fig. 3(B)]. Thus an optimal strategy will both ensure a significant increase in response rates, but will also allow for exclusion of those patients likely to have a serious treatment related side effect.
      Figure thumbnail gr3
      Fig. 3(A) Treatment of the entire patient population, without preselection of patients, resulting in 30% response rates. (B) Preselecting 50% of patients may result in a 60% response rate, thus drastically improving the cost-benefit and benefit-risk assessments. Importantly, PHC is of benefit for not only identifying those patients who will respond safely to treatment, but also those patients who should not be exposed to the drug. Thus, an optimal strategy will both ensure an enrichment of response rates, and lower adverse event rates through exclusion of patients from the treatment pool for whom the therapy poses a serious risk. Importantly, SAEs can and may also occur in selected patients, as illustrated by the “half-man” being a non-responder as well as subject to an SAE.
      A range of technologies may assist the implementation of PHC. While such technologies must be technically robust, relatively inexpensive, and easily accessible and simple to use, they are not limited to the current generation of non-invasive biochemical markers. A type of molecular marker has been developed that can quantify the levels of tissue destruction associated with a progressing pathology
      • Karsdal M.A.
      • Woodworth T.
      • Henriksen K.
      • Maksymowych W.P.
      • Genant H.
      • Vergnaud P.
      • et al.
      Biochemical markers of ongoing joint damage in rheumatoid arthritis – current and future applications, limitations and opportunities.
      • Lotz M.
      • Martel-Pelletier J.
      • Christiansen C.
      • Brandi M.L.
      • Bruyere O.
      • Chapurlat R.
      • et al.
      Value of biomarkers in osteoarthritis: current status and perspectives.
      . In contrast to total protein measurements of established inflammatory biomarkers, such as interleukin (IL)-1, IL-6 and C-reactive protein (CRP), these new biomarkers rely on the measurement of specific, circulating tissue protein fragments generated by up-regulated, active proteolytic enzymes. As an example, the major extracellular matrix (ECM) proteins of connective tissue in joints are type I, II and III collagen. Matrix metallo protease (MMP)-mediated degradation of these collagens results in the generation of the specific biomarkers, e.g., C1M, C2M and C3M respectively
      • Bay-Jensen A.C.
      • Liu Q.
      • Byrjalsen I.
      • Li Y.
      • Wang J.
      • Pedersen C.
      • et al.
      Enzyme-linked immunosorbent assay (ELISAs) for metalloproteinase derived type II collagen neoepitope, CIIM – increased serum CIIM in subjects with severe radiographic osteoarthritis.
      • Barascuk N.
      • Veidal S.S.
      • Larsen L.
      • Larsen D.V.
      • Larsen M.R.
      • Wang J.
      • et al.
      A novel assay for extracellular matrix remodeling associated with liver fibrosis: an enzyme-linked immunosorbent assay (ELISA) for a MMP-9 proteolytically revealed neo-epitope of type III collagen.
      . Additional protein fingerprints are generated by distinct enzymatic processing of collagen or other ECM proteins represented by the e.g., products like CTX-I, ICTP, CTX-II, C2C, ARGS (aggrecan), FFGV (aggrecan), C4M (type IV collagen), C5M (type V collagen) and C6M (type VI collagen)
      • Karsdal M.A.
      • Henriksen K.
      • Leeming D.J.
      • Mitchell P.
      • Duffin K.
      • Barascuk N.
      • et al.
      Biochemical markers and the FDA critical path: how biomarkers may contribute to the understanding of pathophysiology and provide unique and necessary tools for drug development.
      • Karsdal M.A.
      • Nielsen M.J.
      • Sand J.M.
      • Henriksen K.
      • Genovese F.
      • Bay-Jensen A.C.
      • et al.
      Extracellular matrix remodeling: the common denominator in connective tissue diseases possibilities for evaluation and current understanding of the matrix as more than a passive architecture, but a key player in tissue failure.
      . The measurement of different degradation products of a protein may elucidate different and frequently opposing metabolic mechanisms. This concept is exemplified by e.g., products of metabolism of type I collagen like: PINP, CTX-I, C1M and ICTP – each of these products provides distinct and unique information. PINP is a pro-peptide released during protein synthesis and can be measured as a surrogate biomarker of both bone formation and fibrogenesis
      • Veidal S.S.
      • Vassiliadis E.
      • Bay-Jensen A.C.
      • Tougas G.
      • Vainer B.
      • Karsdal M.A.
      Procollagen type I N-terminal propeptide (PINP) is a marker for fibrogenesis in bile duct ligation-induced fibrosis in rats.
      • Wislowska M.
      • Jakubicz D.
      • Stepien K.
      • Cicha M.
      Serum concentrations of formation (PINP) and resorption (Ctx) bone turnover markers in rheumatoid arthritis.
      . CTX-I is a widely used biomarker for Cathepsin K mediated destruction of type I collagen, reflecting bone resorption
      • Rosenquist C.
      • Fledelius C.
      • Christgau S.
      • Pedersen B.J.
      • Bonde M.
      • Qvist P.
      • et al.
      Serum CrossLaps One Step ELISA. First application of monoclonal antibodies for measurement in serum of bone-related degradation products from C-terminal telopeptides of type I collagen.
      . ICTP is a triple cross-linked carboxy-terminal telopeptide of type I collagen generated by MMP-activity, but destroyed by the activity of other proteases
      • Garnero P.
      • Ferreras M.
      • Karsdal M.A.
      • NicAmhlaoibh R.
      • Risteli J.
      • Borel O.
      • et al.
      The type I collagen fragments ICTP and CTX reveal distinct enzymatic pathways of bone collagen degradation.
      . ICTP is mainly released from connective tissue turnover. Finally, C1M is also released by the action of MMP's from the helical domain of type I collagen. These different products of type I collagen provide different and complementary information on tissue integrity and turnover even though they in essence are measures of the same protein, albeit in different ways.

      What does this mean for the OA field?

      The list of failures and limited successes in clinical development for OA is growing, and includes among others, Inducible nitric oxide synthase (iNOS), strontium ranelate, Calcitonin, MMP inhibitors, Cathepsin K inhibitors and bisphosphonates
      • Karsdal M.A.
      • Byrjalsen I.
      • Henriksen K.
      • Riis B.J.
      • Lau E.M.
      • Arnold M.
      • et al.
      The effect of oral salmon calcitonin delivered with 5-CNAC on bone and cartilage degradation in osteoarthritic patients: a 14-day randomized study.
      • Spector T.D.
      • Conaghan P.G.
      • Buckland-Wright J.C.
      • Garnero P.
      • Cline G.A.
      • Beary J.F.
      • et al.
      Effect of risedronate on joint structure and symptoms of knee osteoarthritis: results of the BRISK randomized, controlled trial [ISRCTN01928173].
      • Alexandersen P.
      • Karsdal M.A.
      • Byrjalsen I.
      • Christiansen C.
      Strontium ranelate effect in postmenopausal women with different clinical levels of osteoarthritis.
      • Reginster J.Y.
      • Badurski J.
      • Bellamy N.
      • Bensen W.
      • Chapurlat R.
      • Chevalier X.
      • et al.
      Efficacy and safety of strontium ranelate in the treatment of knee osteoarthritis: results of a double-blind, randomised placebo-controlled trial.
      . These disappointing results may in part be due to hypothesis failures but also a result of applying these therapies to non-selected patient populations. Figure 4 offers a very preliminary but possibly clinically meaningful and clinically feasible strategy for selecting OA patients. OA might be divided into at least three different subtypes based on the most active joint tissue at a particular stage of disease (bone, cartilage and inflamed synovium) or the tissue with the predominant manifestations in particular patient populations. As examples, Treatment 1 may target early cartilage pathology such as manifested in traumatic OA, with high protease activities in the articular cartilage such as MMP and aggrecanase activities. Treatment 2 would target both bone and cartilage involvement, such as manifested in postmenopausal generalized OA. Treatment 3 may target late stage inflammation driven OA. Clearly these three different subtypes of OA would require different interventions and diagnostic algorithms, and one treatment that may be successful in one patient population may fail in another.
      Figure thumbnail gr4
      Fig. 4The drivers of OA may be divided into at least three different categories: bone, cartilage and inflamed synovium. These may represent different disease subgroups; alternatively they may represent the predominant tissue pathology during a particular stage of disease. Optimal therapy may be considered the ability to detect and target each of these stages or subgroups of disease.
      Based on even this simplistic subdivision of patients groups, the following treatment scenarios might be suggested.
      • Patient phenotype 1: Traumatic OA, in the early disease course most likely involving a high level of protease activity destroying the cartilage subsequent to cartilage injury. These patients might benefit from a protease inhibitor treatment.
      • Patient phenotype 2: Generalized OA, with high turnover of bone and progression of cartilage damage characterized by an intimate relationship between bone and cartilage in the pathogenesis of OA. These patients may benefit from an anti-resorptive treatment inhibiting bone resorption and subchondral bone turnover.
      • Patient phenotype 3: Episodic subacute and acute inflammatory events precipitating worsening in stages. Macrophages play a role in this process
        • Bondeson J.
        Activated synovial macrophages as targets for osteoarthritis drug therapy.
        and likely will yield pathways to inform specific anti-inflammatory therapies for this phenotype or stage of disease.
      This approach underscores the need to better understand the multiple pathways that result in the common end stage we refer to as OA. While these delineations may overlap to some extent and be dependent on the time and stage of disease, the scenarios represent testable hypotheses that call for further research to validate and refine algorithms for clinically meaningful patient phenotyping.

