Published online Apr 28, 2016. doi: 10.4329/WJR.v8.i4.397
Peer-review started: August 16, 2015
First decision: November 7, 2015
Revised: January 15, 2016
Accepted: February 14, 2016
Article in press: February 16, 2016
Published online: April 28, 2016
AIM: To investigate feasibility of a quantitative study of prostate cancer using three dimensional (3D) fiber tractography.
METHODS: In this institutional review board approved retrospective study, 24 men with biopsy proven prostate cancer underwent prostate magnetic resonance imaging (MRI) with an endorectal coil on a 1.5 T MRI scanner. Single shot echo-planar diffusion weighted images were acquired with b = 0.600 s/mm2, six gradient directions. Open-source available software TrackVis and its Diffusion Toolkit were used to generate diffusion tensor imaging (DTI) map and 3D fiber tracts. Multiple 3D spherical regions of interest were drawn over the areas of tumor and healthy prostatic parenchyma to measure tract density, apparent diffusion coefficient (ADC) and fractional anisotropy (FA), which were statistically analyzed.
RESULTS: DTI tractography showed rich fiber tract anatomy with tract heterogeneity. Mean tumor region and normal parenchymal tract densities were 2.53 and 3.37 respectively (P < 0.001). In the tumor, mean ADC was 0.0011 × 10-3 mm2/s vs 0.0014 × 10-3 mm2/s in the normal parenchyma (P < 0.001). The FA values for tumor and normal parenchyma were 0.2047 and 0.2259 respectively (P = 0.3819).
CONCLUSION: DTI tractography of the prostate is feasible and depicts congregate fibers within the gland. Tract density may offer new biomarker to distinguish tumor from normal tissue.
Core tip: Our study identified 24 men with biopsy proven prostate cancer. These patients underwent prostate magnetic resonance imaging with an endorectal coil on a 1.5 T scanner. Software was used to generate a diffusion tensor imaging (DTI) map and three dimensional fiber tracts. DTI tractography demonstrated rich fiber tract anatomy with tract heterogeneity. Tract density may represent a new biomarker to distinguish tumor from normal tissue.