Prediction of tumour volume and pathological stage in radical prostatectomy specimens is not improved by taking more prostate needle-biopsy cores

Objective To determine what. if any, additional prognostic information is a vailable from the prostate needle biopsy by comparing the number of biopsy cores obtained with the pathology assessed from the radical retropubic pros tatectomy (RRP) specimen. Patients and methods The results from 135 consecutive patients who underwen t RRP at a single institution were reviewed. Needle biopsy information (num ber of cores, percentage of positive cores, laterality of the positive core s, and Gleason sum) were compared with the pathological data of the RRP spe cimen. including stage, Gleason sum and tumour volume. Patients were furthe r stratified into those with six or fewer cores (96 men) or more than six c ores (39 men). Clinical data. including biopsy information and pathological findings. were compared using univariate and multivariate models. Results Overall. univariate analysis showed that the total prostate-specifi c antigen (PSA) level. number of positive cores. bilateral positive cores a nd percentage of positive cores were directly correlated with tumour volume (P = 0.01). Also, PSA and percentage of positive cores were directly corre lated with extracapsular extension (P=0.008 and P 0.01. respectively). In t he multivariate model, the most important independent predictors of RRP tum our volume and pathological stage were the preoperative PSA level and perce ntage of cancer in the biopsy (P<0.01). There was no significant relationsh ip between the number of cores obtained and the predicted pathology of the RRP specimen. There were no differences in the number of positive cores. bi lateral positive cores or percentage tumour in the cores between men with m ore or less than six biopsies. In men with more than six core biopsies. the re was no significant increase in prognostic information for tumour volume and extracapsular extension. or a correlation between the Gleason sum on bi opsy and the RRP specimen. Taking more than six biopsies did not result in a significantly greater detection of potentially indolent tumours (defined as a tumour volume of <0.5 mL). Conclusions While taking more prostate needle biopsy cores seems to improve the detection of prostate cancer. there appears to be no major improvement in prognostic information over that gained from traditional sextant biopsi es. Furthermore. the results suggest that the percentage of positive cores is the best predictor of both pathological stage and tumour volume, from am ong the information readily available from prostate needle biopsy. Given th e variability in the number of cores obtained for diagnosis in clinical pra ctice, these results add credence to the use of the percentage of positive cores in the biopsy set, with known predictors such as PSA and Gleason scor e. into future models that attempt to predict tumour biology.