      Do different phenotypes in OA represent different pathological stages and/or different phenotypes?

      It is important to recognize that OA is far from a static disease, and represents very distinct characteristics during the various stages of disease progression, i.e., the above mentioned phenotypes may vary over time. The different phenotypes of OA
      • Bay-Jensen A.C.
      • Slagboom E.
      • Chen-An P.
      • Alexandersen P.
      • Qvist P.
      • Christiansen C.
      • et al.
      Role of hormones in cartilage and joint metabolism: understanding an unhealthy metabolic phenotype in osteoarthritis.
      most likely depend on many factors such as metabolic, traumatic, genetic and lifestyle dependent factors. Currently we may understand that OA may be triggered by a macro- or micro-traumatic injury to the joint, where the trauma can be an acute accident or daily microinjury respectively. Such events result in high turnover of the bone and cartilage compartments, which eventually leads to symptomatic OA and joint failure, as illustrated in Fig. 5.
      Figure thumbnail gr5
      Fig. 5The present figure illustrates key drivers of the disease and their speculated impact on the rate of disease progression. These distinct drivers of disease may be highly OA stage dependent and overlap to some extent. The length of the line is considered the relative importance, as such the line is longer for hormonal than autoimmunity driven OA disease, and consequently autoimmunity may lead to joint failure faster than hormonal regulations. All cases of OA may be caused by a minor or major joint trauma; however the rate of progression may be driven and accelerated by different factors. These factors may describe different OA phenotypes, and these factors may also be applied as tools for PHC allowing early segregation of patient as compared to traditional clinical diagnosis of symptomatic OA. Note that this figure provides a working hypothesis (and is hypothetical) as we do not know whether genetic factors are more important leading to faster progression (shorter length of the line) than e.g., mechano-transduction.
      It will be important to improve our ability to identify factors that identify the various phenotypes of OA; to the extent to which we can more precisely identify these phenotypes, the more capable we will be at developing personalized treatments. Currently several factors related to OA disease progression may be identified including the following (as outlined in Fig. 5), which needs further experimental evidence and independent validation: (1) Low grade autoimmunity possibly plays a role in a subset of patients in whom there is epitope spreading from the injured joint and activation of the innate immune system
      • Frisenda S.
      • Perricone C.
      • Valesini G.
      Cartilage as a target of autoimmunity: a thin layer.
      • Dolzani P.
      • Assirelli E.
      • Pulsatelli L.
      • Addimanda O.
      • Mancarella L.
      • Peri G.
      • et al.
      Systemic inflammation and antibodies to citrullinated peptides in hand osteoarthritis.
      • Nishioka K.
      Autoimmune response in cartilage-delivered peptides in a patient with osteoarthritis.
      ; (2) Inflammation, in which release of factors such as cytokines and matrikines act as pro-inflammatory factors
      • Frisenda S.
      • Perricone C.
      • Valesini G.
      Cartilage as a target of autoimmunity: a thin layer.
      • Cooke T.D.
      Significance of immune complex deposits in osteoarthritic cartilage.
      ; (3) Genetic, wherein specific Single nucleotide polymorphisms (SNPs) may predispose to fast progression through loss of function of key proteins
      • Reynard L.N.
      • Loughlin J.
      Genetics and epigenetics of osteoarthritis.
      ; (4) Hormonal, representing an imbalance in the endocrine system that may result in loss of skeletal protections, as observed in some postmenopausal women
      • Stevenson J.C.
      A woman's journey through the reproductive, transitional and postmenopausal periods of life: impact on cardiovascular and musculo-skeletal risk and the role of estrogen replacement.
      ; (5) Metabolic, in which an unhealthy phenotype, such as obesity, may drive OA through adiopokines
      • Bay-Jensen A.C.
      • Slagboom E.
      • Chen-An P.
      • Alexandersen P.
      • Qvist P.
      • Christiansen C.
      • et al.
      Role of hormones in cartilage and joint metabolism: understanding an unhealthy metabolic phenotype in osteoarthritis.
      ; and (6) Mechano-transduction, in which mechanical loading may drive a “wear and tear” effect. Importantly, some factors may predispose at individual stages to either slow or fast progression; these are discussed and outlined in Table I. Of key importance, the risk factors and analysis methods outlined in Table I, may optimally be used in combination rather than stand alone technologies, in which imaging [X-ray and magnetic resonance imaging (MRI)], standard demographic description, genetic and novel biomarkers together provide value to patients. Such combination of modalities, imaging, demographic and serological biomarkers has in a few clinical settings been shown to provide additional improvement in odds ratios for identification of i.e., progression
      • Dam E.B.
      • Loog M.
      • Christiansen C.
      • Byrjalsen I.
      • Folkesson J.
      • Nielsen M.
      • et al.
      Identification of progressors in osteoarthritis by combining biochemical and MRI-based markers.
      .
      Table IPotential biomarkers for clinical and molecular phenotyping of patients for PHC in OA
      PhenotypeQuantifiable measure in synovial fluid, blood or urineReference
      Low grade autoimmunityIncrease in specific autoantibodies [e.g., anti-CCP, anti-modified citrullinated vimentin (MCV) antibodies, antibodies against citrullinated type I and II collagen fragments]

      Increase in autoantigens [e.g., citrullinated and MMP degraded vimentin (VICM), citrullinated fibrinogen, Fibulin-4]
      • Dolzani P.
      • Assirelli E.
      • Pulsatelli L.
      • Addimanda O.
      • Mancarella L.
      • Peri G.
      • et al.
      Systemic inflammation and antibodies to citrullinated peptides in hand osteoarthritis.
      • Bay-Jensen A.C.
      • Karsdal M.A.
      • Vassiliadis E.
      • Wichuk S.
      • Marcher-Mikkelsen K.
      • Lories R.
      • et al.
      Circulating citrullinated vimentin fragments reflect disease burden in ankylosing spondylitis and have prognostic capacity for radiographic progression.
      • Koivula M.K.
      • Aman S.
      • Karjalainen A.
      • Hakala M.
      • Risteli J.
      Are there autoantibodies reacting against citrullinated peptides derived from type I and type II collagens in patients with rheumatoid arthritis?.
      • Yoshida M.
      • Tsuji M.
      • Kurosaka D.
      • Kurosaka D.
      • Yasuda J.
      • Ito Y.
      • et al.
      Autoimmunity to citrullinated type II collagen in rheumatoid arthritis.
      • Xiang Y.
      • Sekine T.
      • Nakamura H.
      • Imajoh-Ohmi S.
      • Fukuda H.
      • Yudoh K.
      • et al.
      Fibulin-4 is a target of autoimmunity predominantly in patients with osteoarthritis.
      • Wildbaum G.
      • Nahir M.A.
      • Karin N.
      Beneficial autoimmunity to proinflammatory mediators restrains the consequences of self-destructive immunity.
      • Clague R.B.
      • Morgan K.
      • Collins I.
      • Pattrick M.
      • Doherty M.
      Absence of autoimmunity to type II collagen in generalised nodal osteoarthritis.
      • Bari A.S.
      • Carter S.D.
      • Bell S.C.
      • Morgan K.
      • Bennett D.
      Anti-type II collagen antibody in naturally occurring canine joint diseases.
      Auto-inflammationLocal and systemic increase in matrikines from cartilage degradation and synovial turnover [e.g., Hyaluron (HA), fibronectin fragments]
      • Yan X.
      • Yin L.
      • Wang Y.
      • Zhao Y.
      • Chang X.
      The low binding affinity of ADAMTS4 for citrullinated fibronectin may contribute to the destruction of joint cartilage in rheumatoid arthritis.
      • Yasuda T.
      Comparison of hyaluronan effects among normal, osteoarthritis, and rheumatoid arthritis cartilages stimulated with fibronectin fragment.
      • Bastow E.R.
      • Byers S.
      • Golub S.B.
      • Clarkin C.E.
      • Pitsillides A.A.
      • Fosang A.J.
      Hyaluronan synthesis and degradation in cartilage and bone.
      • Barilla M.L.
      • Carsons S.E.
      Fibronectin fragments and their role in inflammatory arthritis.
      • Sokolove J.
      • Zhao X.
      • Chandra P.E.
      • Robinson W.H.
      Immune complexes containing citrullinated fibrinogen costimulate macrophages via Toll-like receptor 4 and Fcgamma receptor.
      • Goldberg V.M.
      • Kresina T.F.
      Immunology of articular cartilage.
      • de J.H.
      • Berlo S.E.
      • Hombrink P.
      • Otten H.G.
      • van E.W.
      • Lafeber F.P.
      • et al.
      Cartilage proteoglycan aggrecan epitopes induce proinflammatory autoreactive T-cell responses in rheumatoid arthritis and osteoarthritis.
      • Girardelli M.
      • Bianco A.M.
      • Marcuzzi A.
      • Crovella S.
      A comparative analysis of serologic parameters and oxidative stress in osteoarthritis and rheumatoid arthritis: reply to Mishra and colleagues.
      • Deberg M.
      • Dubuc J.E.
      • Labasse A.
      • Sanchez C.
      • Quettier E.
      • Bosseloir A.
      • et al.
      One-year follow-up of Coll2-1, Coll2-1NO2 and myeloperoxydase serum levels in osteoarthritis patients after hip or knee replacement.
      • Siebuhr A.S.
      • Wang J.
      • Karsdal M.
      • Bay-Jensen A.C.
      • Y J.
      • Q Z.
      Matrix metalloproteinase-dependent turnover of cartilage, synovial membrane, and connective tissue is elevated in rats with collagen induced arthritis.
      • Catterall J.B.
      • Barr D.
      • Bolognesi M.
      • Zura R.D.
      • Kraus V.B.
      Post-translational aging of proteins in osteoarthritic cartilage and synovial fluid as measured by isomerized aspartate.
      GeneticSNPs in genes associated with joint tissue protection (e.g., GDF5, IL-1, DIO2, IL1RN, WNT pathway)

