Inside-Out Nonlinear Gradients to Improve Diffusion MRI

Project Summary/Abstract The lifetime risk of prostate cancer is 1 in 9 and it remains the 2nd leading cancer killer of men. Yet many patients with low grade disease receive premature or excessive treatment, which can lead to decades of incontinence and erectile dysfunction. One reason for the simultaneous over- and under- diagnosis of prostate

cancer is that current imaging methods are unreliable, which in turn make biopsies unreliable. A noninvasive imaging method with good sensitivity and specificity is one of the greatest needs in prostate cancer management, and diffusion weighted MRI (DWI) is the most promising candidate. However, prostate MRI has

extremely low image quality due to the long echo times needed for adequate diffusion encoding. To improve DWI of prostate, we introduce a new concept in MRI hardware: an accessory whose sole purpose is to provide diffusion encoding to a target organ. Unlike imaging gradients which are volume-encompassing

and stringently linear, this “inside-out” diffusion gradient is a wand placed between the upper legs which creates strong nonlinear fields at the prostate. The geometry allows for standard spatial encoding, standard receiver placement, and portability between scanners. For prostate imaging, this approach achieves a gradient that is >10-fold stronger over the region of interest,

increasing both overall SNR and sensitivity to epithelial/cellular volume, a hallmark of malignancy. The increase in field strength is predicted to double contrast in DWI of prostate, improving active surveillance, biopsy hit rates, and treatment monitoring. Taken together, the preliminary data on (1) achievable hardware

specs, (2) prostate imaging simulations, and (3) phantom experimental results make a strong case for the feasibility of this project and the significant impact it will have on prostate DWI. Approach: Aim 1 will install and characterize a custom diffusion gradient to demonstrate DWI in a prostate lesion phantom

which is consistent with conventional methods, but with higher reproducibility and CNR. Aim 2 applies the new method to prostate of healthy controls to establish positioning, PNS thresholds, accuracy, and improved image quality (precision, contrast, reproducibility and qualitative ratings) in vivo. Aim 3 will image patients with biopsy-naïve suspected lesions, which will be outlined and scored by multiple

radiologists to quantify contrast, diagnostic confidence and inter-reader variability in a realistic population. Aim 4 will acquire, in radical prostatectomy patients, coregistered presurgical in vivo MRI, ex vivo MRI of the excised prostate, and whole-mount pathology to show correlation between ADC and true Gleason score.

Together, these aims will provide strong evidence for the increased precision, contrast, diagnostic confidence, and clinical validity of DWI acquired with this hardware.