Radiologists and scientists who interpret and study prostate MRI.Restriction Spectrum Imaging (RSI) is a multiple b-value, multidirectional advanced diffusion weighted imaging (DWI) technique which aims to minimize signal derived from the extracellular hindered as well as the cylindrically restricted pools of water molecules and focus contribution from the isotropic, intracellular, truly restricted pool (1). This has been used to advantage in Neuroimaging, producing greater accuracy than conventional imaging measures in differentiating areas of glioblastoma multiforme infiltrative tumor involvement from normal appearing cerebral white matter (2). We hypothesize that this technique can perform similarly in the prostate, resulting in improved conspicuity for prostate cancer (PCa) tumors compared to conventional DWI as well as dynamic contrast enhancement (DCE) techniques, which currently serve as the mainstay for multiparametric magnetic resonance imaging (MP-MRI) of the prostate. We evaluated the ability of RSI-derived normalized z-scores to quantitatively differentiate between PCa and normal prostate compared to quantitative conventional multiparametric magnetic resonance imaging (MP-MRI) parameters.We retrospectively analyzed preoperative RSI and MP-MRI in 33 patients completed prior to prostatectomy. Imaging was performed on a 3T GE Signa HDx unit, without an endorectal coil. RSI data were collected at b = 125, 375, and 1000 s/mm2 at 6, 6 and 15 directions, respectively, with a TE of 72 ms. Conventional DWI was performed at b-values of 0 and 1000 s/mm2. RSI was corrected for spatial distortion as performed previously in the brain (2,3). Whole mount sectioning was performed on the surgical specimens, and tumor contours were defined by an experienced genitourinary pathologist. ROIs corresponding to the sites of confirmed PCa and normal prostate were drawn on T2-weighted images based on the whole mount pathology sections. RSI-derived normalized z-scores, as well as the quantitative MP-MRI parameters Ktrans (generated via Dynacad software) and ADC values were obtained for all ROIs. Receiver operating characteristic (ROC) curves were used to evaluate the sensitivity and specificity of each method for delineating tumor from benign prostatic tissue. Area under the ROC curve (AUC) was greater for RSI z-score than for ADC or Ktrans (Fig 1). Specifically, the AUC for RSI z-score, conventional ADC, and Ktrans was 93.6%, 84.8%, and 79.2% respectively. Using DeLong’s test, we find that the AUC for RSI z-score achieved statistical significance over conventional ADC (P=0.04) and Ktrans (P=0.03). Additionally, tumor is qualitatively more conspicuous on RSI “cellularity maps” (CM) relative to ADC and Ktrans maps, as can be seen in our representative examples below (Fig 2). In these examples, RSI-MRI more conspicuously demarcates PCa for targeted biopsy than T2, DCE, or conventional DWI techniques as confirmed by post-prostatectomy whole mount pathology. RSI z-score demonstrates greater sensitivity and specificity for discriminating between tumor and normal prostate than conventional ADC or Ktrans.MP-MRI, including DCE and conventional DWI, serves as the current standard of care after many years of effort to improve the detection and localization of PCa via imaging. Many studies support the added benefit of DCE and, particularly, DWI to standard anatomic T1 and T2 sequences (4,5). However, DCE requires intravenous contrast administration, resulting in the burden of longer scan time, patient discomfort, and the risk of adverse reaction. Conventional DWI is compromised by severe spatial distortion, limiting its ability to be coregistered to anatomic images, which is necessary for tumor localization. Additionally, conventional DWI fails to exclude signal from the extracellular hindered and cylindrically restricted water pools, decreasing tumor conspicuity thought to be possible with advanced DWI techniques such as RSI (1). Our data raise the possibility of a rapid RSI-based non-invasive imaging test (without intravenous contrast or an endorectal coil), to accurately discriminate between PCa and benign tissue.