Search

Abstract The standard method for identifying active Brown Adipose Tissue (BAT) is [18F]-Fluorodeoxyglucose ([18F]-FDG) PET/CT imaging, which is costly and exposes patients to radiation, making it impractical for population studies. These issues can be addressed with computational methods that predict [18F]-FDG uptake by BAT from CT; earlier population studies pave the way for developing such methods by showing some correlation between the Hounsfield Unit (HU) of BAT in CT and the corresponding [18F]-FDG uptake in PET. In this study, we propose training convolutional neural networks (CNNs) to predict [18F]-FDG uptake by BAT from unenhanced CT scans in the restricted regions that are likely to contain BAT. Using the Attention U-Net architecture, we perform experiments on datasets from four different cohorts, the largest study to date. We segment BAT regions using predicted [18F]-FDG uptake values, achieving 23% to 40% better accuracy than conventional CT thresholding. Additionally, BAT volumes computed from the segmentations distinguish the subjects with and without active BAT with an AUC of 0.8, compared to 0.6 for CT thresholding. These findings suggest CNNs can facilitate large-scale imaging studies more efficiently and cost-effectively using only CT.

Abstract Glucagon‐like peptide‐1 receptor (GLP‐1R) agonists induce weight loss in patients with type 2 diabetes mellitus (T2DM), but the underlying mechanism is unclear. Recently, the mechanism by which metformin induces weight loss could be explained by an increase in growth differentiation factor 15 (GDF15), which suppresses appetite. Therefore, we aimed to investigate whether the GLP‐1R agonist liraglutide modifies plasma GDF15 levels in patients with T2DM. GDF15 levels were measured in plasma samples obtained from Dutch Europids and Dutch South Asians with T2DM before and after 26 weeks of treatment with daily liraglutide (n = 44) or placebo (n = 50) added to standard care. At baseline, circulating GDF15 levels did not differ between South Asians and Europids with T2DM. Treatment with liraglutide, compared to placebo, decreased body weight, but did not modify plasma GDF15 levels in all patients, or when data were split by ethnicity. Also, the change in plasma GDF15 levels after treatment with liraglutide did not correlate with changes in body weight or HbA1c levels. In addition, the dose of metformin used did not correlate with baseline plasma GDF15 levels. Compared to placebo, liraglutide treatment for 26 weeks does not modify plasma GDF15 levels in Dutch Europid or South Asian patients with T2DM. Thus, the weight loss induced by liraglutide is likely explained by other mechanisms beyond the GDF15 pathway. Highlights What is the central question of this study? Growth differentiation factor 15 (GDF15) suppresses appetite and is increased by metformin: does the GLP‐1R agonist liraglutide modify plasma GDF15 levels in patients with type 2 diabetes mellitus (T2DM)? What is the main finding and its importance? Plasma GDF15 levels did not differ between South Asians and Europids with T2DM and were not modified by 26 weeks of liraglutide in either ethnicity. Moreover, there was no correlation between the changes in plasma GDF15 levels and dosage of metformin administered, changes in body weight or HbA1c levels. The appetite‐suppressing effect of liraglutide is likely exerted via pathways other than GDF15.

Objectives: Cold exposure is linked to cardiometabolic benefits. Cold activates brown adipose tissue (BAT), increases energy expenditure, and induces secretion of the hormones fibroblast growth factor 21 (FGF21) and growth differentiation factor 15 (GDF15). The cold-induced increase in energy expenditure exhibits a diurnal rhythm in men. Therefore, we aimed to investigate the effect of cold exposure on serum FGF21 and GDF15 levels in humans and whether cold-induced changes in FGF21 and GDF15 levels differ between morning and evening in males and females. Method: In this randomized cross-over study, serum FGF21 and GDF15 levels were measured in healthy lean males (n = 12) and females (n = 12) before, during, and after 90 min of stable cold exposure in the morning (07:45 h) and evening (19:45 h) with a 1-day washout period in between. Results: Cold exposure increased FGF21 levels in the evening compared to the morning both in males (+61% vs −13%; P < 0.001) and in females (+58% vs +8%; P < 0.001). In contrast, cold exposure did not significantly modify serum GDF15 levels, and no diurnal variation was found. Changes in FGF21 and GDF15 levels did not correlate with changes in cold-induced energy expenditure in the morning and evening. Conclusion: Cold exposure increased serum FGF21 levels in the evening, but not in the morning, in both males and females. GDF15 levels were not affected by cold exposure. Thus, this study suggests that the timing of cold exposure may influence cold-induced changes in FGF21 levels but not GDF15 levels and seems to be independent of changes in energy expenditure.

Summary: Background: There is a subset of individuals with overweight/obesity characterized by a lower risk of cardiometabolic complications, the so-called metabolically healthy overweight/obesity (MHOO) phenotype. Despite the relatively higher levels of subcutaneous adipose tissue and lower visceral adipose tissue observed in individuals with MHOO than individuals with metabolically unhealthy overweight/obesity (MUOO), little is known about the differences in brown adipose tissue (BAT). Methods: This study included 53 young adults (28 women) with a body mass index (BMI) ≥25 kg/m2 which were classified as MHOO (n = 34) or MUOO (n = 19). BAT was assessed through a static 18F-FDG positron emission tomography/computed tomography scan after a 2-h personalized cooling protocol. Energy expenditure, skin temperature, and thermal perception were assessed during a standardized mixed meal test (3.5 h) and a 1-h personalized cold exposure. Body composition was assessed by dual-energy x-ray absorptiometry, energy intake was determined during an ad libitum meal test and dietary recalls, and physical activity levels were determined by a wrist-worn accelerometer. Findings: Participants with MHOO presented higher BAT volume (+124%, P = 0.008), SUVmean (+63%, P = 0.001), and SUVpeak (+133%, P = 0.003) than MUOO, despite having similar BAT mean radiodensity (P = 0.354). In addition, individuals with MHOO exhibited marginally higher meal-induced thermogenesis (P = 0.096) and cold-induced thermogenesis (+158%, P = 0.050). Moreover, MHOO participants showed higher supraclavicular skin temperature than MUOO during the first hour of the postprandial period and during the cold exposure, while no statistically significant differences were observed in other skin temperature parameters. We observed no statistically significant differences between MHOO and MUOO in thermal perception, body composition, outdoor ambient temperature exposure, resting metabolic rate, energy intake, or physical activity levels. Interpretation: Adults with MHOO present higher BAT volume and activity than MUOO. The higher meal- and cold-induced thermogenesis and cold-induced supraclavicular skin temperature are compatible with a higher BAT activity. Overall, these results suggest that BAT presence and activity might be linked to a healthier phenotype in young adults with overweight or obesity. Funding: See acknowledgments section.