Detection of human brown adipose tissue by MRI with hyperpolarized Xe-129 gas and validation by FDG-PET/MRI
Rosa Tamara Branca1,2, Le Zhang3,4, Alex Burant1,4, Laurence Katz5, and Andrew McCallister1,4

1Physics and Astronomy, University of North Carolina at Chapel Hill, Chapel Hill, NC, United States, 2Biomedical Research Imaging Center, Chapel Hill, NC, United States, 3Material Science, University of North Carolina at Chapel Hill, Chapel Hill, NC, United States, 4Biomedical Research Imaging Center, University of North Carolina at Chapel Hill, Chapel Hill, NC, United States, 5Emergency Medicine, University of North Carolina at Chapel Hill, Chapel Hill, NC, United States

Synopsis

Despite histological evidence that all humans have brown adipose tissue, the detection of this tissue in overweighs and obese subjects has proven to be a challenge. A recent study showed that MRI by hyperpolarized xenon gas (HP129Xe) enables the detection of this tissue in both lean and obese animal phenotype, with enhanced sensitivity in the latter with respect to the gold standard, FDG-PET. Here we demonstrate that HP129Xe gas MRI can also be used to detect human BAT with better sensitivity than FDG-PET.

Purpose

Recent studies have shown that Brown Adipose Tissue1 (BAT) plays a regulatory role in human energy balance and glucose homeostasis and may have a protective role against obesity and diabetes as it has in mice[2,3]. As new therapeutic treatments that specifically target this tissue are being developed, non-invasive detection of BAT tissue and thermogenic activity remains an unmet need, especially in obese and overweight subjects[4–6]. Recently we demonstrated that magnetic resonance with HyperPolarized 129Xe gas (HP129Xe) can be used in rodents to detect BAT mass with better sensitivity than FDG-PET for the lipid-rich and thermogenically inactive BAT of obese mice[7]. Here we present our first preliminary results on the detection of human BAT by HP129Xe MRI with validation by FDG-PET/ MRI.

Method

Six healthy volunteers, between the age of 20-53 and with a BMI ranging between 19 and 28, underwent a xenon/1H MRI scan followed by a FDG-PET/1H-MRI scan. The 129Xe/1H MRI scan was performed on a 3T TIM-Trio System (Siemens Medical Solutions, Erlangen, Germany) by using a dual tuned 129Xe/1H surface coil (8 cm diameter). Anatomical images as well as 129Xe spectra and images were acquired both at thermoneutrality and during mild cold exposure, achieved by using MR-compatible water-perfused cooling pads. For the acquisition of 129Xe spectra and images, subjects inhaled 600ml of isotopically enriched xenon (∼86% 129Xe), polarized up to 10% by using a commercial polarizer (Polarean, Inc., Research Triangle Park, NC), and held their breath for 15-20s . Spectra were collected from the beginning of the breath hold up to 1 minute, while xenon images were acquired once, after gas exhalation. The FDG-PET/1H MRI scan was performed within 24 hours from the 129Xe/1H MRI scan, on a Biograph mMR (Siemens Healthcare, Erlangen, Germany). For this study, subjects were exposed to mild cold for 1 hour before the injection of 500mCi of FDG and for 45-60 minutes before the acquisition of static FDG-PET images. 1H MRI images acquired on the 3T and on the Biograph using the same sequences and parameters were used for co-registration of the 129Xe scan with the FDG-PET scan.