      SNPs and dysregulation of bone specific proteins (e.g., RANK, RANK-L, Sclerostin, Cathepsin K, V-ATPase, CLC-7, MMP-9 and -2)
      • Dodd A.W.
      • Syddall C.M.
      • Loughlin J.
      A rare variant in the osteoarthritis-associated locus GDF5 is functional and reveals a site that can be manipulated to modulate GDF5 expression.
      • Henriksen K.
      • Karsdal M.A.
      • John M.T.
      Osteoclast-derived coupling factors in bone remodeling.
      • Henriksen K.
      • Neutzsky-Wulff A.V.
      • Bonewald L.F.
      • Karsdal M.A.
      Local communication on and within bone controls bone remodeling.
      • Kaarvatn M.H.
      • Jotanovic Z.
      • Mihelic R.
      • Etokebe G.E.
      • Mulac-Jericevic B.
      • Tijanic T.
      • et al.
      Associations of the interleukin-1 gene locus polymorphisms with risk to hip and knee osteoarthritis: gender and subpopulation differences.
      • Meulenbelt I.
      • Min J.L.
      • Bos S.
      • Riyazi N.
      • Houwing-Duistermaat J.J.
      • van der Wijk H.J.
      • et al.
      Identification of DIO2 as a new susceptibility locus for symptomatic osteoarthritis 1165.
      • Valdes A.M.
      • Spector T.D.
      Genetic epidemiology of hip and knee osteoarthritis.
      HormonalEstrogen loss (e.g., E2, FSH, Progesterone)

      Increase in thyroid hormone (e.g., PTH)
      • Bay-Jensen A.C.
      • Slagboom E.
      • Chen-An P.
      • Alexandersen P.
      • Qvist P.
      • Christiansen C.
      • et al.
      Role of hormones in cartilage and joint metabolism: understanding an unhealthy metabolic phenotype in osteoarthritis.
      • Bay-Jensen A.C.
      • Tabassi N.C.
      • Sondergaard L.V.
      • Andersen T.L.
      • Dagnaes-Hansen F.
      • Garnero P.
      • et al.
      The response to oestrogen deprivation of the cartilage collagen degradation marker, CTX-II, is unique compared with other markers of collagen turnover.
      MetabolicIncrease in systemic levels of adipokines (e.g., leptin, resistin, adiponectin, visfatin)
      • Berry P.A.
      • Jones S.W.
      • Cicuttini F.M.
      • Wluka A.E.
      • Maciewicz R.A.
      Temporal relationship between serum adipokines, biomarkers of bone and cartilage turnover, and cartilage volume loss in a population with clinical knee osteoarthritis.
      • Gomez R.
      • Lago F.
      • Gomez-Reino J.
      • Dieguez C.
      • Gualillo O.
      Adipokines in the skeleton: influence on cartilage function and joint degenerative diseases.
      • Lajeunesse D.
      • Pelletier J.P.
      • Martel-Pelletier J.
      Osteoarthritis: a metabolic disease induced by local abnormal leptin activity?.
      • Presle N.
      • Pottie P.
      • Dumond H.
      • Guillaume C.
      • Lapicque F.
      • Pallu S.
      • et al.
      Differential distribution of adipokines between serum and synovial fluid in patients with osteoarthritis. Contribution of joint tissues to their articular production.
      • Duan Y.
      • Hao D.
      • Li M.
      • Wu Z.
      • Li D.
      • Yang X.
      • et al.
      Increased synovial fluid visfatin is positively linked to cartilage degradation biomarkers in osteoarthritis.
      Mechano-transductionAbove normal levels of cartilage degradation fragments (e.g., cartilage oligomeric matrix protein (COMP), dCOMP, CTX-II, C2M, C2C, AGNx1, fibronectin)

      Increased synovial fluid proteolytical activity (e.g., MMP1, MMP3)

      Bone sclerosis and turnover (e.g., alpha-CTX-I, sclerostin, RANK-L, osteocalcin)
      • Sokolove J.
      • Zhao X.
      • Chandra P.E.
      • Robinson W.H.
      Immune complexes containing citrullinated fibrinogen costimulate macrophages via Toll-like receptor 4 and Fcgamma receptor.
      • Goldberg V.M.
      • Kresina T.F.
      Immunology of articular cartilage.
      • de J.H.
      • Berlo S.E.
      • Hombrink P.
      • Otten H.G.
      • van E.W.
      • Lafeber F.P.
      • et al.
      Cartilage proteoglycan aggrecan epitopes induce proinflammatory autoreactive T-cell responses in rheumatoid arthritis and osteoarthritis.
      • Girardelli M.
      • Bianco A.M.
      • Marcuzzi A.
      • Crovella S.
      A comparative analysis of serologic parameters and oxidative stress in osteoarthritis and rheumatoid arthritis: reply to Mishra and colleagues.
      • Clark A.G.
      • Jordan J.M.
      • Vilim V.
      • Renner J.B.
      • Dragomir A.D.
      • Luta G.
      • et al.
      Serum cartilage oligomeric matrix protein reflects osteoarthritis presence and severity: the Johnston County Osteoarthritis Project.
      • Pelletier J.P.
      • Raynauld J.P.
      • Caron J.
      • Mineau F.
      • Abram F.
      • Dorais M.
      • et al.
      Decrease in serum level of matrix metalloproteinases is predictive of the disease-modifying effect of osteoarthritis drugs assessed by quantitative MRI in patients with knee osteoarthritis.
      • Catterall J.B.
      • Hsueh M.F.
      • Stabler T.V.
      • McCudden C.R.
      • Bolognesi M.
      • Zura R.
      • et al.
      Protein modification by deamidation indicates variations in joint extracellular matrix turnover.
      • Karsdal M.A.
      • Henriksen K.
      • Leeming D.J.
      • Woodworth T.
      • Vassiliadis E.
      • Bay-Jensen A.C.
      Novel combinations of post-translational modification (PTM) neo-epitopes provide tissue-specific biochemical markers – are they the cause or the consequence of the disease?.
      DemographicAge, significant trauma, obesity, altered gait, altered biomechanics (varus or valgus deformity), and excessive loading on the affected joint
      • Gelber A.C.
      • Hochberg M.C.
      • Mead L.A.
      • Wang N.Y.
      • Wigley F.M.
      • Klag M.J.
      Joint injury in young adults and risk for subsequent knee and hip osteoarthritis.
      • Gelber A.C.
      • Hochberg M.C.
      • Mead L.A.
      • Wang N.Y.
      • Wigley F.M.
      • Klag M.J.
      Body mass index in young men and the risk of subsequent knee and hip osteoarthritis.
      ImagingMRI assessment of the joint including; cartilage surface, damage/lesions, synovitis, Bone morrow edema (BME), osteophytes and subchondral bone (shape and trabecular structure) in addition to other advanced technologies
      • Dam E.B.
      • Loog M.
      • Christiansen C.
      • Byrjalsen I.
      • Folkesson J.
      • Nielsen M.
      • et al.
      Identification of progressors in osteoarthritis by combining biochemical and MRI-based markers.
      • Kinds M.B.
      • Marijnissen A.C.
      • Viergever M.A.
      • Emans P.J.
      • Lafeber F.P.
      • Welsing P.M.
      Identifying phenotypes of knee osteoarthritis by separate quantitative radiographic features may improve patient selection for more targeted treatment.
      • Holla J.F.
      • van der L.M.
      • Heymans M.W.
      • Roorda L.D.
      • Bierma-Zeinstra S.M.
      • Boers M.
      • et al.
      Three trajectories of activity limitations in early symptomatic knee osteoarthritis: a 5-year follow-up study.
      A few factors of relevance to the different type/stages of OA are given attention. The release of citrullinated fragments of different ECM proteins as well as intracellular proteins has been shown to act as autoantigens. To date, one such fragment, VICM, has been identified and shown to predict radiographic joint disease progression in a subset of patients
      • Bay-Jensen A.C.
      • Karsdal M.A.
      • Vassiliadis E.
      • Wichuk S.
      • Marcher-Mikkelsen K.
      • Lories R.
      • et al.
      Circulating citrullinated vimentin fragments reflect disease burden in ankylosing spondylitis and have prognostic capacity for radiographic progression.
      . Genetic factors may also be of particular importance as GDF5 mutations
      • Dodd A.W.
      • Syddall C.M.
      • Loughlin J.
      A rare variant in the osteoarthritis-associated locus GDF5 is functional and reveals a site that can be manipulated to modulate GDF5 expression.
      have been shown to predispose to a lack of repair potential, which may be of paramount importance after traumatic injury. In addition, there are a long list of genes that are related to cartilage and bone metabolism such as Receptor activator of nuclear factor kappa-B ligand (RANK), RANK-L, Sclerostin, Cathepsin K, V-ATPase, CLC-7, MMP-9 and -2, that may influence the traumatic and high turnover stage of disease
      • Henriksen K.
      • Karsdal M.A.
      • John M.T.
      Osteoclast-derived coupling factors in bone remodeling.
      • Henriksen K.
      • Neutzsky-Wulff A.V.
      • Bonewald L.F.
      • Karsdal M.A.
      Local communication on and within bone controls bone remodeling.
      . Further, our current understanding of OA suggests that inflammation may be a driver of disease albeit not an initiator of disease, as mutations in IL-1, predispose to inflammation and consequently fast progression in OA
      • Attur M.
      • Belitskaya-Levy I.
      • Oh C.
      • Krasnokutsky S.
      • Greenberg J.
      • Samuels J.
      • et al.
      Increased interleukin-1beta gene expression in peripheral blood leukocytes is associated with increased pain and predicts risk for progression of symptomatic knee osteoarthritis.
      • Attur M.
      • Wang H.Y.
      • Kraus V.B.
      • Bukowski J.F.
      • Aziz N.
      • Krasnokutsky S.
      • et al.
      Radiographic severity of knee osteoarthritis is conditional on interleukin 1 receptor antagonist gene variations.
      . This would be consistent with the biology of IL-1 and IL-1 receptor in chondrocyte function and cartilage degradation
      • Goldring M.B.
      • Berenbaum F.
      The regulation of chondrocyte function by proinflammatory mediators: prostaglandins and nitric oxide.
      . In parallel to RA where early radiographic damage in RA is a paramount predictor of later damage
      • Hashimoto J.
      • Garnero P.
      • van der H.D.
      • Miyasaka N.
      • Yamamoto K.
      • Kawai S.
      • et al.
      A combination of biochemical markers of cartilage and bone turnover, radiographic damage and body mass index to predict the progression of joint destruction in patients with rheumatoid arthritis treated with disease-modifying anti-rheumatic drugs.
      , a similar concept is applied to OA where different imaging modalities clearly may assist in delineating different phenotypes and progression
      • Dam E.B.
      • Loog M.
      • Christiansen C.
      • Byrjalsen I.
      • Folkesson J.
      • Nielsen M.
      • et al.
      Identification of progressors in osteoarthritis by combining biochemical and MRI-based markers.
      • Kinds M.B.
      • Marijnissen A.C.
      • Viergever M.A.
      • Emans P.J.
      • Lafeber F.P.
      • Welsing P.M.
      Identifying phenotypes of knee osteoarthritis by separate quantitative radiographic features may improve patient selection for more targeted treatment.
      • Holla J.F.
      • van der L.M.
      • Heymans M.W.
      • Roorda L.D.
      • Bierma-Zeinstra S.M.
      • Boers M.
      • et al.
      Three trajectories of activity limitations in early symptomatic knee osteoarthritis: a 5-year follow-up study.
      . These combined highlighted modalities, in addition to the traditional risk factors; age, significant trauma, obesity, altered gait, altered biomechanics (for example, a varus or valgus deformity), and excessive loading on the affected joint
      • Gelber A.C.
      • Hochberg M.C.
      • Mead L.A.
      • Wang N.Y.
      • Wigley F.M.
      • Klag M.J.
      Joint injury in young adults and risk for subsequent knee and hip osteoarthritis.
      • Gelber A.C.
      • Hochberg M.C.
      • Mead L.A.
      • Wang N.Y.
      • Wigley F.M.
      • Klag M.J.
      Body mass index in young men and the risk of subsequent knee and hip osteoarthritis.
      , may be part of a proposition for PHC in OA.
      One relevant example of PHC in OA may be found in the recent fibroblast growth factor (FGF)-18 studies. In a retrospective analysis of data from a clinical trial with Sprifermin (rhFGF18)
      • Mcpherson R.
      • Felchsenhar K.
      • Hellot S.
      • Eckstein F.
      A randomized, double-blind, placebo-controlled, multicenter study of rhfgf18 administered intraarticularly using single or multiple ascending doses in patients with primary knee osteoarthritis (OA), not expected to require knee surgery within 1 year.
      , a combination of two SNPs in the IL1RN gene could be indicative of disease severity/progression as well as potential response to Sprifermin in defined genetic groups
      • Valsesia A.
      • Cara Di
      • Berton A.
      • Debaillel V.
      • Wojcik J.
      • Farmer P.
      • et al.
      Exploratory biomarker analysis reveals agenetic epistatic interaction in interleukin 1 receptor antagonist (il1rn) gene associated to cartilage volume growth measured by MRI in response to sprifermin therapy in patients with knee radiographic osteoarthritis.
      . Four groups of patients were identified and tested for a statistical association with change in cartilage thickness/volume (as measured by MRI) and WOMAC scores. Groups of patients stratified for the SNPs rs9005 and rs315952 showed better response to Sprifermin therapy in terms of cartilage volume
      • Valsesia A.
      • Cara Di
      • Berton A.
      • Debaillel V.
      • Wojcik J.
      • Farmer P.
      • et al.
      Exploratory biomarker analysis reveals agenetic epistatic interaction in interleukin 1 receptor antagonist (il1rn) gene associated to cartilage volume growth measured by MRI in response to sprifermin therapy in patients with knee radiographic osteoarthritis.
      . This genetic stratification may be combined with serological technologies or other biomarkers such as described in Table I to further improve the prediction.

      Which technological advances may facilitate the advent of PHC for the OA field?

