Conventional metabolic imaging techniques using nuclear methods have been limited in the prostate due to the gland’s proximity to the extraction route (urinary tract). Hyperpolarized (HP) MRI provides a means of dramatically enhancing the signals of MR active nuclei, thus providing a means of utilizing MRI to visualize these molecules and their metabolic products non-invasively, in vivo and in real time [1]. Since the lifetime of the HP molecule is short (< 2min) imaging data acquired in less than 1 min has demonstrated that this method is predominantly before urinary clearance, thus providing contrast without background. In preclinical models, prostate cancer cells show increased conversion of HP pyruvate to lactate due to upregulated glycolytic metabolism [2], showcasing a potential application for HP MRSI in the clinic for distinguishing tumor from normal as well as grading prostate cancer. The feasibility and safety of HP [1-¹³C] pyruvate has recently been demonstrated utilizing a proof of concept hyperpolarizer [3]. In this work, we demonstrate the feasibility of acquiring dynamic 2D HP ¹³C spectra in humans using the SpinLab hyperpolarizer at 5T and accompanying sterile fluid-paths.Pre-surgical prostate cancer patients were imaged using a clinical prostate screening protocol followed by a dynamic 2D ¹³C spectroscopy sequence. To evaluate repeatability, patients are scanned again following a second injection. GMP [1-¹³C] pyruvic acid (Isotec) was mixed with a stable organic free radical (15mM, GLP AH11501 sodium salt, GE Healthcare) under sterile conditions and laser welded in a sterile fluid-path (GE Healthcare). Patients were injected with 0.43 mL/kg followed by a 20 mL saline flush. All MR data were acquired on a 3 T wide-bore scanner (GE Healthcare) using a clamshell transmit coil (GE Healthcare) and a dual-tuned ¹H/¹³C endorectal receiver coil (GE Healthcare). A 2D dynamic EPSI sequence was initiated 5 s following the flush, with a 4.3 s temporal resolution. The EPSI waveform was designed to acquire 16 spectra across a 16-cm FOV. Phase encoding was used to acquire 1x1x1.5 cm voxels.Three pre-surgical prostate cancer patients (64±11 years, PSA 9.6±2.6 ng/mL) have been imaged. The QC measurements for the patient doses (n=5) are: 244±4 mM pyruvate, pH 7.1±0.5, 1.9±1.3 µM free radical, 35.0±1.6°C, and 19.0±0.5% polarization; the injections occurred 44.4±10.0 s after the dissolution. Figure 1 shows data from a 71-year-old man with predominantly Gleason 3+4 in the left base of the prostate (PSA 6.05). 2D spectra from an early time point are shown overlaid on a T2-weighted image. Spectra from a single voxel (arrow) are shown at an early and late time point. In the early time point, a large pyruvate peak is visible, as well as the pyruvate hydrate peak. Later, the lactate peak is visible. The dynamics of the pyruvate deliver and its conversion to lactate are also shown. Figure 2 shows the delivery of pyruvate to the prostate for the 5 injections. When corrected for the arrival of the pyruvate in the neighboring vessels, the delivery dynamics are consistent among patients.We successfully imaged three prostate cancer patients using the SpinLab hyperpolarizer and sterile fluid-paths. We observed HP pyruvate delivery and its conversion to lactate, with higher lactate visible in the region of the tumor. No adverse effects have been reported by any of the subjects. This study will test repeatability and robustness of studying prostate cancer metabolism in patients. The dynamics we’ve observed from 5 injections in 3 patients indicate the need for bolus tracking for initiating a static acquisition. Otherwise, the pyruvate-to-lactate ratio may be affected by differences in perfusion.With the high SNR higher achievable with hyperpolarization, it will be possible to observe the delivery of pyruvate to the prostate, conversion of pyruvate to lactate. The acquisition of 2D data allows comparison of spectra with prostate cancer histopathology.