Encrypted login | home

Program Information

Potential Role of BOLD MRI in Discrimination of Aggressive Tumor Habitat in Prostate Cancer

J Ford

J Ford*, C Lopez , Y Tschudi , A Breto , K Padgett , A Pollack , R Stoyanova , University of Miami Miller School of Medicine, Miami, Florida


SU-D-207A-3 (Sunday, July 31, 2016) 2:05 PM - 3:00 PM Room: 207A

Purpose: To determine whether blood oxygenation level dependent (BOLD) MRI signal measured in prostate cancer patients, in addition to quantitative diffusion and perfusion parameters from multiparametric (mp)MRI exams, can help discriminate aggressive and/or radioresistant lesions.

Methods: Several ongoing clinical trials in our institution require mpMRI exam to determine eligibility (presence of identifiable tumor lesion on mpMRI) and prostate volumes for dose escalation. Upon consent, patients undergo fiducial markers placement and a T2*-weighted imaging at the time of CT sim to facilitate the fusion. In a retrospective analysis eleven clinical trial patients were identified who had undergone mpMRI on GE 3T magnet, followed by T2*-weighted imaging (time-period mean±SD = 48±20 days) using a consistent protocol (gradient echo, TR/TE=30/11.8ms, flip angle=12, matrix=256x256x75, voxel size=1.25x1.25x2.5mm). ROIs for prostate tumor lesions were automatically determined using ADC threshold ≤1200 μm2/s. Although the MR protocol was not intended for BOLD analysis, we utilized the T2*-weighted signal normalized to that in nearby muscle; likewise, T2-weighted lesion signal was normalized to muscle, following rigid registration of the T2 to T2* images. The ratio of these normalized signals, T2*/T2, is a measure of BOLD effect in the prostate tumors. Perfusion parameters (Ktrans, ve, kep) were also calculated.

Results: T2*/T2 (mean±SE) was found to be substantially lower for Gleason score (GS) 8&9 (0.82±0.04) compared to GS 7 (1.08±0.07). A k-means cluster analysis of T2*/T2 versus kep = Ktrans/ve revealed two distinct clusters, one with higher T2*/T2 and lower kep, containing only GS 7 lesions, and another with lower T2*/T2 and higher kep, associated with tumor aggressiveness. This latter cluster contained all GS 8&9 lesions, as well as some GS 7.

Conclusion: BOLD MRI, in addition to ADC and kep, may play a role (perhaps orthogonal to Gleason score) in identifying prostate lesions that would benefit from more aggressive radiotherapy.

Contact Email: