There are two major unmet needs in the field of RA that could be addressed by the development and validation of biomarkers. The first represents the risk of development of RA in the patient with arthralgia and the second is the risk of radiographic progression in early RA. Multivariate models show that known predictors contribute only about a third of the total risk for these endpoints. The OMERACT Soluble Biomarker working group has focused on the development of validation criteria for soluble biomarkers considered to reflect structural damage endpoints in RA, PsA, AS, according to the OMERACT Filter (truth, feasibility, discrimination) as well as a scheme for levels of evidence aimed at concluding that the soluble biomarker is sufficiently validated that it can substitute for radiographic damage parameters in clinical trials [1,2]. A desirable objective for an RA biomarker is that it acts as a modifiable lead indicator of development of the endpoint of interest, change in the biomarker reflects/predicts interval change in radiographic progression independently of known predictors regardless of treatment approach, the biomarker is more responsive than currently used predictors, and finally, it adds prognostic information over and above the combined information obtained from all other known predictors at both the group and individual patient level. The unmet need in SpA focuses on biomarkers of disease activity and disease progression. One desirable objective for an SpA biomarker is that it reflects disease activity as quantified by MRI scores for bone marrow edema irrespective of treatment. A second objective for an SpA biomarker is that early short-term change in the biomarker predicts change in the radiographic endpoint independently of known predictors. Accordingly, validation criteria have been categorized into essential criteria that focus on the demonstration of an independent association with the structural damage endpoint (van der Heijde modification of Sharp score for RA/PsA, mSASSS for SpA) in patients at all stages of disease with different drug classes at the level of both absolute and relative change in prospective cohort studies and randomized controlled trials (discrimination), reproducibility of measurement, and assay feasibility. Non-essential but desirable criteria focus on pre-clinical studies that address animal studies and relation of the biomarker to joint degradation (truth). A longitudinal study design template has been formulated focusing on a minimum set of standards with respect to principle inclusion criteria, study duration, treatment and rules for changes in treatment, timing of clinical and biomarker assessments, principle outcomes, processing of biomarker samples, documentation of potential confounders, and analysis of radiographic endpoint. A level of evidence scheme has been adapted from a generic framework for surrogates proposed by the OMERACT group which ranks the evidence supporting validation of a disease centered variable towards a patient centered surrogate according to 4 domains: target, study design, penalties, and statistical strength. For soluble biomarkers radiography constitutes the primary patient-centered target outcome, observational studies and RCTs are equally ranked in the study design domain, and the statistical strength domain is focused on regression based modeling and standardized mean difference to estimate the level of association between the biomarker and damage. The addition of a new domain on biomarker assay performance is considered essential, and it is recommended that this criterion be met in its entirety before starting clinical validation studies.
Maksymowych WP, et al. J Rheumatol 2009;36:1785.
Maksymowych WP, et al. J Rheumatol 2009;36:1792
Disclosure of Interest None declared