Back

Microbiome variations in osteoarthritis reflect aging and metabolic factors, not the disease

Bevc, K.; Malfertheiner, L.; Neuenschwander, S.; Tran, V. D. T.; Pagni, M.; Pantiukh, K.; Jagura, M.; Niiranen, T.; Knight, R. T.; Salomaa, V.; Havulinna, A. S.; Org, E.; Eklund, K. K.; Barreto, G.; Zenobi, M.; von Mering, C.

2025-08-20 microbiology
10.1101/2025.06.24.661261 bioRxiv
Show abstract

The gut microbiome is crucial for human health. Its disruption has been linked to several chronic diseases, including joint disorders. The gut-joint axis has been implicated in the pathogenesis of osteoarthritis (OA), but conflicting findings and study limitations have led to uncertainty regarding the role of microbiota. We conducted a multi-cohort gut microbiota analysis in 1,395 OA patients from four European cohorts (Lifelines, EstMB, FINRISK 2002, TwinsUK), using stringent exclusion criteria and matched controls. When assessing microbial diversity, taxa, functional gene profiles, and gut permeability biomarkers, no significant differences were found between OA and controls. Although this does not exclude a causal contribution of the microbiota in the gut-joint-axis, its dysbiosis does not seem to be linked with OA disease progression. Instead, age and BMI appeared as the main contributing factors to microbiome changes. Microbiome studies in complex diseases often face challenges such as small sample sizes, batch effects, and limited ability to match appropriate controls, particularly in single-cohort designs. By combining data from multiple large cohorts, we were able to mitigate these limitations and provide a more robust assessment of microbiome association with OA. Our findings emphasize the need for rigorous study design in microbiome research and challenge the OA-gut dysbiosis hypothesis.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.