Osteoarthritis (OA) has traditionally been viewed as a non-inflammatory arthropathy and has not been considered a ‘serious disease’. However, this view has radically changed in recent years, due to the complexity and heterogeneity of the patient populations, spiralling socio-economical costs and long-term impact on the quality of life of affected individuals. There is an acute need for objective and non-invasive diagnostic biomarkers in OA, markers that can stratify patient subtypes and thereby direct therapeutic treatments at an earlier disease stage (read personal health care (PHC)) (Conaghan, 2013). Increased interest in the development of new diagnostic and prognostic tests for early forms of OA may incorporate the use of blood-based biomarkers; however, both research and regulated development and approval are still needed to reach a diagnostically important significant point where a given biomarker will benefit the clinical management of the patient.
1. The OA Biomarker Landscape Today
2. What Is the Medical Need for Biomarkers?
Translational biomarkers, which allow better characterization of a drug in preclinical development, ensuring of selection of the most viable projects
Early identification of efficacy of intervention; Go/no-go decision-making already in phase 1b/2a studies, which normally do not include efficacy measures.
Phase II and Phase III trial enrichment; reduction in study size, and a particular OA phenotype tailored for a selective interventions, which will recuse length of the clinical study to allow more efficient and less costly trials
Identification of patients who are fast progressors and as such in greatest need of treatment.
Identification of super responders to a specific treatment; patients with high efficacy and low safety concerns
Biomarkers of disease activity; as OA is not a stabile disease, there is a need for devices for identification with high disease activity and potential progression
Easy accessible monitoring devices – point of care; post marketing patient care and personalized medicine
Although there are clear overlaps in the above, it is clear that no single biomarker will be the answer for all.
3. Message From the Regulators
On the regulatory side, the FDA issued a position document describing the need and road ahead for personalized medicine “FDA: Paving the Way for Personalized Medicine”,3
which later resulted in new guidelines to faster biomarker tool development by the guidelines “Identifying Potential Biomarkers for Qualification and Describing Contexts of Use To Address Areas Important to Drug Development”,4
which are in addition to the standard guidelines for in vitro companion diagnostic device. This has led to the discussion on prospective-retrospective biomarker analysis for regulatory consideration, by the white paper from the industry pharmacogenomics working group (Patterson et al., 2011). This will greatly assist precision medicine and PHC by guiding the discussion on how to implement a “prospective-retrospective biomarker analysis”. The prospective-retrospective biomarker analysis approach is developed to “rescue” failed phase III trials. Qualified biomarkers are to be measured in certified, high-quality laboratories and analyzed using predefined statistical analysis plans to test hypotheses related to retrospective analysis of technically and biologically validated biomarkers.
According to the FDA, a prognostic in vitro diagnostic biomarker would need a 510 K or de novo approval (class II device), whereas a predictive biomarker would need ldt pre-market approval (PMA, class III device). The main separating factor is that a prognostic biomarker provides you with an estimate for progression, whereas a predictive biomarker would be used to decide the exact treatment regimen for individual patients, and would therefore have a significant impact on the patient’s life. A predictive biomarker will often become a companion diagnostic.5
In addition, the recent “drug development tool (DDT) box” guidelines are also allowing for regulatory assessment of tools to assist in clinical drug development, such as the fibrinogen enrichment of patients in COPD clinical studies with a more severe outcome (fast progressors), which is now classified as a DDT.
No biomarkers have yet been qualified as biomarkers for OA, however several biomarkers have been developed targeting cartilage degradation and formation (e.g. CTX-II, ARGS, PIIANP), joint inflammation (e.g.C3M, Col2-NO2), bone remodelling (e.g. alpha CTX I, osteocalcin) as well as inflammation and metabolic factors (Bay-Jensen et al., 2016a, Bay-Jensen et al., 2016b). The scientists and clinicians working in the biomarker field cannot expect a “one size fits all” solution for OA. Consequently it is important to test and validate a biomarker to a specific hypothesis. This can be done under the laboratory-developed test (LDT) (Sarata and Johnson, 2014), which is a type of in vitro diagnostic test that is designed, manufactured and used within a single laboratory.
4. How Do We Move Forward?
In summary, a great deal of collaborative work needs to be done in this area to develop more predictive, prognostic, objective and complementary biomarkers for OA management and DMOAD development.