Brown fat tracks with fewer heart and psychiatric diagnoses in 101,243 patients, and AI pinpoints the axillary depot
Synopsis
Among 101,243 patients at a cancer center, those with brown fat had fewer cardiovascular and psychiatric diagnoses; machine learning quantified brown fat activity in about 8,700 people, showing higher activity, especially in the axillary depot, tracked with lower odds of type 2 diabetes and coronary disease, and led to losartan as an activator in mice.
Interpretation
Brown fat presence tracks with a broader disease spectrum than cardiovascular and metabolic conditions alone. Earlier cohorts of this kind grouped patients only by brown fat presence or absence and focused mainly on cardiometabolic outcomes, leaving other organ systems largely unexamined. In a cohort matched on age, sex, BMI and ambient temperature, 104 of 511 diagnostic codes were significantly associated with brown fat, 98 remained significant after adjustment, and 92 were less frequent in brown-fat-positive patients.
Brown fat metabolic activity can be quantified semi-automatically and resolved to individual anatomic depots. Large-scale depot-level activity analysis had not been feasible, so patients could only be classified as having brown fat or not. A machine learning workflow measured total and six-depot activity in 8,709 brown-fat-positive patients, defining activity as the product of mean standardized uptake value and volume.
Higher activity tracks with better metabolic and cardiac indices, most strongly for the axillary depot. Stratifying by activity tertiles against brown-fat-negative patients reveals dose-dependent differences. Higher total activity was associated with stepwise lower prevalence of type 2 diabetes, atrial fibrillation or flutter, coronary artery disease and congestive heart failure, while axillary activity was associated with lower odds of every queried cardiometabolic disease and the strongest negative association with visceral fat area.
Medication analysis points to angiotensin receptor blockers, and losartan was validated as a brown adipocyte activator in cells and mice. No approved, widely used medication had been confirmed to leverage the cardiometabolic benefits linked to brown fat. Losartan and valsartan were positively associated with brown fat presence; losartan raised UCP1 expression and norepinephrine-dependent oxygen consumption in mature brown adipocytes, and treated mice showed increased energy expenditure and cold-stimulated thermogenesis.
Perspective
What is demonstrated is an association in a retrospective cohort plus proof-of-concept cell and mouse experiments, in a cancer-center population with access to 18F-FDG-PET–CT; the authors call for large prospective studies and consider a large prospective PET–CT study in the non-oncologic setting highly implausible at present.
The authors note that scans were obtained without cold exposure, which may underestimate brown fat prevalence; the cohort comes from a cancer center, so effects of tumor types or cancer therapies cannot be excluded; 18F-FDG uptake may not correspond to thermogenic capacity and can be influenced by insulin resistance; and the lower bone density association contrasts with previous reports and may be cohort-specific.