Introduction

Proton magnetic resonance spectroscopy (1H-MRS) is a non-invasive technique that has been used successfully to characterize various metabolic disorders and malignancies in the brain, breast, liver, and prostate via estimations of lipids, choline, and other metabolites.¹ As for the kidney, the feasibility of renal MRS has been reported in healthy volunteers, and patients with renal cell carcinoma (RCC).² Due to the lipid-rich tumor cells of RCC, differentiation from other histological subtypes could be feasible by the detection of intracellular lipid contents. This suggests that MRS could be useful for RCC characterization and tumor grading. However, renal MRS remains technically challenging. Although single-voxel MRS can differentiate these lipid peaks in the kidney, their distribution within heterogeneous tumors cannot be mapped. Conventional magnetic resonance spectroscopic imaging (MRSI) overcomes this by providing spatially resolved spectra over many voxels, but it requires a long acquisition time. In this work, our major goal was to investigate the feasibility of a renal high-resolution density-weighted concentric ring trajectory (DW-CRT) metabolite cycling MRSI acquisition in healthy volunteers with the intent to establish the signature of the lipid composition of healthy renal tissues, which will be used as a future tool for non-invasive characterization of RCC.

Methods

In-vivo test-retest scans were performed on five healthy volunteers (four males, and one female; age: 31 ± 5 years) using a Siemens Prisma 3 Tesla scanner (Siemens Healthineers, Germany). The volunteers were asked to lie on a spine coil in a head-first supine position, and a flexible (18-channel body) coil was placed above the abdominal region. After a short break outside of the scanner, the repeat scan was acquired using the same scanning protocol. The acquisition protocol included two sequences, 1) a high-resolution structural MRI sequence [TE/TR: 80 ms/1.5 s, FA: 150°, 128 averages, resolution: 0.9×0.9×4 mm³, FOV: 280x280 mm², echo train length: 88], and 2) a density-weighted concentric ring trajectory (DW-CRT)³ metabolite cycling FID-MRSI acquisition [FOV: 240×240 mm², matrix size: 48×48, slice thickness: 10 mm, acquisition delay: 4 ms, repetition time: 1 s, TA: 192 s. DW-CRT was prescribed using a Hanning-window and the following parameters: points-per-ring: 64, temporal samples = 512, extractable voxel resolution = 5×5×10 mm³, number of rings = 24, spatial interleaves = 4, time acquire = 96 s and spectral bandwidth = 1250 Hz. To enhance B₀ homogeneity, the left kidney area was shimmed before acquiring the MRSI data. MRSI data was acquired from an axial slice that showed the same anatomy, marked by the kidney hilum (Figure 1). For assessing the quantification reliability, a region of interest (ROI) covering the medulla anatomy of the left kidney was used (Figure 2). Voxel-wise spectral fitting was performed using LCModel² to get information on each lipid component. Finally, the repeatability reproducibility of the fat fraction (FF= fat signal/total signal) was used to evaluate the reliability of the MRSI outcomes.

Results

In addition to producing images representing the structural distribution (Figure 1), MRSI provided high spectral resolution (Figure 3). The measured spectral linewidth was 24.5 ± 0.4 Hz. The differentiated lipid peaks are labeled in Figure 3. the fitted peak information was used to generate separate maps for each lipid component (Figure 4). Table 1 summarizes the reproducibility results from one subject represented through the coefficient of variance (CV%). In addition to the total FF, reproducibility evaluation was performed for each lipid component (Figure 3). The coefficient of variance of the proposed MRSI acquisition scheme was less than 1 % for FF. However, the CV of each lipid component varied due to a suboptimal choice of basis set used for LCModel fitting.

Discussion and Conclusion

The results show that our proposed MRSI sequence can generate separate high-resolution lipid distribution maps in a short scan time (~3 min). The high degree of reproducibility of water and fat fraction supports the future application of this technique for non-invasive characterization of renal malignancy.

Acknowledgements

Supported by the Indiana CTSI, funded in part by grant #UL1TR001108 from the NIH, NCATS, CTS Award.