      The OA field is somewhat behind the RA field, although also in the RA field stratified phenotype directed treatment is also still in development This difference in the available toolbox, critical to the success of PHC, is most likely due to the current lack of successful intervention strategies in OA. This is further underscored by the almost accepted postulation in the OA field that novel treatments may more readily be identified and developed by identification of the right subpopulation of OA patients. Consequently, this discussion is reminiscent of the chicken and the egg discussion. However, with the emerging BIPED categorization and stratification of biochemical markers in the OA field
      • Bauer D.C.
      • Hunter D.J.
      • Abramson S.B.
      • Attur M.
      • Corr M.
      • Felson D.
      • et al.
      Classification of osteoarthritis biomarkers: a proposed approach.
      , tools are beginning to be identified that may assist in the advent of PHC for OA. Different methodologies are being classified as reflecting Burden of disease (B), Investigative (I), Prognostic (P), Efficacy of intervention (E), Diagnostic (D) and (S) Safety
      • Bauer D.C.
      • Hunter D.J.
      • Abramson S.B.
      • Attur M.
      • Corr M.
      • Felson D.
      • et al.
      Classification of osteoarthritis biomarkers: a proposed approach.
      • Kraus V.B.
      • 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.
      . Of most importance for PHC would be Prognostic markers at baseline that could predict optimal responders and Safety markers. Table I lists some of the genetic risk factors identified in GWAS studies, biochemical markers, and imaging markers which may facilitate the establishment of a PHC paradigm in OA by helping to define optimal patient populations for specific interventions.
      We suggest that there needs to be a focused effort on PHC that could be initiated in existing cohorts. PHC efforts could be grounded in analyses of markers such as those listed in Table I, to identify patients most likely to respond to a particular treatment. In addition, in other fields, marker panels are emerging for specific organ toxicity, such as the acute kidney injury panel
      • Slocum J.L.
      • Heung M.
      • Pennathur S.
      Marking renal injury: can we move beyond serum creatinine?.
      , which is in direct alignment with the suggestion of the “FDA critical path” that the efficacy marker of one pathology may be the safety marker of another
      • Karsdal M.A.
      • Henriksen K.
      • Leeming D.J.
      • Mitchell P.
      • Duffin K.
      • Barascuk N.
      • et al.
      Biochemical markers and the FDA critical path: how biomarkers may contribute to the understanding of pathophysiology and provide unique and necessary tools for drug development.
      . Such public efforts may assist in generating a better road-map for drug development in the OA field, by identification of the OA patient in whom safety indicators are optimized. Such efforts should also provide quicker and safer decision making in drug development that are of paramount importance for pharmaceutical companies to commit the needed investments in later stage clinical development. These efforts are expected to further lower the investment barrier, allowing more optimal clinical trial design for critical POC studies allowing for trials with fewer patients and shorter trial durations to reach an objective decision point. The best biomarkers would improve success related to the following:
      • 1.
        Identification of fast progressors
      • 2.
        Identification of the optimal patients for a specific type of intervention
      • 3.
        Enablement of phase II and III studies with fewer patients
      • 4.
        Decreasing cost of trials
      • 5.
        Minimizing drug-related adverse events
      • 6.
        Potentially providing early confidence in mechanism of action
      • 7.
        Surveillance for off-target effects

      Conclusion

      In conclusion, experience to date suggests that most medicines need to be developed to target specific subpopulations. We propose that OA is a suitable candidate for the application of PHC. As novel technologies and advanced clinical chemistry approaches, such as protein fingerprints or other robust methodologies, become a reality, the feasibility of PHC increases. The ever increasing prospect of PHC offers hope for a better future for patients, physicians and payers.

      Author contributions

      All authors discuss the concept of the manuscript. MK drafted the first version. All authors edited and approved the final version of the manuscript.

      Conflict of interests

      MK, ACBJ, KH and CC are full time employees of Nordic Bioscience. A company engaged in the discovery and development of biochemical makers. MK and CC own stock in Nordic Bioscience. CL is a full time employee of Merck-Serono, a company engaged in the development of treatment for OA.

      Acknowledgment

      The Danish Research Foundation , “den danske forskningsfond (DDF)” is acknowledged for supporting this project.

