To compare the performance of PI-RADS v2¹ scores versus quantitative ADC measures alone to discriminate high-grade from low-grade PCa. We retrospectively identified 92 treatment naive male patients with pathology proven PCa who had mpMRI examination at 3T. Tumors were pathologically classified into two groups: low-grade (Gleason score of 3+3) and high-grade (Gleason score of >3+3). A single radiologist blinded to the pathology results performed an overall PI-RADS v2 assessment. On a separate occasion, the index tumor (T) and normal (N) prostate tissue were outlined on the de-identified ADC (b 0, 1400) maps using 3D Slicer (http://slicer.org). Mean ADC and ADCratio were subsequently calculated. ADCratio was defined as ADCtumor / ADCN. We compared ADC parameters by paired-ttest, and evaluated association between PI-RADS and PCa Gleason scores via Chi-square method. Receiver operation characteristic curve plot was used to evaluate performance of ADC parameters and PI-RADS in identifying high grade PCa. Only two-sided p-value at preset significant value of 0.05 was reported. All statistical analysis was performed using STATA ( Version 11.2 StataCorp, College Station, Texas USA).There were 21 cases of low-grade and 71 cases of high-grade PCa. The majority of index lesions (85/92) were located in the peripheral zone (PZ), with only (7/92) in the transition zone (TZ). The breakdown of PI-RADS v2 scores based on Gleason pattern is presented in Figure 1. A significant association between increasing PI-RADS score and increasing Gleason score was observed (p = 0.007). Mean ADCtumor was significantly lower in areas of high-grade tumor, compared to low-grade tumor (879±183 x10-6 mm2/sec vs. 1146±163 10-6 mm2/sec, p<0.0001), as expected. ADCratio was also lower in areas of high-grade tumor compared to low-grade tumor (0.64±0.14 vs. 0.73±0.08, p=0.01). The ROC curve for ADC in differentiating high/low grade cancer in all cases revealed a significantly higher AUC for mean ADCtumor (0.86) compared to ADCratio (0.72) (p= 0.0012 ) (Figure 2). The ROC curve for overall PI-RADS v2 in differentiating high/low grade cancer revealed an AUC of 0.67. The performance of mean ADCtumor was significantly better than PI-RADS score (p= 0.0082)(Figure 2).Quantitative ADC measures may provide better discrimination between high-grade and low-grade PCa when compared to qualitative PI-RADS v2 assessment. A possible explanation for this may be that overall PI-RADS v2 score may be negatively impacted by DCE assessment, but this requires further investigation. In addition, further studies are necessary to investigate the repeatability and reliability of quantitative ADC, and as PI-RADS v2 has only recently been introduced into the clinical workflow at our institution, further study is necessary to fully evaluate its repeatability and clinical performance characteristics.