Beyond symptomatic alignment: evaluating the integration of causal mechanisms in matching animal models with human pathotypes in osteoarthritis research

Abstract

Osteoarthritis (OA) is a highly prevalent and disabling condition lacking curative treatments, with only symptomatic relief available. Recognizing OA as a heterogenous disorder with diverse aetiologies and molecular foundations underscores the need to classify patients by both phenotypes and molecular pathomechanisms (endotypes). Such stratification could enable the development of targeted therapies to surmount existing treatment barriers. From a scientific, economic, and ethical perspective, it is crucial to employ animal models that accurately represent the endotype of the target patient population, not merely their clinical symptoms. These models must also account for intrinsic and extrinsic factors, like age, sex, metabolic status, and comorbidities, which impact OA's pathogenesis and its clinical and molecular variability and can profoundly influence not only structural and symptomatic disease severity and progression but also the underlying molecular pathophysiology. The molecular definition of the OA subpopulation must also be reflected in the read-outs, as the traditional methods—macroscopic and histological scoring, along with limited gene expression profiling of established biomarkers for cartilage degradation, extracellular matrix (ECM) turnover, and synovial inflammation—are inadequate for discovering new, phenotype- and endotype-specific biomarkers or therapeutic targets. Thus, animal model characterisation should evolve to include both clinically and pathophysiologically pertinent measures of disease progression and response to treatment. This review evaluates the utility and accuracy of current animal models in OA research, focusing on their capacity to replicate the disease’s pathophysiological processes.