Deciphering osteoarthritis cartilage heterogeneity: gene expression-guided patient stratification and microbiome correlations.

Demand for arthroplasty due to osteoarthritis (OA) is rising with population aging and need for revision surgeries. Efforts to slow OA progression and prevent post-operative complications have been ineffective, likely due to inadequate identification of the primary disease-driving mechanisms underlying cartilage destruction. The aim of this study was to identify the primary disease-driving mechanisms underlying OA cartilage destruction. Patient stratification was performed using a Gaussian mixture model according to the expression profiles of 14 OA indicators in the cartilage collected from total or unicompartmental knee arthroplasty and verified by cartilage microbiome. Six endotypes were identified from 122 OA cartilage samples. Regarding extracellular matrix remodeling, Cluster 1 had high COL1A1 and was dominated by the Pseudomonadaceae family, potentially associated with fibrosis. Cluster 2 showed high COMP transcription and low MMP13, with a high level of species Acinetobacter johnsonii, which may affect matrix regeneration. Cluster 5 had high ADAMTS4 and abundant Streptococcus agalactiae, highlighting cartilage degradation in OA progression and infection risk in arthroplasty. Inflammation-driven destruction, as indicated by high MCP1 and CCR2 levels, was pronounced in Cluster 4 patients, with nine key taxa correlated with NTRK1, suggesting their role in pain sensitization. Regarding mechanosensitivity, Cluster 3 featured high TRPV4 expression and a predominance of the Mycobacteriales order, suggesting a link between mechanosensitivity and aging. Cluster 6 had secondary high levels of CCR2 and PIEZO1, indicating mechanoflammation. This study unveiled OA cartilage heterogeneity and highlighted the potential of using replaced cartilage, which may offer valuable clues for joint health assessment and future research into arthroplasty prognosis.