      References

        • Karsdal M.A.
        • Henriksen K.
        • Leeming D.J.
        • Mitchell P.
        • Duffin K.
        • Barascuk N.
        • et al.
        Biochemical markers and the FDA critical path: how biomarkers may contribute to the understanding of pathophysiology and provide unique and necessary tools for drug development.
        Biomarkers. 2009 May; 14: 181-202
        • Abrahamsen B.
        • Eiken P.
        • Eastell R.
        Cumulative alendronate dose and the long-term absolute risk of subtrochanteric and diaphyseal femur fractures: a register-based national cohort analysis.
        J Clin Endocrinol Metab. 2010 Dec; 95 (Epub 2010 Sep 15.): 5258-5265https://doi.org/10.1210/jc.2010-1571
        • Cooper C.
        • Emkey R.D.
        • McDonald R.H.
        • Hawker G.
        • Bianchi G.
        • Wilson K.
        • et al.
        Efficacy and safety of oral weekly ibandronate in the treatment of postmenopausal osteoporosis.
        J Clin Endocrinol Metab. 2003 Oct; 88: 4609-4615
        • Cummings S.R.
        • San M.J.
        • McClung M.R.
        • Siris E.S.
        • Eastell R.
        • Reid I.R.
        • et al.
        Denosumab for prevention of fractures in postmenopausal women with osteoporosis.
        N Engl J Med. 2009 Aug 20; 361: 756-765
        • Ravn P.
        • Hosking D.
        • Thompson D.
        • Cizza G.
        • Wasnich R.D.
        • McClung M.
        • et al.
        Monitoring of alendronate treatment and prediction of effect on bone mass by biochemical markers in the early postmenopausal intervention cohort study.
        J Clin Endocrinol Metab. 1999 Jul; 84: 2363-2368
        • Pao W.
        • Hutchinson K.E.
        Chipping away at the lung cancer genome.
        Nat Med. 2012 Mar; 18: 349-351
        • Syversen S.W.
        • Goll G.L.
        • van der H.D.
        • Landewe R.
        • Lie B.A.
        • Odegard S.
        • et al.
        Prediction of radiographic progression in rheumatoid arthritis and the role of antibodies against mutated citrullinated vimentin: results from a 10-year prospective study.
        Ann Rheum Dis. 2010 Feb; 69: 345-351
        • Bijlsma J.W.
        Optimal treatment of rheumatoid arthritis: EULAR recommendations for clinical practice.
        Pol Arch Med Wewn. 2010 Sep; 120: 347-353
        • Knevel R.
        • Schoels M.
        • Huizinga T.W.
        • Aletaha D.
        • Burmester G.R.
        • Combe B.
        • et al.
        Current evidence for a strategic approach to the management of rheumatoid arthritis with disease-modifying antirheumatic drugs: a systematic literature review informing the EULAR recommendations for the management of rheumatoid arthritis.
        Ann Rheum Dis. 2010 Jun; 69: 987-994
        • Nam J.L.
        • Winthrop K.L.
        • van Vollenhoven R.F.
        • Pavelka K.
        • Valesini G.
        • Hensor E.M.
        • et al.
        Current evidence for the management of rheumatoid arthritis with biological disease-modifying antirheumatic drugs: a systematic literature review informing the EULAR recommendations for the management of RA.
        Ann Rheum Dis. 2010 Jun; 69: 976-986
        • Gaujoux-Viala C.
        • Smolen J.S.
        • Landewe R.
        • Dougados M.
        • Kvien T.K.
        • Mola E.M.
        • et al.
        Current evidence for the management of rheumatoid arthritis with synthetic disease-modifying antirheumatic drugs: a systematic literature review informing the EULAR recommendations for the management of rheumatoid arthritis.
        Ann Rheum Dis. 2010 Jun; 69: 1004-1009
        • Upchurch K.S.
        • Kay J.
        Evolution of treatment for rheumatoid arthritis.
        Rheumatology (Oxford). 2012 Dec; 51: vi28-vi36
        • Kay J.
        • Upchurch K.S.
        ACR/EULAR 2010 rheumatoid arthritis classification criteria.
        Rheumatology (Oxford). 2012 Dec; 51: vi5-vi9
        • Reynolds A.
        • Koenig A.S.
        • Bananis E.
        • Singh A.
        When is switching warranted among biologic therapies in rheumatoid arthritis?.
        Expert Rev Pharmacoecon Outcomes Res. 2012 Jun; 12: 319-333
        • Smolen J.S.
        • Aletaha D.
        Forget personalised medicine and focus on abating disease activity.
        Ann Rheum Dis. 2013 Jan; 72: 3-6
        • Smolen J.S.
        • Aletaha D.
        • Bijlsma J.W.
        • Breedveld F.C.
        • Boumpas D.
        • Burmester G.
        • et al.
        Treating rheumatoid arthritis to target: recommendations of an international task force.
        Ann Rheum Dis. 2010 Apr; 69: 631-637
        • Verweij C.L.
        Pharmacogenetics: anti-TNF therapy in RA – towards personalized medicine?.
        Nat Rev Rheumatol. 2011 Mar; 7: 136-138
        • Bergman G.J.
        • Hochberg M.C.
        • Boers M.
        • Wintfeld N.
        • Kielhorn A.
        • Jansen J.P.
        Indirect comparison of tocilizumab and other biologic agents in patients with rheumatoid arthritis and inadequate response to disease-modifying antirheumatic drugs.
        Semin Arthritis Rheum. 2010 Jun; 39: 425-441
        • Devine E.B.
        • fonso-Cristancho R.
        • Sullivan S.D.
        Effectiveness of biologic therapies for rheumatoid arthritis: an indirect comparisons approach.
        Pharmacotherapy. 2011 Jan; 31: 39-51
        • Bijlsma J.W.
        • Berenbaum F.
        • Lafeber F.P.
        Osteoarthritis: an update with relevance for clinical practice.
        Lancet. 2011 Jun 18; 377: 2115-2126
        • Bay-Jensen A.C.
        • Slagboom E.
        • Chen-An P.
        • Alexandersen P.
        • Qvist P.
        • Christiansen C.
        • et al.
        Role of hormones in cartilage and joint metabolism: understanding an unhealthy metabolic phenotype in osteoarthritis.
        Menopause. 2013 May; 20: 578-586
        • Bos S.D.
        • Beekman M.
        • Maier A.B.
        • Karsdal M.A.
        • Kwok W.Y.
        • Bay-Jensen A.C.
        • et al.
        Metabolic health in families enriched for longevity is associated with low prevalence of hand osteoarthritis and influences OA biomarker profiles.
        Ann Rheum Dis. 2013 Oct; 72 (http://dx.doi.org/10.1136/annrheumdis-2012-202013. Epub 2012 Oct 27): 1669-1674
        • Karsdal M.A.
        • Madsen S.H.
        • Christiansen C.
        • Henriksen K.
        • Fosang A.J.
        • Sondergaard B.C.
        Cartilage degradation is fully reversible in the presence of aggrecanase but not matrix metalloproteinase activity.
        Arthritis Res Ther. 2008; 10: R63
        • Attur M.
        • Belitskaya-Levy I.
        • Oh C.
        • Krasnokutsky S.
        • Greenberg J.
        • Samuels J.
        • et al.
        Increased interleukin-1beta gene expression in peripheral blood leukocytes is associated with increased pain and predicts risk for progression of symptomatic knee osteoarthritis.
        Arthritis Rheum. 2011 Jul; 63: 1908-1917
        • Karsdal M.A.
        • Leeming D.J.
        • Dam E.B.
        • Henriksen K.
        • Alexandersen P.
        • Pastoureau P.
        • et al.
        Should subchondral bone turnover be targeted when treating osteoarthritis?.
        Osteoarthritis Cartilage. 2008 Jun; 16: 638-646
        • Isaacs J.D.
        • Cohen S.B.
        • Emery P.
        • Tak P.P.
        • Wang J.
        • Lei G.
        • et al.
        Effect of baseline rheumatoid factor and anticitrullinated peptide antibody serotype on rituximab clinical response: a meta-analysis.
        Ann Rheum Dis. 2013 Mar; 72 (http://dx.doi.org/10.1136/annrheumdis-2011-201117. Epub 2012 Jun 11): 329-336
        • van der Helm-van Mil A.H.
        • Knevel R.
        • Cavet G.
        • Huizinga T.W.
        • Haney D.J.
        An evaluation of molecular and clinical remission in rheumatoid arthritis by assessing radiographic progression.
        Rheumatology (Oxford). 2013 May; 52 (http://dx.doi.org/10.1093/rheumatology/kes378): 839-846
        • Li W.
        • Sasso E.H.
        • Emerling D.
        • Cavet G.
        • Ford K.
        Impact of a multi-biomarker disease activity test on rheumatoid arthritis treatment decisions and therapy use.
        Curr Med Res Opin. 2013 Jan; 29: 85-92
        • Eastman P.S.
        • Manning W.C.
        • Qureshi F.
        • Haney D.
        • Cavet G.
        • Alexander C.
        • et al.
        Characterization of a multiplex, 12-biomarker test for rheumatoid arthritis.
        J Pharm Biomed Anal. 2012 Nov; 70: 415-424
        • Keystone E.C.
        • Combe B.
        • Smolen J.
        • Strand V.
        • Goel N.
        • van V.R.
        • et al.
        Sustained efficacy of certolizumab pegol added to methotrexate in the treatment of rheumatoid arthritis: 2-year results from the RAPID 1 trial.
        Rheumatology (Oxford). 2012 Sep; 51: 1628-1638
        • Curtis J.R.
        • Yang S.
        • Chen L.
        • Park G.S.
        • Bitman B.
        • Wang B.
        • et al.
        Predicting low disease activity and remission using early treatment response to antitumour necrosis factor therapy in patients with rheumatoid arthritis: exploratory analyses from the TEMPO trial.
        Ann Rheum Dis. 2012 Feb; 71: 206-212
        • Karsdal M.A.
        • Woodworth T.
        • Henriksen K.
        • Maksymowych W.P.
        • Genant H.
        • Vergnaud P.
        • et al.
        Biochemical markers of ongoing joint damage in rheumatoid arthritis – current and future applications, limitations and opportunities.
