Phosphorus Magnetic Resonance Spectroscopic Imaging At 7 T In Patients With Prostate Cancer

Objectives: The aim of this study was to identify characteristics of phosphorus (P-31) spectra of the human prostate and to investigate changes of individual phospholipid metabolites in prostate cancer through in vivo P-31 magnetic resonance spectroscopic imaging (MRSI) at 7 T.Materials and Methods: In this institutional review board-approved study, 15 patients with biopsy-proven prostate cancer underwent T-2-weighted magnetic resonance imaging and 3-dimensional P-31 MRSI at 7 T. Voxels were selected at the tumor location, in normal-appearing peripheral zone tissue, normal-appearing transition zone tissue, and in the base of the prostate close to the seminal vesicles. Phosphorus metabolite ratios were determined and compared between tissue types.Results: Signals of phosphoethanolamine (PE) and phosphocholine (PC) were present and well resolved in most P-31 spectra in the prostate. Glycerophosphocholine signals were observable in 43% of the voxels in malignant tissue, but in only 10% of the voxels in normal-appearing tissue away from the seminal vesicles. In many spectra, independent of tissue type, 2 peaks resonated in the chemical shift range of inorganic phosphate, possibly representing 2 separate pH compartments. The PC/PE ratio in the seminal vesicles was highly elevated compared with the prostate in 5 patients. A considerable overlap of P-31 metabolite ratios was found between prostate cancer and normal-appearing prostate tissue, preventing direct discrimination of these tissues. The only 2 patients with high Gleason scores tumors (>= 4+5) presented with high PC and glycerophosphocholine levels in their cancer lesions.Conclusions: Phosphorus MRSI at 7 T shows distinct features of phospholipid metabolites in the prostate gland and its surrounding structures. In this exploratory study, no differences in P-31 metabolite ratios were observed between prostate cancer and normal-appearing prostate tissue possibly because of the partial volume effects of small tumor foci in large MRSI voxels.