Microvessel density in prostate cancer: lack of correlation with tumor grade, pathologic stage, and clinical outcome.

OBJECTIVES Angiogenesis is believed to play an important role in tumor progression and metastasis. Previous studies have suggested that the microvessel density (MVD) of prostate tumors may be of prognostic value. This study investigated the reliability of assessing MVD in radical prostatectomy specimens and its value as an independent prognostic indicator in men with clinically localized prostate cancer. METHODS One hundred radical prostatectomy specimens from 1993 to 1995 were randomly selected for this study. Thirteen cases were excluded because the patients had undergone neoadjuvant hormonal therapy or tissue blocks were unavailable. The median follow-up time was 36 months. Tumor blocks were immunostained using the endothelial-specific antibody CD31. MVD was counted in areas with the greatest microvessel immunostaining, which were designated "hot spots." MVD was analyzed for associations with clinical and pathologic factors. In a subset of 60 cases, the same observer repeated the counts three times. RESULTS Intraobserver reliability for MVD counting was excellent (reliability coefficient 0.82), demonstrating that this method could be reproduced by a single observer. MVD was not associated with Gleason sum, tumor stage, surgical margin status, or seminal vesicle invasion. Of the 87 patients, 20 (23%) had a prostate-specific antigen (PSA) failure during a 36-month median follow-up time. As expected, Gleason sum and tumor stage were strong predictors of PSA failure, with risk ratios of 2.1 and 2.3, respectively. In contrast, MVD was not associated with PSA failure. CONCLUSIONS MVD, as determined by CD31, can be reliably measured by a single observer, but it is not a useful prognostic indicator for men with clinically localized prostate cancer.