        Arthritis Res Ther. 2011 Apr 28; 13: 215
        • Lotz M.
        • Martel-Pelletier J.
        • Christiansen C.
        • Brandi M.L.
        • Bruyere O.
        • Chapurlat R.
        • et al.
        Value of biomarkers in osteoarthritis: current status and perspectives.
        Ann Rheum Dis. 2013 Nov 1; 72: 1756-1763
        • Bay-Jensen A.C.
        • Liu Q.
        • Byrjalsen I.
        • Li Y.
        • Wang J.
        • Pedersen C.
        • et al.
        Enzyme-linked immunosorbent assay (ELISAs) for metalloproteinase derived type II collagen neoepitope, CIIM – increased serum CIIM in subjects with severe radiographic osteoarthritis.
        Clin Biochem. 2011 Apr; 44: 423-429
        • Barascuk N.
        • Veidal S.S.
        • Larsen L.
        • Larsen D.V.
        • Larsen M.R.
        • Wang J.
        • et al.
        A novel assay for extracellular matrix remodeling associated with liver fibrosis: an enzyme-linked immunosorbent assay (ELISA) for a MMP-9 proteolytically revealed neo-epitope of type III collagen.
        Clin Biochem. 2010 Jul; 43: 899-904
        • Karsdal M.A.
        • Nielsen M.J.
        • Sand J.M.
        • Henriksen K.
        • Genovese F.
        • Bay-Jensen A.C.
        • et al.
        Extracellular matrix remodeling: the common denominator in connective tissue diseases possibilities for evaluation and current understanding of the matrix as more than a passive architecture, but a key player in tissue failure.
        Assay Drug Dev Technol. 2013 Mar; 11 (http://dx.doi.org/10.1089/adt.2012.474. Epub 2012 Oct 9): 70-92
        • Veidal S.S.
        • Vassiliadis E.
        • Bay-Jensen A.C.
        • Tougas G.
        • Vainer B.
        • Karsdal M.A.
        Procollagen type I N-terminal propeptide (PINP) is a marker for fibrogenesis in bile duct ligation-induced fibrosis in rats.
        Fibrogenesis Tissue Repair. 2010 Apr 1; 3: 5
        • Wislowska M.
        • Jakubicz D.
        • Stepien K.
        • Cicha M.
        Serum concentrations of formation (PINP) and resorption (Ctx) bone turnover markers in rheumatoid arthritis.
        Rheumatol Int. 2009 Oct; 29: 1403-1409
        • Rosenquist C.
        • Fledelius C.
        • Christgau S.
        • Pedersen B.J.
        • Bonde M.
        • Qvist P.
        • et al.
        Serum CrossLaps One Step ELISA. First application of monoclonal antibodies for measurement in serum of bone-related degradation products from C-terminal telopeptides of type I collagen.
        Clin Chem. 1998 Nov; 44: 2281-2289
        • Garnero P.
        • Ferreras M.
        • Karsdal M.A.
        • NicAmhlaoibh R.
        • Risteli J.
        • Borel O.
        • et al.
        The type I collagen fragments ICTP and CTX reveal distinct enzymatic pathways of bone collagen degradation.
        J Bone Miner Res. 2003 May; 18: 859-867
        • Karsdal M.A.
        • Byrjalsen I.
        • Henriksen K.
        • Riis B.J.
        • Lau E.M.
        • Arnold M.
        • et al.
        The effect of oral salmon calcitonin delivered with 5-CNAC on bone and cartilage degradation in osteoarthritic patients: a 14-day randomized study.
        Osteoarthritis Cartilage. 2010 Feb; 18: 150-159
        • Spector T.D.
        • Conaghan P.G.
        • Buckland-Wright J.C.
        • Garnero P.
        • Cline G.A.
        • Beary J.F.
        • et al.
        Effect of risedronate on joint structure and symptoms of knee osteoarthritis: results of the BRISK randomized, controlled trial [ISRCTN01928173].
        Arthritis Res Ther. 2005; 7: R625-R633
        • Alexandersen P.
        • Karsdal M.A.
        • Byrjalsen I.
        • Christiansen C.
        Strontium ranelate effect in postmenopausal women with different clinical levels of osteoarthritis.
        Climacteric. 2011 Apr; 14: 236-243
        • Reginster J.Y.
        • Badurski J.
        • Bellamy N.
        • Bensen W.
        • Chapurlat R.
        • Chevalier X.
        • et al.
        Efficacy and safety of strontium ranelate in the treatment of knee osteoarthritis: results of a double-blind, randomised placebo-controlled trial.
        Ann Rheum Dis. 2013 Feb; 72 (http://dx.doi.org/10.1136/annrheumdis-2012-202231. Epub 2012 Nov): 179-186
        • Bondeson J.
        Activated synovial macrophages as targets for osteoarthritis drug therapy.
        Curr Drug Targets. 2010 May; 11: 576-585
        • Frisenda S.
        • Perricone C.
        • Valesini G.
        Cartilage as a target of autoimmunity: a thin layer.
        Autoimmun Rev. 2013 Mar; 12: 591-598
        • Dolzani P.
        • Assirelli E.
        • Pulsatelli L.
        • Addimanda O.
        • Mancarella L.
        • Peri G.
        • et al.
        Systemic inflammation and antibodies to citrullinated peptides in hand osteoarthritis.
        Clin Exp Rheumatol. 2011 Nov; 29: 1006-1009
        • Nishioka K.
        Autoimmune response in cartilage-delivered peptides in a patient with osteoarthritis.
        Arthritis Res Ther. 2004; 6: 6-7
        • Cooke T.D.
        Significance of immune complex deposits in osteoarthritic cartilage.
        J Rheumatol. 1987 May; 14 (Spec. No: 77–79)
        • Reynard L.N.
        • Loughlin J.
        Genetics and epigenetics of osteoarthritis.
        Maturitas. 2012 Mar; 71: 200-204
        • Stevenson J.C.
        A woman's journey through the reproductive, transitional and postmenopausal periods of life: impact on cardiovascular and musculo-skeletal risk and the role of estrogen replacement.
        Maturitas. 2011 Oct; 70: 197-205
        • Dam E.B.
        • Loog M.
        • Christiansen C.
        • Byrjalsen I.
        • Folkesson J.
        • Nielsen M.
        • et al.
        Identification of progressors in osteoarthritis by combining biochemical and MRI-based markers.
        Arthritis Res Ther. 2009; 11: R115
        • Bay-Jensen A.C.
        • Karsdal M.A.
        • Vassiliadis E.
        • Wichuk S.
        • Marcher-Mikkelsen K.
        • Lories R.
        • et al.
        Circulating citrullinated vimentin fragments reflect disease burden in ankylosing spondylitis and have prognostic capacity for radiographic progression.
        Arthritis Rheum. 2013 Apr; 65: 972-980
        • Dodd A.W.
        • Syddall C.M.
        • Loughlin J.
        A rare variant in the osteoarthritis-associated locus GDF5 is functional and reveals a site that can be manipulated to modulate GDF5 expression.
        Eur J Hum Genet. 2013 May; 21: 517-521
        • Henriksen K.
        • Karsdal M.A.
        • John M.T.
        Osteoclast-derived coupling factors in bone remodeling.
        Calcif Tissue Int. 2013 May 23;
        • Henriksen K.
        • Neutzsky-Wulff A.V.
        • Bonewald L.F.
        • Karsdal M.A.
        Local communication on and within bone controls bone remodeling.
        Bone. 2009 Jun; 44 (http://dx.doi.org/10.1016/j.bone.2009.03.671. Epub 2009 Apr 5): 1026-1033
        • Attur M.
        • Wang H.Y.
        • Kraus V.B.
        • Bukowski J.F.
        • Aziz N.
        • Krasnokutsky S.
        • et al.
        Radiographic severity of knee osteoarthritis is conditional on interleukin 1 receptor antagonist gene variations.
        Ann Rheum Dis. 2010 May; 69: 856-861
        • Goldring M.B.
        • Berenbaum F.
        The regulation of chondrocyte function by proinflammatory mediators: prostaglandins and nitric oxide.
        Clin Orthop Relat Res. 2004 Oct; : S37-S46
        • Hashimoto J.
        • Garnero P.
        • van der H.D.
        • Miyasaka N.
        • Yamamoto K.
        • Kawai S.
        • et al.
        A combination of biochemical markers of cartilage and bone turnover, radiographic damage and body mass index to predict the progression of joint destruction in patients with rheumatoid arthritis treated with disease-modifying anti-rheumatic drugs.
        Mod Rheumatol. 2009; 19: 273-282
        • Kinds M.B.
        • Marijnissen A.C.
        • Viergever M.A.
        • Emans P.J.
        • Lafeber F.P.
        • Welsing P.M.
        Identifying phenotypes of knee osteoarthritis by separate quantitative radiographic features may improve patient selection for more targeted treatment.
        J Rheumatol. 2013 Jun; 40: 891-902
        • Holla J.F.
        • van der L.M.
        • Heymans M.W.
        • Roorda L.D.
        • Bierma-Zeinstra S.M.
        • Boers M.
        • et al.
        Three trajectories of activity limitations in early symptomatic knee osteoarthritis: a 5-year follow-up study.
        Ann Rheum Dis. 2013 May 28;
        • Gelber A.C.
        • Hochberg M.C.
        • Mead L.A.
        • Wang N.Y.
        • Wigley F.M.
        • Klag M.J.
        Joint injury in young adults and risk for subsequent knee and hip osteoarthritis.
        Ann Intern Med. 2000 Sep 5; 133: 321-328
        • Gelber A.C.
        • Hochberg M.C.
        • Mead L.A.
        • Wang N.Y.
        • Wigley F.M.
        • Klag M.J.
        Body mass index in young men and the risk of subsequent knee and hip osteoarthritis.
        Am J Med. 1999 Dec; 107: 542-548
        • Mcpherson R.
        • Felchsenhar K.
        • Hellot S.
        • Eckstein F.
        A randomized, double-blind, placebo-controlled, multicenter study of rhfgf18 administered intraarticularly using single or multiple ascending doses in patients with primary knee osteoarthritis (OA), not expected to require knee surgery within 1 year.
        Osteoarthritis Cartilage. 1-4-2011; 19 (Ref Type: Abstract)
