2187 CORRELATING MULTIPARAMETRIC PROSTATE MRI (MP-MRI) SUSPICION FOR PROSTATE CANCER WITH MRI/ULTRASOUND (MR/US) FUSION GUIDED BIOPSY RESULTS AND GLEASON GRADE

Abstract

INTRODUCTION AND OBJECTIVES: MRI has been suggested to improve prostate cancer detection. Herein, we aim to determine the utility of MP-MRI in predicting cancer presence and grade found on subsequent MR/US fusion guided biopsies. METHODS: Patients enrolled at the National Cancer Institute between 2007 and 2012 in a prospective trial assessing MRI/US fusion guided prostate biopsy with electromagnetic tracking were analyzed. A total of 671 cases of MP-MRI followed by MR/US fusion guided biopsies were evaluated in 583 patients including those with repeat biopsies on active surveillance. On MP-MRI, lesions were identified and scored as low, moderate, or high suspicion for prostate cancer. An MR/US fusion guided biopsy platform was used to identify and biopsy these target lesions in addition to conducting a systematic 12-core biopsy. Correlations between the highest assigned MP-MRI cancer suspicion and both presence of cancer and biopsy Gleason Score (bGS) were separately assessed per patient based upon first MR/US fusion guided biopsy session. A separate analysis was also done to assess the diagnostic yield of cancer presence and correlation with bGS on an individual lesion basis. RESULTS: Patient-based analysis revealed a significant correlation between the highest assigned MP-MRI cancer suspicion and presence of prostate cancer (p 0.0001) and bGS (p 0.0001). On multivariable analyses controlling for age, PSA, and prostate volume, MP-MRI cancer suspicion score was an independent prognosticator of bGS (p 0.0001). Parallel analysis of 1751 lesions revealed significant correlations between the individual lesion cancer suspicion score and the presence of cancer (p 0.0001) and bGS of that lesion (p 0.0001). Lesions with a high suspicion had an odds ratio (OR) of 19.4 for cancer detection vs those with low suspicion (p 0.0001). Similarly, the OR for high vs moderate and moderate vs low were 5.8 (p 0.0001) and 3.4 (p 0.0001), respectively. We also assessed MP-MRI as a diagnostic tool for bGS 6, 7, and 8 cancers separately, and ROC analysis demonstrated increasing accuracy of MP-MRI for increasing grade of disease (AUC of 0.64, 0.69, and 0.72, respectively). CONCLUSIONS: Lesions identified by MP-MRI demonstrate a high correlation with the presence of prostate cancer on targeted biopsies. Furthermore, levels of prebiopsy radiographic suspicion were highly correlated with and independently predicted higher bGS. MPMRI is a clinically useful modality in the evaluation and detection of prostate cancer, specifically for higher grade disease.