Results

Figure 1 shows a FDG-PET SUV map, a axial 1H fat images, and fused PET/MRI axial image acquired on the Biograph along with a 1H fat image and 129Xe spectra acquired at 3T using the dual tuned surface coil at thermoneutrality and during cold exposure. The xenon spectrum acquired at thermoneutrality shows only the gas-phase xenon resonance, whereas the xenon spectrum acquired during cold exposure shows the appearance of two dissolved-phase resonances: a narrow peak around 190ppm, which lasted for several seconds after gas exhalation, corresponding to xenon dissolved in lipids, and a broader peak around 200ppm, corresponding to xenon dissolved in blood plasma, which disappeared few seconds after exhalation. Comparison of the 1H images acquired on the 3T and those acquired on the PET/MRI scanner clearly shows that the sensitive region of the surface coil from which 129Xe spectra were acquired contains metabolically active BAT. Figure 2 shows a HP129Xe image, acquired several seconds after gas exhalation, fused to the corresponding anatomical 1H image, along with the fused FDG-PET/1H image. This image shows a clear selective uptake of xenon in the supraclavicular fat pads, which also shows enhanced glucose uptake.

Discussion

Our preliminary studies in humans mirror the results obtained in mice[7] in that they show a specific and strong increase in the downstream accumulation of inhaled HP129Xe in BAT during stimulation of thermogenesis by cold exposure. The increase in xenon uptake by BAT is such that imaging of BAT by HP129Xe MRI became feasible. The presence of a blood-dissolved peak during cold exposure reflects the expected increase in tissue perfusion during thermogenic activity, which leads to the increased accumulation of 129Xe into BAT. The increase in the dissolved–phase signal was seen only during cold exposure and in all subject analyzed, despite two of them turning up BAT-negative on FDG-PET scans. On the other hand, the relative intensity of blood/lipid peak varied greatly across our subjects and seemed to be correlated with tissue hydration and perfusion.

Conclusions

We have demonstrated detection of BAT by 129Xe/MRI in humans. As already demonstrated in mice, this technique seems to be more sensitive than FDG-PET to the lipid rich and thermogenically less active BAT characteristic of obese phenotype, underlying the well known limitations of FDG-PET scans for the detection of human BAT and finally enabling the detection of this tissue in overweight and perhaps obese subjects, the target population for anti-obesity therapies that aim to target BAT.

Acknowledgements

This work was supported by the North Carolina National Center for Advancing Translational Sciences (NCATS), through NIH grant award UL1TR001111, and by the NIDDK through grant award R01 DK108231

References

1. Cannon, B. & Nedergaard, J. Brown adipose tissue: function and physiological significance. Physiol Rev 84, 277–359 (2004). 2. Ouellet, V. et al. Brown adipose tissue oxidative metabolism contributes to energy expenditure during acute cold exposure in humans. J. Clin. Invest. 122, 545–552 (2012). 3. Chechi, K., Nedergaard, J. & Richard, D. Brown adipose tissue as an anti-obesity tissue in humans. Obes. Rev. 15, 1–15 (2013). 4. Matsushita, M. et al. Impact of brown adipose tissue on body fatness and glucose metabolism in healthy humans. Int. J. Obes. (Lond). 38, 1–6 (2013). 5. Vijgen, G. H. E. J. et al. Brown adipose tissue in morbidly obese subjects. PLoS One 6, 2–7 (2011). 6. Cypess, A. M., Haft, C. R., Laughlin, M. R. & Hu, H. H. Brown fat in humans: consensus points and experimental guidelines. Cell Metab. 20, 408–15 (2014). 7. Branca, R. T. et al. Detection of brown adipose tissue and thermogenic activity in mice by hyperpolarized xenon MRI. Proc. Natl. Acad. Sci. 111, 18001–18006 (2014).

Figures

SUV maximum intensity projection map (a), fused PET/MRI axial image (c), 1H Fat-image (d), and fused PET/1H Fat image (e) acquired on the PET/MRI scanner. These images show increased glucose uptake in the supraclavicular fat pad and the location of the surface coil with which the 1H fat image (f) and the HP129Xe spectra (b) were acquired on the MRI scanner.

Fused FDG-PET/1H MR image (a) and 129Xe/1H MR image of human BAT (b). The fused 129Xe/1H MR image shows increased xenon signal from the supraclavicular fat pad, which also shows increased glucose uptake.



Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)
1054