Room: Room 207
Purpose: To develop a data acquisition and reconstruction framework specialized for spectroscopic imaging of hyperpolarized 13C compounds in small animals.
Methods: A novel pulse sequence for hyperpolarized MRI was implemented in ParaVision 6.0.1 for a 7T Bruker micro-MRI system equipped with BGA12SHP gradients. This sequence allows arbitrary gradient and RF waveforms to be read from text files and played out for customized excitation, preparation and readout of non-renewable hyperpolarized signals. Dynamic or single timepoint imaging of hyperpolarized metabolites can be performed in a variety of modes not possible within the vendor pulse sequence library—including spectrally selective excitation with variable tip angles, spectroscopic readout or phase-sensitive decomposition of echo time-shifted data (IDEAL). Publicly available MATLAB code for design of spectral-spatial excitations was adapted for use with this sequence, and new tools for implementation of flyback-EPI readout gradients were created for rapid calibrationless imaging. The various acquisition modes were tested in 13C enriched phantoms and small animal experiments interrogating tumor metabolism with hyperpolarized [1-13C]pyruvate.
Results: Phantom data demonstrate that the spectroscopic imaging methods implemented achieve efficient and rapid encoding of 13C polarization with chemical shift specificity. Spectroscopic imaging of hyperpolarized [1-13C]pyruvate uptake and metabolism shows increased lactate production in orthotopic prostate and thyroid tumors, consistent with the upregulation of glycolysis characteristic of cancer metabolism.
Conclusion: Pulse sequences for imaging of metabolic hyperpolarized agents must efficiently use a non-renewable signal to encode the spatial distribution of multiple chemical shifts. This work provides a method for achieving this within the brief lifetime of hyperpolarized signals with flexibility in the method of spectroscopic and spatial encoding. Since the RF and gradient waveforms employed in this pulse sequence are read directly from text files, users can design their imaging experiment using their preferred tools and implement them without modification to the underlying sequence.
Funding Support, Disclosures, and Conflict of Interest: This work was supported in part by R01-CA211150, R01-DK105346, P30-CA016672, R21-CA178450, and CPRIT RP170366.