MP36-15\u2003TO BIOPSY, OR NOT TO BIOPSY, THAT IS THE QUESTION: RISK STRATIFICATION OF PATIENTS WITH A NEGATIVE MRI

Abstract

INTRODUCTION AND OBJECTIVES: The use of multiparametric MRI (mp-MRI) of the prostate has been shown to aid in the detection of clinically significant prostate cancer (CSPC). A subset of patients without a suspicious lesion on mp-MRI will harbor occult CSPC. Our objectives were twofold: to determine predictive utility of prostate mp-MRI and to determine the value of PSA Density (PSAD) in improving risk stratification for CSPC in patients with a negative MRI that undergo a prostate biopsy. METHODS: We performed a retrospective review of a database of 4294 patients who underwent a prostate mp-MRI from 2012 - 2016. Patients with a previous prostate cancer diagnosis were excluded. Sensitivity (Sn), specificity (SP), PPV, and NPV were calculated for mp-MRI of the prostate, with PI RADS 1 or 2 being considered negative and 3-5 considered positive. Analysis was then performed on patients with negative MRI that had a subsequent prostate biopsy. CSPC was defined as either Gleason grade group 2 or greater, or grade group 1 with more than 1/3 of total cores positive. Univariate Chi square, logistic regression and receiver operating characteristic (ROC) curve analyses were performed to evaluate the association between PSAD values and CSPC. RESULTS: For all mp-MRI in our patient cohort PPV was 54%, NPV 85%, Sn 95% and Sp 27%. Of the 1857 men with PI-RADS scores of 1 or 2 on mp-MRI, 200 men were followed by our practice and had undergone subsequent biopsy. Average age was 67 years, average PSA 7.3 ng/mL, average prostate size 71 mL, and average PSAD .13. Overall cancer detection rate on biopsy was 22.5% (45 patients) and for CSPC 8% (16 patients). On logistic regression analysis, PSAD was found to be a statistically significant predictor of CSPC (p=<.001), with the area under the ROC being .68 (Fig 1). Using a cutoff of 0.1 NPV is 96%, PPV 11%, Sn 75% and Sp 51%. Using a cutoff of 0.2 NPV is 94%, PPV 22.2%, Sn 37%, and Sp 88% and 36% PPV for any cancer on biopsy. CONCLUSIONS: Similar to prior reports, a negative MRI will miss <10% of CSPC. PSAD appears to have statistically significant prognostic value for patients with a negative MRI. Based on our data cohort, patients with negative mp-MRI and a PSAD score above 0.2 still have a 22% risk of CSPC and 36% of any cancer. Our results strongly confirm the utility of PSAD to stratify which patients with a negative prostate mp-MRI should undergo a prostate biopsy. Figure. No caption available. Source of Funding: none