0724

First Hyperpolarized [2-13C]Pyruvate MR Studies of Human Brain Metabolism
Brian T Chung1,2, Hsin-Yu Chen1, Jeremy Gordon1, Daniele Mammoli1, Renuka Sriram1, Adam Autry1, Lydia Le Page1,3, Myriam Chaumeil1,3, Peter Shin1, James Slater1, Chou T Tan4, Chris Suszczynski4, Susan Chang5, Robert Bok1, Sabrina Ronen1,2, Peder EZ Larson1,2, John Kurhanewicz1,2, and Daniel B Vigneron1,2

1Department of Radiology and Biomedical Imaging, University of California San Francisco, San Francisco, CA, United States, 2UCSF – UC Berkeley Graduate Program in Bioengineering, University of California, San Francisco, CA, United States, 3Department of Physical Therapy and Rehabilitation Science, University of California San Francisco, San Francisco, CA, United States, 4ISOTEC Stable Isotope Division, MilliporeSigma, Merck KGaA, Miamisburg, OH, United States, 5Department of Neurological Surgery, University of California San Francisco, San Francisco, CA, United States

Synopsis

We investigated hyperpolarized (HP) [2-13C]pyruvate conversion to [2-13C]lactate and [5-13C]glutamate for the first time in the healthy human brain, with a focus on the development of hyperpolarization and preparation techniques for sterile [2-13C]pyruvate with FDA-IND & IRB approval. HP [2-13C]pyruvate, [2-13C]lactate, [5-13C]glutamate and other metabolites were successfully observed and quantitatively measured for the first time in four volunteers, and initial EPI studies confirmed a feasibility of imaging [2-13C]pyruvate to [5-13C]glutamate conversion, demonstrating a significant first step for HP metabolic imaging to diagnose and detect early stage neurological disorders.

Motivation

Dissolution Dynamic Nuclear Polarization (dDNP) provides over 10,000 fold signal enhancement for carbon-13 MRI, enabling an emerging stable-isotope molecular imaging technique for preclinical and recently clinical research studies1. Hyperpolarized (HP) 13C-pyruvate MR metabolic imaging is presently applied to identify tumor metabolism, assess aggressiveness, evaluate treatment response, and probe organ function2,3. The investigation of HP [1-13C]pyruvate conversion to [1-13C]lactate catalyzed by lactate dehydrogenase (LDH) demonstrated clinical potential spotlighting the hallmark Warburg Effect in tumors with greatly upregulated LDH activity in cancer patients and successful translation to Phase I trials4,5. In approaching the tricarboxylic acid (TCA) cycle however, [1-13C]pyruvate can also be enzymatically metabolized by pyruvate dehydrogenase and thereby converted to 13CO2, preventing direct detection of downstream metabolites. Prior animal studies using HP pyruvate labeled in the 2-position ([2-13C]pyruvate) have successfully shown direct detection as the HP 13C atoms are metabolized into Acetyl-CoA and then onto the TCA cycle and/or acetyl-carnitine and other molecules2,3. Therefore, HP [2-13C]pyruvate provides new metabolic information from its unique position atop multiple anapleurotic and catapleurotic metabolic cascades in the TCA cycle with known fast conversions. The goal of this study was to develop methods for the hyperpolarization and preparation of sterile [2-13C]pyruvate with FDA-IND and IRB approval for first-ever human studies. We sought to investigate HP [2-13C]pyruvate conversion to [2-13C]lactate and [5-13C]glutamate in the healthy brain in four human volunteers, demonstrating a significant first step for HP metabolic imaging to diagnose and detect early stage neurological disorders.

Methods

[2-13C]pyruvate was produced and supplied by MilliporeSigma Isotec Stable Isotopes (Miamisburg, OH) following Good Manufacturing Practices (GMP) for its first-ever use in human HP MR studies. A 5 Tesla GE SPINlab polarizer was used to hyperpolarize [2-13C]pyruvate prior to injection. A 400 μsec hard pulse excitation provided an approximately 2.5 kHz bandwidth, with a nominal flip angle of 40° at a center frequency of about 141 ppm, calibrated using a built-in urea phantom on a 32-channel array receiver. The [2-13C]pyruvate, [5-13C]glutamate and [2-13C]lactate doublet resonances saw 7°, 30°, 5° and 2.1° flip angles respectively. The acquisition used temporal and spectral resolutions of 2 seconds and 2.4 Hz. The 32-channel data was combined using a phase-sensitive summation followed by line broadening of 5 Hz.

Results

HP [2-13C]pyruvate, [2-13C]lactate, [5-13C]glutamate and other metabolites were successfully observed and quantitatively measured for the first time in four volunteers. T1 values for [2-13C]pyruvate known to be shorter than [1-13C]pyruvate were determined from independent solid-state buildup measurements. Polarization levels were back-calculated from the time of dissolution given exponential T1 decay. In humans, dynamic spectroscopic data was collected and summed yielding kinetic rates and curves. Quantitative post-processing analysis was performed across MestReNova (Santiago de Compostela, Spain) and MATLAB (Natick, MA). Peak identifications were assigned following those by Park et al. from their study of HP [2-13C]pyruvate in the murine brain6.

