Program Information
DCE Perfusion MRI for Renal Tumor Subtyping
Y Wu*, P Murillo , K Shah , Rutgers University, New Brunswick, NJ
Presentations
MO-L-GePD-IT-1 (Monday, July 31, 2017) 1:15 PM - 1:45 PM Room: Imaging ePoster Theater
Purpose: Tumor histologic subtype provides valuable information for treatment planning of renal tumor. Using dynamic contrast-enhanced (DCE) perfusion MRI, perfusion (blood flow at capillary level) can be measured quantitatively. In this study, we investigate capability of DCE perfusion MRI in renal tumor subtyping.
Methods: Baseline DCE perfusion MRI study was conducted on 14 patients with renal lesions, whose histologic subtype was determined from subsequent surgery.DCE images were acquired in shallow free breathing. The imaging protocol was optimized as follows: 3D SPGR sequence with minimum TR/TE, a flip angle of 12°, 8 coronal slices with 6mm thickness (ZIP to 16 slices with 3mm thickness), a FOV of 36cm with partial PE (256x192 pixels), resulting in a resolution of 2x2x3mm3. One hundred frames of images were acquired dynamically. In addition, pre-contrast images was acquired for T1 mapping with variable flip angle of 2°, 8°, 12°, 16°, and 20°.Post-processing was applied on DCE and pre-contrast images. ROIs were delineated within aorta and around tumor. From the dynamic change in signal intensity, the decrease of T1 was inferred and used to calculate contrast agent concentration. The changes in contrast agent concentration within aorta and within tumor were used to derive tumor perfusion as described by Pharmacokinetic model using nonlinear least square error fitting.
Results: Out of 14 patients, two had oncocytoma, and others had renal cell carcinoma (RCC), including six clear cell, three papillary, and three chromophobe. The mean perfusion level for each subtype of renal lesion is significantly different. Specifically, benign oncocytoma has lower perfusion than malignant RCC, where clear cell RCC has higher perfusion than non-clear cell RCC.
Conclusion: DCE perfusion MRI is promising in renal tumor subtype prediction, which could be used to guide treatment planning.
Contact Email: