Format

Send to:

Choose Destination
See comment in PubMed Commons below
Eur J Surg Oncol. 2005 Apr;31(3):309-13.

Prediction by quantitative histology of pathological stage in prostate cancer.

Author information

  • 1Urology Unit, Cannizzaro Hospital, Via Messina 829, 95126 Catania, Italy. piepepe@hotmail.com

Abstract

AIMS:

To find a predictor of extraprostatic extension in clinically localized prostate cancer (PCa), pre-operative ultrasound-guided prostate needle biopsies and clinico-pathological data were reviewed.

METHODS:

One hundred and eighty-three consecutive patients who underwent radical retropubic prostatectomy for clinical T1-T2 PCa and serum PSA <10 ng/ml were reviewed. Pre-operative biopsy was performed according to an extended protocol and whole-mount prostatectomy specimens were processed. The following biopsy variables were categorized to this analysis: Gleason score (< or =6, >6), TPC (< or =20%; >20%), GPC (< or =50%; >50%), cancer-positive cores (< or =2; >2), cancer-positive cores in both lateral portions (yes; no), PCa (monolateral; bilateral).

RESULTS:

Only 60/183 specimens showed an organ-confined PCa; the remaining ones showed pT3a in 57 cases, pT3b in 11 and pT3 with positive surgical margins in 55. A locally advanced PCa was found in 60.2 and 76.8% of T1c and T2 clinical stage, respectively. The positive predictive value and negative predictive value of biopsy findings to predict a locally advanced PCa was 89.9 and 75%, respectively. All biopsy variables associations were statistically significant; however, among these variables (non-categorized), in multivariate logistic regression analysis, only GPC was significantly associated with pathologic stage (odds ratio estimate was 1.075, 95% CI: 1.053-1.098).

CONCLUSIONS:

Quantitative histology, especially GPC, seems to be helpful for pre-operative staging of PCa in patients with T1c-T2 clinical stage and PSA < 10 ng/ml.

PMID:
15780569
[PubMed - indexed for MEDLINE]
PubMed Commons home

PubMed Commons

0 comments
How to join PubMed Commons

    Supplemental Content

    Full text links

    Icon for Elsevier Science
    Loading ...
    Write to the Help Desk