        • Valsesia A.
        • Cara Di
        • Berton A.
        • Debaillel V.
        • Wojcik J.
        • Farmer P.
        • et al.
        Exploratory biomarker analysis reveals agenetic epistatic interaction in interleukin 1 receptor antagonist (il1rn) gene associated to cartilage volume growth measured by MRI in response to sprifermin therapy in patients with knee radiographic osteoarthritis.
        Osteoarthritis Cartilage. 1-4-2013; 21 (Ref Type: Abstract)
        • Bauer D.C.
        • Hunter D.J.
        • Abramson S.B.
        • Attur M.
        • Corr M.
        • Felson D.
        • et al.
        Classification of osteoarthritis biomarkers: a proposed approach.
        Osteoarthritis Cartilage. 2006 Aug; 14: 723-727
        • Kraus V.B.
        • 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 May; 19: 515-542
        • Slocum J.L.
        • Heung M.
        • Pennathur S.
        Marking renal injury: can we move beyond serum creatinine?.
        Transl Res. 2012 Apr; 159: 277-289
        • Koivula M.K.
        • Aman S.
        • Karjalainen A.
        • Hakala M.
        • Risteli J.
        Are there autoantibodies reacting against citrullinated peptides derived from type I and type II collagens in patients with rheumatoid arthritis?.
        Ann Rheum Dis. 2005 Oct; 64: 1443-1450
        • Yoshida M.
        • Tsuji M.
        • Kurosaka D.
        • Kurosaka D.
        • Yasuda J.
        • Ito Y.
        • et al.
        Autoimmunity to citrullinated type II collagen in rheumatoid arthritis.
        Mod Rheumatol. 2006; 16: 276-281
        • Xiang Y.
        • Sekine T.
        • Nakamura H.
        • Imajoh-Ohmi S.
        • Fukuda H.
        • Yudoh K.
        • et al.
        Fibulin-4 is a target of autoimmunity predominantly in patients with osteoarthritis.
        J Immunol. 2006 Mar 1; 176: 3196-3204
        • Wildbaum G.
        • Nahir M.A.
        • Karin N.
        Beneficial autoimmunity to proinflammatory mediators restrains the consequences of self-destructive immunity.
        Immunity. 2003 Nov; 19: 679-688
        • Clague R.B.
        • Morgan K.
        • Collins I.
        • Pattrick M.
        • Doherty M.
        Absence of autoimmunity to type II collagen in generalised nodal osteoarthritis.
        Ann Rheum Dis. 1991 Nov; 50: 769-771
        • Bari A.S.
        • Carter S.D.
        • Bell S.C.
        • Morgan K.
        • Bennett D.
        Anti-type II collagen antibody in naturally occurring canine joint diseases.
        Br J Rheumatol. 1989 Dec; 28: 480-486
        • Yan X.
        • Yin L.
        • Wang Y.
        • Zhao Y.
        • Chang X.
        The low binding affinity of ADAMTS4 for citrullinated fibronectin may contribute to the destruction of joint cartilage in rheumatoid arthritis.
        Clin Exp Rheumatol. 2013 Mar; 31: 201-206
        • Yasuda T.
        Comparison of hyaluronan effects among normal, osteoarthritis, and rheumatoid arthritis cartilages stimulated with fibronectin fragment.
        Biomed Res. 2010 Feb; 31: 63-69
        • Bastow E.R.
        • Byers S.
        • Golub S.B.
        • Clarkin C.E.
        • Pitsillides A.A.
        • Fosang A.J.
        Hyaluronan synthesis and degradation in cartilage and bone.
        Cell Mol Life Sci. 2008 Feb; 65: 395-413
        • Barilla M.L.
        • Carsons S.E.
        Fibronectin fragments and their role in inflammatory arthritis.
        Semin Arthritis Rheum. 2000 Feb; 29: 252-265
        • Sokolove J.
        • Zhao X.
        • Chandra P.E.
        • Robinson W.H.
        Immune complexes containing citrullinated fibrinogen costimulate macrophages via Toll-like receptor 4 and Fcgamma receptor.
        Arthritis Rheum. 2011 Jan; 63: 53-62
        • Goldberg V.M.
        • Kresina T.F.
        Immunology of articular cartilage.
        J Rheumatol. 1987 May; 14 (Spec. No: 73–76)
        • de J.H.
        • Berlo S.E.
        • Hombrink P.
        • Otten H.G.
        • van E.W.
        • Lafeber F.P.
        • et al.
        Cartilage proteoglycan aggrecan epitopes induce proinflammatory autoreactive T-cell responses in rheumatoid arthritis and osteoarthritis.
        Ann Rheum Dis. 2010 Jan; 69: 255-262
        • Girardelli M.
        • Bianco A.M.
        • Marcuzzi A.
        • Crovella S.
        A comparative analysis of serologic parameters and oxidative stress in osteoarthritis and rheumatoid arthritis: reply to Mishra and colleagues.
        Rheumatol Int. 2012 May 6;
        • Deberg M.
        • Dubuc J.E.
        • Labasse A.
        • Sanchez C.
        • Quettier E.
        • Bosseloir A.
        • et al.
        One-year follow-up of Coll2-1, Coll2-1NO2 and myeloperoxydase serum levels in osteoarthritis patients after hip or knee replacement.
        Ann Rheum Dis. 2008 Feb; 67 (Epub 2007 Jul 20): 168-174
        • Siebuhr A.S.
        • Wang J.
        • Karsdal M.
        • Bay-Jensen A.C.
        • Y J.
        • Q Z.
        Matrix metalloproteinase-dependent turnover of cartilage, synovial membrane, and connective tissue is elevated in rats with collagen induced arthritis.
        J Transl Med. 2012; 10: 195
        • Catterall J.B.
        • Barr D.
        • Bolognesi M.
        • Zura R.D.
        • Kraus V.B.
        Post-translational aging of proteins in osteoarthritic cartilage and synovial fluid as measured by isomerized aspartate.
        Arthritis Res Ther. 2009; 11: R55
        • Kaarvatn M.H.
        • Jotanovic Z.
        • Mihelic R.
        • Etokebe G.E.
        • Mulac-Jericevic B.
        • Tijanic T.
        • et al.
        Associations of the interleukin-1 gene locus polymorphisms with risk to hip and knee osteoarthritis: gender and subpopulation differences.
        Scand J Immunol. 2013 Feb; 77: 151-161
        • Meulenbelt I.
        • Min J.L.
        • Bos S.
        • Riyazi N.
        • Houwing-Duistermaat J.J.
        • van der Wijk H.J.
        • et al.
        Identification of DIO2 as a new susceptibility locus for symptomatic osteoarthritis 1165.
        Hum Mol Genet. 2008 Jun 15; 17: 1867-1875
        • Valdes A.M.
        • Spector T.D.
        Genetic epidemiology of hip and knee osteoarthritis.
        Nat Rev Rheumatol. 2011 Jan; 7: 23-32
        • Bay-Jensen A.C.
        • Tabassi N.C.
        • Sondergaard L.V.
        • Andersen T.L.
        • Dagnaes-Hansen F.
        • Garnero P.
        • et al.
        The response to oestrogen deprivation of the cartilage collagen degradation marker, CTX-II, is unique compared with other markers of collagen turnover.
        Arthritis Res Ther. 2009; 11: R9
        • Berry P.A.
        • Jones S.W.
        • Cicuttini F.M.
        • Wluka A.E.
        • Maciewicz R.A.
        Temporal relationship between serum adipokines, biomarkers of bone and cartilage turnover, and cartilage volume loss in a population with clinical knee osteoarthritis.
        Arthritis Rheum. 2011 Mar; 63: 700-707
        • Gomez R.
        • Lago F.
        • Gomez-Reino J.
        • Dieguez C.
        • Gualillo O.
        Adipokines in the skeleton: influence on cartilage function and joint degenerative diseases.
        J Mol Endocrinol. 2009 Jul 1; 43: 11-18
        • Lajeunesse D.
        • Pelletier J.P.
        • Martel-Pelletier J.
        Osteoarthritis: a metabolic disease induced by local abnormal leptin activity?.
        Curr Rheumatol Rep. 2005 Apr; 7: 79-81
        • Presle N.
        • Pottie P.
        • Dumond H.
        • Guillaume C.
        • Lapicque F.
        • Pallu S.
        • et al.
        Differential distribution of adipokines between serum and synovial fluid in patients with osteoarthritis. Contribution of joint tissues to their articular production.
        Osteoarthritis Cartilage. 2006 Jul; 14: 690-695
        • Duan Y.
        • Hao D.
        • Li M.
        • Wu Z.
        • Li D.
        • Yang X.
        • et al.
        Increased synovial fluid visfatin is positively linked to cartilage degradation biomarkers in osteoarthritis.
        Rheumatol Int. 2011 Jan 19;
        • Clark A.G.
        • Jordan J.M.
        • Vilim V.
        • Renner J.B.
        • Dragomir A.D.
        • Luta G.
        • et al.
        Serum cartilage oligomeric matrix protein reflects osteoarthritis presence and severity: the Johnston County Osteoarthritis Project.
        Arthritis Rheum. 1999 Nov; 42: 2356-2364
        • Pelletier J.P.
        • Raynauld J.P.
        • Caron J.
        • Mineau F.
        • Abram F.
        • Dorais M.
        • et al.
        Decrease in serum level of matrix metalloproteinases is predictive of the disease-modifying effect of osteoarthritis drugs assessed by quantitative MRI in patients with knee osteoarthritis.
        Ann Rheum Dis. 2010 Dec; 69 (http://dx.doi.org/10.1136/ard.2009.122002. Epub 2010 Jun 22.): 2095-2101
        • Catterall J.B.
        • Hsueh M.F.
        • Stabler T.V.
        • McCudden C.R.
        • Bolognesi M.
        • Zura R.
        • et al.
        Protein modification by deamidation indicates variations in joint extracellular matrix turnover.
        J Biol Chem. 2012 Feb 10; 287: 4640-4651
        • Karsdal M.A.
        • Henriksen K.
        • Leeming D.J.
        • Woodworth T.
        • Vassiliadis E.
        • Bay-Jensen A.C.
        Novel combinations of post-translational modification (PTM) neo-epitopes provide tissue-specific biochemical markers – are they the cause or the consequence of the disease?.
        Clin Biochem. 2010 Jul; 43: 793-804