Discussion

Dynamic conversion of HP [2-13C]pyruvate to [2-13C]lactate and [5-13C]glutamate from each volunteer were represented with kinetic traces. Area-under-curve (AUC) metabolite ratios and [2-13C]pyruvate to [2-13C]lactate conversion rates (kPL) from nonlinear models were calculated and found in accordance with prior [1-13C]pyruvate data. Studies demonstrating a feasibility of imaging spatially localized metabolism of HP [2-13C]pyruvate to [5-13C]glutamate using an Echo Planar Imaging sequence were performed with histologically-defined regions transposed onto 1H data.

Conclusion

We developed methods for the hyperpolarization and preparation of sterile [2-13C]pyruvate with FDA-IND and IRB approval for first-ever human studies. Using a 32-channel 13C-headcoil, MR spectroscopy was acquired following injection of HP [2-13C]pyruvate in four human volunteers. We were able to detect the dynamic conversion of HP [2-13C]pyruvate to [2-13C]lactate, [5-13C]glutamate and other compounds in the normal brain, demonstrating a significant first step for HP metabolic imaging to diagnose and detect early stage neurodisorders.

HP metabolic information can be linked with other modalities such as functional and diffusion MRI to build increasingly comprehensive representations of neural function, structure and metabolism. Centrality metrics processing higher-order descriptors of multivalued metabolite kinetics with advances in machine learning may elucidate new methods of detecting early stage neurodisorders.

Acknowledgements

Special thanks to the UCSF Mission Bay Surbeck Lab for Advanced Imaging, NIH grant P41EB013598, NICO grant, and the UC Berkeley – UCSF Graduate Program in Bioengineering for outstanding support.

References

[1] Ardenkjær-Larsen JH, Fridlund B, Gram A, et al. Increase in signal-to-noise ratio of > 10,000 times in liquid-state NMR. Proceedings of the National Academy of Sciences Sep. 2003;100(18):10158-10163.

[2] Schroeder MA, Atherton HJ, Dodd MS, et al. The cycling of acetyl-coenzyme A through acetylcarnitine buffers cardiac substrate supply: a hyperpolarized 13C magnetic resonance study. Circ Cardiovasc Imaging. 2012 Mar;5(2):201-9.

[3] Schroeder MA, Atherton HJ, Ball DR, et al. Real-time assessment of Krebs cycle metabolism using hyperpolarized 13C magnetic resonance spectroscopy. FASEB J. 2009 Aug;23(8):2529-38.

[4] Albers MJ, Bok R, Chen AP, et al. Hyperpolarized 13C lactate, pyruvate, and alanine: noninvasive biomarkers for prostate cancer detection and grading. Cancer Res. 2008 Oct 15;68(20):8607-15.

[5] Brindle KM, Bohndiek SE, Gallagher FA, et al. Tumor imaging using hyperpolarized 13C magnetic resonance spectroscopy. Magn Reson Med. 2011 Aug;66(2):505-19.

[6] Park JM, Josan S, Grafendorfer T, et al. Measuring mitochondrial metabolism in rat brain in vivo using MR Spectroscopy of hyperpolarized [2-¹³C]pyruvate. NMR Biomed. 2013 Oct;26(10):1197-203.

Figures

Human volunteer summed spectra and dynamic kinetics of [2-13C]pyruvate conversion to [2-13C]lactate and [5-13C]glutamate with temporal resolution = 2 seconds.

[2-13C]pyruvate to [2-13C]lactate conversion rates (kPL) were calculated and found in accordance with prior [1-13C]pyruvate data (May 2017). The measured pyruvate magnetization serves as the input for the kinetic model, and lactate magnetization is fit based on a minimization of the least-squares error computed with a constrained, non-linear solver given relaxation and loss constants. [Methods available from the Hyperpolarized MRI Toolbox via the Hyperpolarized Technology Resource Center training and dissemination website: https://radiology.ucsf.edu/research/labs/hyperpolarized-mri-tech/training and https://github.com/LarsonLab/hyperpolarized-mri-toolbox]

Results of an EPI sequence study demonstrating a feasibility of imaging spatially localized metabolism of HP [2-13C]pyruvate to [5-13C]glutamate (TR = 3 seconds, 2.5 x 2.5 cm2 in-plane resolution, 5 cm slice thickness, flip angle = 10 deg for pyruvate and 60 deg for glutamate). Histologically defined regions were transposed onto 1H data with strongest glutamate signals recorded from the cortex.

Proc. Intl. Soc. Mag. Reson. Med. 27 (2019)
0724