Osteoarthritis and cartilage · 2026
Molecular clustering in osteoarthritis primary tissues identifies shared inflammatory and tissue-specific pathway profiles.
СтатьяНазвание и аннотация: Перевод пока не подготовлен
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Stergiou Odysseas Sotirios, Bittner Norbert, Katsoula Georgia, Swift Diane, Wilkinson Jeremy Mark, Zeggini Eleftheria
Аннотация
OBJECTIVES: To disentangle the molecular heterogeneity of knee osteoarthritis (OA) through the classification and characterization of transcriptomic clusters in multiple joint tissues, and to uncover distinct biological pathways that will facilitate improved patient stratification.
METHODS: We analyzed RNA sequencing data from 330 knee OA patients across low- and high-grade OA knee cartilage, synovium and infrapatellar fat pad tissues. We used unsupervised machine learning to identify distinct transcriptomic clusters and subsequently performed cluster-specific differential expression and pathway enrichment analyses. We applied multi-omics factor analysis in low-grade cartilage to construct a gene expression-based classifier for subtype prediction, which we validated in an independent knee OA RNA sequencing dataset.
RESULTS: We identified robust clusters across all four joint tissues. In low-grade cartilage, we identified two patient groups separated by differences in inflammation and transcriptional regulation. A gene classifier distinguished these groups with a cross-validated accuracy of 94.5% (95% CI 91.1-96.6%). We also reproduced these subtypes in an external cohort in which the same axis similarly separated the subgroups. In high-grade cartilage there were three distinct clusters, characterized by inflammatory, neuroactive receptor-signaling, and housekeeping-transcriptional programs. Both synovium and infrapatellar fat pad showed two distinct subgroups. Despite the histological differences between these two tissues, subgrouping was based on shared biological functions related to immune activation, alongside disease tissue-specific ones.
CONCLUSIONS: Our findings identify gene expression-based patient clusters in different primary joint tissues and point to shared and disease tissue-specific molecular programs in OA, thus setting the foundation for transcription signature-based patient stratification.
METHODS: We analyzed RNA sequencing data from 330 knee OA patients across low- and high-grade OA knee cartilage, synovium and infrapatellar fat pad tissues. We used unsupervised machine learning to identify distinct transcriptomic clusters and subsequently performed cluster-specific differential expression and pathway enrichment analyses. We applied multi-omics factor analysis in low-grade cartilage to construct a gene expression-based classifier for subtype prediction, which we validated in an independent knee OA RNA sequencing dataset.
RESULTS: We identified robust clusters across all four joint tissues. In low-grade cartilage, we identified two patient groups separated by differences in inflammation and transcriptional regulation. A gene classifier distinguished these groups with a cross-validated accuracy of 94.5% (95% CI 91.1-96.6%). We also reproduced these subtypes in an external cohort in which the same axis similarly separated the subgroups. In high-grade cartilage there were three distinct clusters, characterized by inflammatory, neuroactive receptor-signaling, and housekeeping-transcriptional programs. Both synovium and infrapatellar fat pad showed two distinct subgroups. Despite the histological differences between these two tissues, subgrouping was based on shared biological functions related to immune activation, alongside disease tissue-specific ones.
CONCLUSIONS: Our findings identify gene expression-based patient clusters in different primary joint tissues and point to shared and disease tissue-specific molecular programs in OA, thus setting the foundation for transcription signature-based patient stratification.