TGF-β signalling in osteoarthritis: from cartilage homeostasis to therapeutic targeting

Osteoarthritis (OA) is the most common degenerative joint disease, characterised by progressive articular cartilage breakdown, subchondral bone remodelling, synovial inflammation, and functional decline. Transforming Growth Factor-β (TGF-β) is a multifunctional cytokine central to articular cartilage biology that maintains joint homeostasis under physiological conditions but contributes to pathological remodelling when its signalling is dysregulated. In articular cartilage, TGF-β signals principally through ALK5 and SMAD2/3, preserving the differentiated chondrocyte phenotype, sustaining extracellular matrix (ECM) production and restraining hypertrophic differentiation. This balance is disturbed in osteoarthritic cartilage. In ageing and in experimental OA, ALK5 expression falls further than ALK1, so the ALK1/ALK5 ratio rises and a larger share of the signal is carried by ALK1 and SMAD1/5/8. Canonical output falls further through induction of the inhibitory mediators SMAD7 and Smurf2 and through destabilisation of the type I receptor. We use TGF-β resistance in this review as a descriptive label for reduced canonical SMAD2/3 output despite preserved or raised ligand, not as an established disease entity comparable to insulin resistance. The consequences are increased matrix catabolism, chondrocyte hypertrophy, osteophyte formation and subchondral bone sclerosis. No disease-modifying osteoarthritis drug (DMOAD) is approved and treatment remains symptomatic. Strategies under investigation include isoform- and receptor-selective modulators, intra-articular delivery platforms, gene-based interventions and cell-based therapies. The problem is not whether to raise or lower TGF-β activity overall, since each direction carries a penalty: broad activation drives fibrosis, osteophyte formation and ectopic mineralisation, while broad inhibition removes chondroprotection and risks systemic effects. What is required is restoration of ALK5/SMAD2/3 output in cartilage without amplifying TGF-β activity in subchondral bone and synovium. A TGF-β-directed cell and gene therapy, TG-C, reach phase III testing in knee OA; the ACTiVION-II trial reported in 2026 did not meet its co-primary efficacy endpoints, illustrating the translational difficulty of this therapeutic approach. Progress toward TGF-β-directed therapy in OA depends on three things the field does not yet have: stratification of patients by pathway activity rather than by clinical phenotype, biomarkers that report pathway output rather than ligand concentration, and translational studies that define the benefit-risk of candidate modulators in each joint compartment separately.