The current value of histological findings in negative prostate biopsies to predict the future risk of clinically significant prostate cancer.

Schwartzmann, I; Celma, A; Regis, L; Planas, J; Roche, S; de Torres, I M; Semidey, M E; Morote, J

INTRODUCTION Repeat prostate biopsy (PBx) is recommended under persistent suspicion of prostate cancer (PCa) or in the face of the following findings: atypical small acinar proliferation (ASAP), extense (≥3 biopsy sites) high-grade prostatic intraepithelial neoplasia (HGPIN), or HGPIN with atypical glands, suspicious for adenocarcinoma (PIN-ATYP). Nowadays, multiparametric magnetic resonance imaging (mpMRI) and mpMRI targeted PBx (MRI-TBx) are recommended in repeat PBx. Our objective was to analyze the current value of ASAP, mHGPIN, PIN-ATYP and other histological findings to predict clinically significant PCa (csPCa) risk. METHODS Retrospective analysis of 377 repeat PBxs. MRI-TBx was performed when Prostate Imaging-Reporting and Data System (PI-RADS) score>3 and 12-core transrectal ultrasound (TRUS) systematic PBx when≤2. ASAP, HGPIN, mHGPIN, PIN-ATYP, and 8 other histological findings were prospectively reported in negative PBx. CsPCa was defined as ISUP group grade>2. RESULTS Incidence of ASAP, multifocal HGPIN (mHGPIN) and PINATYP was 4.2%, 39.7% and 3.7% respectively, and csPCa rate was statistically similar among men with these histological findings. However, the rate of csPCa was 22.2% when proliferative inflammatory atrophy (PIA) was present, and 36.1% when it was not. PIA was the only histological finding which predicted lower risk of csPCa, with an OR of .54 (95% CI: .308-.945, P=.031). In addition, PIA was an independent predictor of a model combining clinical variables and mpMRI which reached area under de ROC curve of .86 (95% CI: .83-.90). CONCLUSIONS PIA emerged as the only predictive histological finding of csPCa risk and can contribute to a predictive model. mHGPIN failed to predict csPCa risk. The low incidence of ASAP (4.2%) and PIN-ATYP (3.7%) prevented us from drawing conclusions.