Background

Hyperpolarized (HP) [1-¹³C]pyruvate imaging is rapidly being translated to humans. Detection of metabolic products such as [1-¹³C]lactate, [¹³C]bicarbonate and [1-¹³C]alanine from HP pyruvate provides direct access to metabolic conditions in organs of interest.^1–3 Although the T₂s of ¹³C-labeled pyruvate and the products are reportedly long,^4,5 the in vivo T₂*s, which are critical for gradient-recalled echo (GRE)-based data acquisitions, are yet poorly understood. Previously, we implemented a metabolite-selective multi-echo spiral imaging (MESI) sequence for measuring T₂*s of in vivo HP signals in rats.⁶ In this study, we compared T₂*s of HP [1-¹³C]pyruvate, [1-¹³C]lactate, [1-¹³C]alanine, and [¹³C]bicarbonate in human heart, brain, kidney, and spleen of healthy subjects.

Methods

The MESI acquisition consists of a spectral-spatial RF pulse and subsequent multiple short spiral readouts. In this study, 4 healthy subjects were recruited and imaged by the MESI for HP [1-¹³C]pyruvate imaging of heart (female, 38 y.o.), brain (female, 19 y.o.), kidney (male, 33 y.o.) and spleen (female, 57 y.o.), respectively. All the data were acquired using a 3T 750w wide-bore MR scanner and a SPINlab polarizer (GE Healthcare, Waukesha, WI USA). For heart, kidney and spleen imaging, a two-loop ¹³C transmit-receive Helmholtz coil was used (diameter = 20 cm; PulseTeq Limited, Chobham, Surrey, UK). Brain imaging was conducted with a ¹³C/¹H dual-frequency birdcage RF coil (Clinical MR Solutions, Brookfield, WI USA).⁷ GMP-grade [1-¹³C]pyruvic acid (Sigma Aldrich, St. Louis, MO USA) was prepared in clinical fluid paths (0.40 mL/kg body weight of 250-mM HP [1-¹³C]pyruvate solution). After 3 – 4 hrs of polarization, the dissolved HP [1-¹³C]pyruvate solution was validated by the QC and administered intravenously to subjects (injection rate = 5 mL/s), followed by a 25-mL saline flush. Multi-echo images were acquired using the MESI sequence in the order of [¹³C]bicarbonate, [1-¹³C]lactate, [1-¹³C]pyruvate and [1-¹³C]alanine (kidney only) at each timepoint. Field-of-view (FOV), nominal spatial resolution and #echoes were prescribed differently for each organ: FOV = 40 cm × 40 cm (heart, kidney and spleen) and 24 cm × 24 cm (brain), spatial resolution = 1.6 cm × 1.6 cm (heart), 1.0 cm × 1.0 cm (kidney and spleen) and 1.5 cm × 1.5 cm (brain), # echoes = 6 (heart), 3 (kidney and spleen) and 8 (brain). For all organs, slice thickness was set to 30 mm and flip angles were 90° for products and 10° for pyruvate. For brain, kidney and spleen, 16 timepoints were acquired with temporal resolution of 5 s with 15 s of delay between the start of HP injection and the start of data acquisition. For cardiac imaging, the subject was instructed to hold their breath in expiration for ~20 s, followed by a shallow breathing to minimize the respiratory motion, and the images were acquired with ECG gating. T₂*s of HP metabolites from each timepoint were calculated by fitting the decay rate of the spatially-averaged signal within structurally identified tissue compartments (heart: left ventricle (LV), right ventricle (RV) and myocardium (Myo); brain: grey matter (GM), white matter (WM) and cerebrospinal fluid (CSF); kidney: left kidney and right kidney) along the echo times as mono-exponential functions using MATLAB, and at least the first 3 echoes were used for the fitting depending on the signal to noise ratios (SNRs).

Results and Discussion

Fig. 1 shows the T₂* measurement of HP ¹³C-labeled metabolites for human heart in a short-axis ventricle view. The multi-echo images of each metabolite were fitted along echo times mono-exponentially in LV, RV and Myo to measure the T₂*s. Similarly, the results from T₂* measurement of the in vivo HP ¹³C-labeled metabolites in brain, kidney and spleen are shown in Fig. 2, Fig. 3 and Fig. 4, respectively. Time-averaged T₂*s of HP metabolites for each organ are summarized in Table 1. For all of the ¹³C-labeled metabolites, T₂*s measured from different timepoint for each organ were consistent along time (coefficients of variation < 0.20). [1-¹³C]Pyruvate exhibited the longest T₂* in all these organs, ranging from 36.3 ms to 142.6 ms. The T₂* of in vivo [1-¹³C]lactate also varied in different organs from 27.5 ms to 56.7 ms. Due to the low SNR, the T₂*s of HP [¹³C]bicarbonate could be measured only from Myo (60.8±3.9 ms) in heart and WM (46.5 ms) in brain, and T₂* of [1-¹³C]alanine was measured only in the left kidney (51.6 ms). The largely varying T₂*s of HP ¹³C-metabolites in different organs/compartments are likely due to various contributing factors such as off-resonance effects and blood flow. In future studies, more subjects will be recruited to investigate the in vivo T₂*s of HP ¹³C-labeled metabolites in each organ, with acquisition of additional information such as B₀ map and perfusion.

Conclusion

In summary, we measured the in vivo T₂*s of HP [¹³C]bicarbonate, [1-¹³C]lactate, [1-¹³C]pyruvate and [1-¹³C]alanine from human heart, brain, kidney, and spleen. The T₂*s of ¹³C-labeled signals will provide insights of in vivo pyruvate metabolism and play critical roles in designing k-space readout trajectories and quantifying the HP metabolites.