The use of osteoarthritis (OA) biomarkers in daily clinical practice is not in the near future. Why would we use OA biomarkers in daily practice, when our current focus is treating the illness (symptoms), and not the disease (structural joint changes)?
In the daily practice of OA research and development of new, disease-modifying OA interventions the situation is different. Monitoring OA in longitudinal or intervention studies requires outcome measures. Patient-reported outcomes on symptoms, function and quality of life are gold standard, the clinical endpoint. Other relevant outcomes include function and performance tests, imaging to monitor structure and quality of joint tissues, and molecular biomarkers to reflect the turnover of joint tissues. OA disease progression is slow, and only a minority of a study cohort progresses appreciably in the usual study window. OA studies therefore require long observation times, and there is a need for biomarkers that in a shorter time predict the long-term clinical outcome and treatment response.
A terminology for OA biomarkers was proposed which classifies biomarkers into 5 categories corresponding to their proposed use: Burden of disease, Investigational, Prognostic, Efficacy of Intervention, Diagnostic, and Safety. Biomarkers likely to have the earliest beneficial impact fall into two categories. The first are markers that would allow us to select for clinical trials the subjects that are most likely to respond or progress (prognostic markers) within a reasonable time. The second category includes those that provide early feedback for preclinical and early clinical decision-making that an intervention has the desired effect on the primary target (efficacy markers). Both types of biomarkers are highly desirable in OA where conventional outcomes may take years to present. They could reduce the burden and cost of trials by speeding up the development cycle.
OA biomarkers exist in various states of validation. Validation refers to evidence in support of marker use as an endpoint, and includes analytical performance characteristics and correlation of the biomarker with a specific biological process.
A second classification system divides biomarkers into four categories according to their level of qualification. Exploration level biomarkers are research tools with in vitro and/or preclinical evidence but without consistent information linking the biomarker to clinical outcomes in humans. Demonstration level biomarkers are associated with clinical outcomes in humans but have not been reproducibly shown in clinical studies. Characterization level biomarkers are reproducibly linked to clinical outcomes in more than one prospective clinical study in humans. Surrogacy level biomarkers can substitute for a clinical endpoint, correspond to a “surrogate end point”, and require approval by regulators as a registrable endpoint, a very challenging target.
Several commercially available OA-related biomarkers exist, and approximations with regard to their level of qualification have been presented. However, qualification is dependent on a specific context, limiting generalizability in the early development phase. Additional biomarkers are explored in the research laboratories. The great majority of OA biomarkers now investigated are at the exploration level, a few at the demonstration level. It is arguable if any are at the characterization level, and there are no biomarkers considered as surrogate clinical endpoints in OA.
The development and validation of biomarkers with utility for osteoarthritis requires the combination of reproducible and specific biomarker assays with large prospective cohort studies and controlled clinical trials. To demonstrate utility of efficacy biomarkers, these trials will need to show structural joint protection and/or clinically relevant efficacy.
Disclosure of Interest None Declared