Widespread high-grade prostatic intraepithelial neoplasia on prostatic needle biopsy: a significant likelihood of subsequently diagnosed adenocarcinoma.

In comparison with earlier studies, recent reports have demonstrated a lower incidence of prostate carcinoma after an initial diagnosis of high-grade prostatic intraepithelial neoplasia (HGPIN). The latter has led to a general tendency to reconsider the absolute need for a rebiopsy in this setting. The current retrospective study assesses the subsequent likelihood of identifying prostatic adenocarcinoma (PCa) in 41 patients with an initial diagnosis of "widespread" HGPIN defined as HGPIN present in 4 or more biopsy cores. All patients underwent at least 1 follow-up (F/U) sampling procedure in a period of 1 to 41 months. PCa was found in 16/41 patients (39%), all except 1 identified on the first F/U biopsy with the remaining patients diagnosed on a transurethral resection after a negative first F/U biopsy. All but 1 prostatic carcinoma diagnoses were obtained within 2 years from initial biopsy with 10 rendered within the first year. On average, prostate cancer was identified at 10.4 months (range: 1 to 36). One-fourth of all identified prostatic carcinomas were of Gleason score 7 or more. In 4 additional patients (9.7%), F/U biopsy revealed HGPIN with adjacent atypical small glands suspicious but not diagnostic of carcinoma (PINATYP). Of 41 patients, 10 (24.3%) continued to show HGPIN with the remaining 11/41 patients (26.8%) showing benign prostatic tissue. Patients >or=70 years of age at the time of initial biopsy had a statistically significant higher rate of PCa or HGPIN/PINATYP diagnosis on repeat biopsy compared with younger patients (P=0.02), with 55% of older men being diagnosed with cancer as compared with 33% in younger men. Patients with fewer cores sampled on initial biopsy were more likely to be diagnosed with carcinoma as opposed to HGPIN/PINATYP on F/U (P=0.015). Other factors such as the number of F/U procedures, serum prostate-specific antigen level before initial HGPIN biopsy, number of cores per F/U biopsy, and F/U interval length did not affect the likelihood of finding carcinoma. In summary, our study reveals a 39% risk of finding PCa on repeat biopsies obtained after an initial diagnosis of widespread HGPIN. Our findings support the need for a repeat biopsy in this subset of patients.