The translational paradox of ADAMTS-5 in osteoarthritis: mechanistic centrality but limited clinical efficacy

Osteoarthritis (OA) is a heterogeneous whole-joint disease for which no disease-modifying therapies have demonstrated consistent clinical efficacy. Among matrix-degrading enzymes, ADAMTS-5 has strong genetic and pharmacological support as a major aggrecanase in experimental OA, whereas its quantitative contribution in human disease is more context dependent. This mechanistic rationale positioned ADAMTS-5 as a leading disease-modifying target. However, the clinical record requires careful separation by development stage: the Phase II ROCCELLA trial of GLPG1972/S201086 showed no significant structural or symptomatic benefit, whereas the anti-ADAMTS-5 nanobody M6495 has so far been evaluated mainly for Phase I safety, pharmacokinetics, and pharmacodynamics rather than efficacy. In this review, we re-evaluate ADAMTS-5 within a network integrating inflammatory signaling, mechanotransduction, extracellular-matrix remodeling, and tissue adaptation. We examine how disease stage, protease redundancy, biomarker-outcome dissociation, delivery limitations, and patient heterogeneity can weaken translation from enzyme inhibition to clinical benefit. We further discuss molecular endotyping, single-cell and spatial profiling, and data-driven stratification as tools to identify patients and disease states in which aggrecanase activity is most likely to be therapeutically relevant. Future strategies should therefore emphasize stage-specific and locally controlled intervention, biomarker-informed enrichment, and rational combination therapy rather than uniform systemic inhibition. ADAMTS-5 thus provides a useful model for understanding why a biologically compelling OA target may fail when therapeutic timing, tissue context, and patient selection are not aligned.