Extent of high-grade prostatic intraepithelial neoplasia is not a predictor of cancer at repeat biopsy

Objective: High-grade prostatic intraepithelial neoplasia (PIN) is a well accepted pre-cursor of invasive prostate cancer. Most investigators agree that a diagnosis of high-grade PIN warrants repeat transrectal ultrasound guided biopsy. We set out to investigate risk factors for cancer among a modern cohort of men with isolated high-grade PIN. Material and Methods: The Princess Margaret Hospital has a comprehensive database of 6249 TRUS procedures over the past 8 years. We searched this dataset for the following parameters: a) diagnosis of high-grade PIN between 1997 and 2003; b) absence of atypia or cancer and c) repeat prostate biopsy to rule out cancer. Addi-tional covariates assessed were: age, pro-state specific antigen (PSA) level at the time of biopsy, digital rectal examination (DRE) findings, transrectal ultrasound (TRUS) stage, TRUS volume, and the amount of PIN at biopsy (defined as one core vs. greater than one core by review of pathology reports). All data were tabulated and univariate (chi-square/t-tests) as well as multivariate (logistic regression) analyses were performed. All significance testing was two-sided with p<0.05 considered as significant. Results: A total of 130 patients had a diagnosis of high-grade PIN and underwent repeat biopsy. Among this cohort, 41 patients (31.5%) had cancer at re-biopsy.The Gleason sum distribution for these tumors was: 6 in 32 patients, 7 in 8 patients and 9 in 1 patient. Among the entire cohort the mean age was 64.1 years (range 37-78); PSA was 8.3 ng/ml (range: 0.28-70.2); and prostate volume was 65.43 cc (range 16-182.9). Fifteen patients (11.5%) had abnormal DRE, 53 (40.8%) had hypo-echoic lesions at TRUS, and 46 (35.4%) had more than one core high-grade PIN. On uni-variate analysis, the presence of an abnormal TRUS (22 of 52 with hypo-echoic lesions vs. 19 of 76 without hypo-echoic lesions, p=0.039) and the prostate volume (mean volume 69.66 cc for benign re-biopsies versus 56.89 cc for positive repeat biopsies, p<0.05) were significantly different between men with and without cancer at repeat biopsy. On logistic regression ana-lysis, however, only the volume remained a significant predictor of cancer (p=0.028). There was no association between Gleason score at biopsy (6 vs > 6) and the extent of PIN at first biopsy (p=0.86). Conclusions: In a modern cohort of men with high-grade PIN, PSA, DRE and age are not predictive of cancer at re-biopsy. There was no association between Gleason score at biopsy (6 vs > 6) and the extent of PIN at first biopsy. The prostate volume was the only significant predictor of cancer at re-biopsy. Résumé Objectifs: La néoplasie intra-épithéliale prostatique de haut grade (PIN) est un précurseur bien connu de cancer de la prostate invasif. La plupart des investigateurs consentent qu'un diagnostic de PIN de haut grade justifie la répétition des biopsies transrectales écho-guidées. Nous avons eu l'intention d'enquêter sur les facteurs de risque pour cancer parmi u