Classification of four distinct osteoarthritis subtypes with a knee joint tissue transcriptome atlas

Area of Science:

• Biomedical research
• Molecular biology
• Genomics

Background:

• Osteoarthritis (OA) lacks detailed molecular classifications, hindering targeted treatments.
• Current diagnostic methods like medical imaging are insufficient for precise OA patient stratification.

Purpose of the Study:

• To classify and molecularly define osteoarthritis (OA) subtypes.
• To identify distinct molecular signatures and potential therapeutic targets for OA patients.

Main Methods:

• Collected cartilage, subchondral bone, and synovium tissues from OA patients.
• Performed unsupervised clustering analysis on cartilage transcriptomes.
• Conducted ligand-receptor crosstalk analysis across multiple tissue types.
• Validated OA subtype classification using independent RNA-seq datasets and random forest modeling.

Main Results:

• Identified four distinct OA molecular subtypes: C1 (glycosaminoglycan metabolic disorder), C2 (collagen metabolic disorder), C3 (activated sensory neuron), and C4 (inflammation).
• Linked molecular functions (e.g., vasculature development) to clinical symptoms (e.g., osteophytes, joint space narrowing) via tissue crosstalk analysis.
• Successfully modeled OA subtypes in independent datasets, confirming the robustness of the classification.

Conclusions:

• Revealed novel molecular subtypes of knee osteoarthritis, distinct from traditional classifications.
• The identified subtypes provide a foundation for developing precise diagnostic tools and targeted therapeutic strategies for OA.
• This molecular subtyping approach may significantly advance personalized medicine in osteoarthritis treatment.