@phdthesis{Langhammer2018,
  author    = {Langhammer, Romy},
  title     = {Metabolomic Imaging for Human Prostate Cancer Detection using MR Spectroscopy at 7T},
  url       = {http://nbn-resolving.de/urn:nbn:de:bvb:20-opus-165772},
  school      = {Universit{\"a}t W{\"u}rzburg},
  year      = {2018},
  abstract  = {BACKGROUND. Prostate cancer (PCa) remains a major health concern in men of the Western World. However, we still lack effective diagnostic tools a) for an effective screening with both high sensitivity and specificity, b) to guide biopsies and avoid histology sampling errors and c) to predict tumor aggressiveness in order to avoid overtreatment. Therefore, a more reliable, highly cancer-specific and ideally in vivo approach is needed. The present study has been designed in order to further develop and test the method of "metabolomic imaging" using magnetic resonance spectroscopy (MRS) at 7T to address those challenges. METHODS. Thirty whole prostates with biopsy-proven PCa were in vitro analyzed with a 7T human MR scanner. A voxel grid containing the spectral information was overlaid with the MR image of the middle transverse cross-sectional plane of each case. Subsequent histopathological evaluation of the prostate specimen followed. After the spectral output was processed, all voxels were compared with a metabolomic PCa profile, which had been established within a preliminary study, in order to create a metabolomic map indicating MRS cancer-suspicious regions. Those regions were compared with the histologically identified tumor lesions regarding location. RESULTS. Sixty-one percent of the histological cancer lesions were detected by metabolomic imaging. Among the cases with PCa on the examined slice, 75\% were identified as cancerous. None of the tested features significantly differed between detected and undetected cancer lesions. A defined "Malignancy Index" (MI) significantly differentiated between MRS-suspicious lesions corresponding with a histological cancer lesion and benign lesions (p = 0.006) with an overall accuracy of 70\%. The MI furthermore showed a positive correlation with the Gleason grade (p = 0.021). CONCLUSION. A new approach within PCa diagnostics was developed with spectral analysis including the whole measureable metabolome - referred to as "metabolomics" - rather than focusing on single metabolites. The MI facilitates precise tumor detection and may additionally serve as a marker for tumor aggressiveness. Metabolomic imaging might contribute to a highly cancer-specific in vivo diagnostic protocol for PCa.},
  subject      = {Prostatakrebs},
  language  = {en}
}