Scientists have found the first evidence that bone marrow fat may help cause osteoporosis – a condition that weakens bones – and potentially fractures and type 2 diabetes.
The study used an innovative AI technique to analyse MRI scans from over 48,000 patients who have donated their data to the UK Biobank project, creating the most detailed picture yet of how bone marrow adiposity – fat stored inside bone marrow – affects human health.
The University of Edinburgh team say their findings suggest bone marrow fat is not just a marker of disease but may actively drive it, opening the door to new treatments.
Hidden fat
Bone marrow is a critical site of fat storage, accounting for approximately 10 per cent of total body fat in lean, healthy adults. Bone marrow fat increases with age and also builds up in conditions such as obesity, osteoporosis, diabetes, and after treatments like radiotherapy.
Bone marrow fat is not as well studied compared with other fat types, despite its prevalence and potential importance, experts say.
Disease links
To investigate, the research team developed AI tools to measure bone marrow fat in scans of key sites such as the spine and major leg bones. By comparing these measurements with participants’ medical records, they found that variation in bone marrow fat levels was linked to the risk of 47 different diseases.
The team analysed genetic data to create a risk score for an individual’s predisposition to have higher or lower bone marrow fat. These genetic scores reflected the same patterns seen in the imaging data and were also associated with multiple diseases, suggesting that bone marrow fat may help predict disease risk.
Finally, by applying a genetic technique that tests cause-and-effect, researchers found that excess bone marrow fat is not just associated with disease but may directly contribute to it. Their analysis indicated that bone marrow fat can drive osteoporosis, and may also play a causal role in fractures and type 2 diabetes.
Next steps
Researchers say further studies are needed to understand how bone marrow fat contributes to disease, whether it could be used as a predictive marker, and if it might be safely targeted for treatment.
They also caution that, while the evidence is strong, additional studies are needed before these findings can translate into therapies.