Fischer et al. EJNMMI Research (2020) 10:85 https://doi.org/10.1186/s13550-020-00665-7 ORIGINAL RESEARCH Open Access Comparison of [18F]FDG PET/CT with magnetic resonance imaging for the assessment of human brown adipose tissue activity Jonas Gabriel William Fischer1, Claudia Irene Maushart1, Anton S. Becker2, Julian Müller3, Philipp Madoerin4, Alin Chirindel5, Damian Wild5, Edwin E. G. W. ter Voert3, Oliver Bieri6, Irene Burger3 and Matthias Johannes Betz1* Abstract Background: Brown adipose tissue (BAT) is a thermogenic tissue which can generate heat in response to mild cold exposure. As it constitutes a promising target in the fight against obesity, we need reliable techniques to quantify its activity in response to therapeutic interventions. The current standard for the quantification of BAT activity is [18F]FDG PET/CT. Various sequences in magnetic resonance imaging (MRI), including those measuring its relative fat content (fat fraction), have been proposed and evaluated in small proof-of-principle studies, showing diverging results. Here, we systematically compare the predictive value of adipose tissue fat fraction measured by MRI to the results of [18F]FDG PET/CT. Methods: We analyzed the diagnostic reliability of MRI measured fat fraction (FF) for the estimation of human BAT activity in two cohorts of healthy volunteers participating in two prospective clinical trials (NCT03189511, NCT03269747). In both cohorts, BAT activity was stimulated by mild cold exposure. In cohort 1, we performed [18F]FDG PET/MRI; in cohort 2, we used [18F]FDG PET/CT followed by MRI. Fat fraction was determined by 2-point Dixon and 6-point Dixon measurement, respectively. Fat fraction values were compared to SUVmean in the corresponding tissue depot by simple linear regression. Results: In total, 33 male participants with a mean age of 23.9 years and a mean BMI of 22.8 kg/m2 were recruited. In 32 participants, active BAT was visible. On an intra-individual level, FF was significantly lower in high-SUV areas compared to low-SUV areas (cohort 1: p < 0.0001 and cohort 2: p = 0.0002). The FF of the supraclavicular adipose tissue depot was inversely related to its metabolic activity (SUVmean) in both cohorts (cohort 1: R2 = 0.18, p = 0.09 and cohort 2: R2 = 0.42, p = 0.009). Conclusion: MRI FF explains only about 40% of the variation in BAT glucose uptake. Thus, it can currently not be used to substitute [18F] FDG PET-based imaging for quantification of BAT activity. Trial registration: ClinicalTrials.gov. NCT03189511, registered on June 17, 2017, actual study start date was on May 31, 2017, retrospectively registered. NCT03269747, registered on September 01, 2017. * Correspondence: [email protected] 1Department of Endocrinology, Diabetes and Metabolism, University Hospital Basel, Petersgraben 4, 4031 Basel, Switzerland, and University of Basel, Basel, Switzerland Full list of author information is available at the end of the article © The Author(s). 2020 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. Fischer et al. EJNMMI Research (2020) 10:85 Page 2 of 12 Introduction PET/CT has several limitations: since the radiolabeled Brown adipose tissue (BAT) is a thermogenic tissue tracer [18F]FDG PET/CT is taken up preferentially by which contributes to energy homeostasis in human metabolically active tissue, BAT has to be activated prior adults. Upon cold exposure, it is activated by the sympa- to PET/CT scanning. Furthermore, it exposes individuals thetic nervous system (SNS) and converts chemical en- to ionizing radiation and is expensive [12]. Magnetic res- ergy stored as lipids within the adipocytes directly into onance imaging (MRI) has been proposed as an alterna- heat [1]. Brown adipocytes differ significantly from white tive to PET/CT [13]. The imaging properties differ adipocytes: they contain a high amount of mitochondria significantly from those of WAT since BAT contains and the intracellular triglycerides are stored in multiple more mitochondria and therefore a higher amount of small lipid droplets allowing for rapid lipolysis. The iron and a lower amount of fat [14]. Therefore, the fat mitochondria in brown adipocytes express high levels of fraction (FF) of BAT is generally lower than in WAT. uncoupling protein 1 (UCP1) which is unique to this cell Recent MRI studies described an inverse correlation of type [2]. When activated by fatty acids, UCP1 allows the tissues FF and the metabolic activity in [18F]FDG protons to flow across the inner mitochondrial mem- PET/CT [13]. However, other studies could not find brane along the proton gradient which has been built up such correlations [15–18]. by the respiratory chain. This is partly uncoupled from In this study, we aimed to quantify the predictive value ATP synthase. As the proton-motive force driving the of MRI FF for BAT by comparing it to the current ATP synthase is reduced higher amounts of ADP accu- standard imaging method which is [18F]FDG PET/CT. mulate in turn activating the citric-acid cycle and the re- spiratory chain. Thus, the activation of UCP1 reduces Materials and methods the efficiency of oxidative phosphorylation and energy Study participants stored in the mitochondrial proton-gradient is dissipated Data from healthy male volunteers participating in two as heat [3]. Activation of thermogenic adipocytes in- prospective clinical trials (cohort 1: NCT03189511, co- creases energy expenditure (EE) and facilitates uptake of hort 2: NCT03269747) was analyzed for this study. glucose and lipids into the tissue [4]. Therefore, BAT is Both studies were approved by the Ethics Committee an appealing potential therapeutic target for treatment for Northwestern and Central Switzerland (EKNZ). In- of obesity and associated metabolic diseases. Since the and exclusion criteria were identical for both studies. increase of EE results upon cold exposure, the raise in We recruited male volunteers with a BMI between 19 EE is called cold-induced thermogenesis (CIT). CIT can and 27 kg/m2 and an age between 18 and 40 years. Ex- be determined by measuring the difference EE during clusion criteria were concomitant disease such as heart-, warm and cold conditions using indirect calorimetry [5]. kidney-, or liver failure, thyroid hormone disorders, In order to reliably evaluate interventions that target en- hypersensitivity to cold (e.g., Raynaud’s syndrome), and ergy homeostasis and BAT activity, accurate techniques contraindications for MRI. Participants in cohort 1 for quantification are needed. Brown adipocytes are underwent PET/MRI scanning after 2 h of cold stimula- found in the cervical, supraclavicular, axillary, and retro- tion and 200 mg of the β3-agonist Mirabegron. Partici- peritoneal region with the cervical and supraclavicular pants in cohort 2 underwent PET-CT scanning after a regions being the most predominant [6]. In these adi- cold stimulus of 2-h duration and an MRI scan on the pose tissue depots, thermogenic adipocytes can emerge following day. The study flow for both cohorts is given from white adipocytes in response to cold stimulation. in Fig. 1. We used a screening visit to identify subjects This lineage of brown adipocytes is different from the with an amount of cold-induced thermogenesis (CIT) of classical BAT found in human newborns. Upon cold at least 5% after a cold stimulus of 2 h. stimulation, these brown-like (brown in white, brite) adi- pocytes transdifferentiate from white adipose tissue BAT stimulation protocols and image acquisition (WAT) [7]. This transitional process from white to Cohort 1 brown is highly dynamic, since conversion can be Participants arrived at the clinical research facility at 8 reverted when cold stimulus disappears [8]. This plasti- AM in a fasted state. After measurement of vital signs city of the tissue can also be observed in humans in tem- and body weight, an i.v. cannula was placed in an ante- perate climate zones in whom the amount of active BAT cubital vein and blood was sampled. Then participants is considerably higher during the cold season [9–11]. received 200 mg of Mirabegron. After 90 min, a con- Currently, BAT activity is most accurately quantified trolled cold exposure was started using a Hilotherm by [18F]fluoro-2-deoxy-D-glucose ([18F]FDG FDG) posi- cooling device (Hilotherm, Argenbühl, Germany) as de- tron emission tomography/computed tomography (PET/ scribed previously [19]. Participants wore a T-shirt and CT) as it allows to quantify and visualize and precisely shorts during the cooling process. Total cold exposure localize metabolically active BAT. However, [18F]FDG lasted 120 min. Immediately after cooling, an i.v. bolus Fischer et al. EJNMMI Research (2020) 10:85 Page 3 of 12 Fig. 1 Schematic overview of study procedures in cohorts 1 and 2 of 75 MBq of [18F]FDG was applied and static PET at-half-maximum of 5.0 mm, followed by a standard image acquisition was performed 30 min later on a GE axial filter with a three-slice kernel using relative weights Medical Signa 3 T PET/MRI Scanner. Images were ac- of 1:4:1 [20].