Abstract
Recent disappointments in late stage developments of anti-osteoarthritic drugs have reinforced efforts to develop better biomarkers for application in both the drug development process as well as in the routine management of these patients. Here we provide a brief review of biochemical tests available for the study of tissue turnover in each of the three compartments of the articular joint, that is the bone, the cartilage, and the synovium. Finally, we provide some perspective to future developments in biomarker discovery and discuss the potential impact such technologies could have on the drug development process.
INTRODUCTION
Obviously, the road to the identification and the clinical demonstration of both efficacy and safety of the chondroprotective drug is paved with numerous obstacles. Over the last few years, the disappointments associated with efforts to develop a disease modifying osteoarthritis drug (DMOAD) have been numerous and still today the millions of patients suffering from the serious, chronic disease can only be offered treatments aimed at improving signs and symptoms of the disease.
This huge, unmet medical need has been the primary driver behind major efforts to develop improved analytical techniques allowing better and more efficient clinical trial design and implementation.1
In this report, we aim to provide a brief and general overview of bone, cartilage, and synovium markers currently available for clinical and research use, and in particular we highlight recent studies investigating biochemical markers in OA. Finally, we provide perspectives on the possible enhancements on future drug development with better use of biomarker information.
BIOCHEMICAL MARKERS OF JOINT TISSUE
Application of biochemical markers in the study of osteoarthritis (OA) has attracted much attention, and the field has been the subject of several reviews over the last several years.2–7
The joint has three major compartments: the bone, the articular cartilage, and the synovium; and all three are affected by the disease,8 which manifests as osteophyte formation, subchondral sclerosis, articular cartilage breakdown, and alterations of the synovium such as inflammation, proliferation, and synovial thickening.
Cartilage
A central hallmark in OA pathogenicity is a gradual destruction of articular cartilage and local denudation leading to loss of joint function.9 The turnover of cartilage is normally maintained by a balance between catabolic and anabolic processes; however, in the case of pathological matrix destruction, the rate of cartilage degradation exceeds the rate of formation, resulting in a net loss of cartilage matrix.10,11 During degradation of articular cartilage, matrix metalloproteinases (MMPs) and aggrecanases are considered the most important proteases for degradation of articular cartilage.10,12 –14
Cartilage is a nonvascularized tissue consisting of chondrocytes and extracellular matrix (ECM). The ECM is a composite network of proteins, that is primarily collagens,15 interacting with polysaccharides and proteoglycans. While collagen type II is the most abundant collagen in articular cartilage, aggrecan is the predominant proteoglycan. Other important molecules in articular cartilage include cartilage oligomeric matrix protein (COMP), link protein, and hyaluronan (or hyaluronic acid).
In a longitudinal intervention study of diacerein, Mazieres and coworkers found by multivariate analysis that Hyaluronic acid (HA) and C-telopeptide of type II collagen (CTX-II), the latter generated by cleavage with MMPs, were significantly (both P < 0.0001) associated with radiographic progression of knee OA. With HA or CTX-II in the highest tertile, the relative risk for progression was 1.69 (95% confidence interval (CI), 1.25–2.27) and 2.00 (95% CI, 1.49–2.70), respectively, compared to the two lowest tertiles.16 This association remained significant after adjustment for baseline clinical, radiological, and treatment variables, and as the two biomarkers were independent, patients with both markers in the highest tertile had a relative risk of progression of 3.73 (95% CI, 2.48–5.61) compared to the two lowest tertiles. In addition, association with disease progression has been reported for in a few other studies, for example, for CTX-II,17,18 COMP,19 and type IIA procollagen amino terminal propeptide (PIIANP).20
In a recent study of oral salmon calcitonin, daily doses of 1 mg induced a significant reduction in both function and pain scores above placebo at days 42 and 84 in subjects with knee OA.21 At day 84, urinary levels of a series of biomarkers, including CTX-II, MMP-3, MMP-13, and HA, were significantly suppressed compared to baseline.
In two parallel, multinational, 2-year, and much larger studies, that is in the knee osteoarthritis structural arthritis (KOSTAR) study, all doses of risedronate failed to improve above placebo signs and symptoms by the Western Ontario and McMaster Osteoarthritis Index (WOMAC) and did not slow radiographic progression.22 Interestingly, in a subanalysis, it was demonstrated that in subjects with accelerated cartilage degradation at baseline (quantitatively assessed by urinary CTX-II), biochemical response after 6 months of risedronate use was associated with a significant reduction in radiological progression compared to subjects with no response in CTX-II (odds ratio 0.57 [95% CI, 0.39–0.85]).23
Recently, another type II collagen degradation marker was developed, that is Helix II, and evaluated in OA,24 and it has been reported to be complementary to CTX-II.25 Recently, however, the molecular specificity of the Helix II test has been questioned,26 and affinity for other types of collagen, for example fragments of type III collagen, was reported.
To compensate for the loss of ECM, the chondrocytes can up-regulate the synthesis of matrix components, including type II collagen, and these molecules can serve as biological markers of anabolic activity in the cartilage. Recently, we described an enzyme-linked immunosorbent assay (ELISA) detecting the N-terminal propeptide of type II collagen, PIINP,27 and reported massive suppression of the synthesis of this molecule in rheumatoid arthritis (RA). This suppression of PIINP expression has later been confirmed in OA,28 and also measurements of the splice variant A, that is PIIANP,29 is suppressed in OA. Others, however, find elevated PIIANP to be associated with radiographic progression of OA over 5 years.30 As PIIANP values are not always consistent,28 further analysis is needed to establish PIIANP as an anabolic marker.
A broad range of aggrecan fragments have been characterized in human synovial fluid,31 but only a few tests have been evaluated in appropriate clinical settings. A sandwich assay employing monoclonal antibody mAb OA-1 recognizing the 374ARGSV neoepitopes after capture by an antibody binding to keratin sulfate detected aggrecan fragments in the 40–100 fmol range in human synovial fluid32; however, the performance of this assay in human serum has not been reported yet. We have previously reported the quantification of aggrecan fragments in human serum using two different immunoassays,33,34 but further evaluation is required to determine the clinical usefulness of these and other aggrecan tests.
Bone
Apart from the articular cartilage, which has attracted most attention in the study of OA pathogenesis, an increasing body of evidence suggests that healthy subchondral bone turnover is prerequisite for preservation of the structural integrity of the articular cartilage (for recent reviews, please refer to Refs. 35–37).
The proliferative abnormalities observed in the skeletal compartment in OA not only encompass the bony sclerosis underneath the eroded cartilage (subchondral sclerosis) and osteophytes, but also ossification at the ligaments and the joint capsule is observed.37 During the development of OA, the subchondral cortical and trabecular bone architecture and properties are modified by cellular processes involving both osteoclasts and osteoblasts.35
Markers of bone formation include osteocalcin, bone alkaline phosphates, and the propeptides of type I collagen (N-terminal propeptide of type I procollagen (PINP), C-terminal propeptide of type I procollage (PICP)), and degradation markers include primarily various fragments of type I collagen (CTX-I, N-telopeptide of type I collage (NTX-I), C-terminal propeptide of type I procollagen (ICTP)).
In a cross-sectional study, Garnero and coworkers detected a 36%, 38%, and 52% reduction in concentrations of osteocalcin, serum CTX-I, and urine CTX-I, respectively,38 in 67 patients with knee OA compared to the same number of matched controls suggesting a general suppression of bone resorption in this disease group. However, several studies have failed to associate bone markers to clinical relevant end points (symptoms and function) as well as to structural damage in the cartilage subcompartment.16,39
Synovium
The vast majority of biomarker research has been focused on the two other compartments: the bone and the articular cartilage, but recent data suggest that synovial involvement, namely inflammation and proliferation, is a key component of OA.8
The synovium consists of the intima, which is a layer of cells (mostly macrophages and specialized fibroblasts), a superficial microvasculature net, and the subintima, which contains numerous lymphatic vessels draining liquid from the synovial cavity. The ECM of the subintima consists of type I and III collagens,40 which to some extent carry unusual glycosylations at the hydroxylysine residues, and hyaluronan as well as glycoproteins such as fibronectin, laminin, entactin, and tenascin.41
In particular, the urinary concentration of the glycosylated pyridinium cross-linker glucosyl-galactosyl-pyridinoline (Glc-Gal-PYD) has been investigated in OA as a marker of synovium tissue destruction.40 The cross-link has been reported to be elevated by 18% (P < 0.05) in knee OA and was significantly associated with total WOMAC index.38 Not surprisingly, Glc-Gal-PYD has been reported to be elevated in RA as well.42
Overview of Biochemical Markers of Joint Tissue
As is described earlier, the joint contains several compartments each of which has a complex biochemical composition, and therefore the biochemical marker potential of the joint is substantial. Table 1 provides an overview of the biochemical marker repertoire currently available for quantitative assessment of the tissue turnover in the joint.
FUTURE PERSPECTIVES
This review reflects an intensive area of biomarker research and it is expected that new markers and procedures for their better use will become available for application in clinical studies of OA as well as for improved managing of patients with this serious disease.
The recent disappointments in late stage OA drug development have renewed the debate on the potential impact of the biomarker repertoire on this process. In particular, it should be acknowledged that current inclusion criteria for OA trials favors the selection of study participants with relatively progressed disease, and furthermore several phase III clinical trials have reported modest structural progression rates in the untreated study population, for example 13% in the 2-year KOSTAR study discussed earlier,22 while the MMP inhibitor PG-11680062 was investigated in a trial with a 0.134 mm reduction in joint space width over 12 months. Obviously, both these important factors, progressed disease stage and modest progression, pose significant challenges for the drug under investigation. However, biomarkers, in particular the biochemical markers, carry the potential for identification at an early stage of individuals with elevated risk of structural disease progression.17,18,23,63
It is anticipated that future biomarker discovery will aim at combining technologies, as it seems unlikely that any single marker will offer sufficient sensitivity and specificity to allow efficient prediction of disease progression as well as rapid detection of clinically relevant response to medical intervention. Recently, Bauer and coworkers, under the Osteoarthritis Biomarkers Network funded by the National Institutes of Health/National Institute of Arthritis, Musculoskeletal, and Skin Disease (NIH/NIAMS) proposed a classification scheme for biomarkers termed BIPED, an acronym for Burden of Disease, Investigative, Prognostic, Efficacy of Intervention, and Diagnostic.64 The objective of the BIPED classification system is to provide specific biomarker definitions for improving development capabilities and analysis OA biomarkers and of communicating advances within a common framework. In brief, the five categories are characterized by the following key features:
This scheme offers a framework for evaluating the outcome of future biomarker discovery, and it contemplates integration/combination of independent markers as a single marker will not be applicable across all BIPED categories. Hopefully, biomarker discovery will soon provide the analytical tools required for increasing the effectiveness of the drug development process in joint diseases.
Footnotes
AUTHOR DISCLOSURE STATEMENT
A.-C.B.-J., B.C.S., C.C., M.A.K., S.H.M., and P.Q. are employees of Nordic Bioscience A/S.