PURPOSE: Angiogenesis is an important part in tumor progression and intratumoral microvessel density (MVD) has proven a prognostic factor in several solid tumors. However, its value in prostate cancer is still unclear, especially in small biopsy samples. We evaluated the prognostic potential of MVD in a large, homogeneous cohort of prostate cancers and its correlation with other pathologic parameters. METHODS: We used a tissue microarray (TMA) containing samples of 3,261 prostatectomy specimens from patients with prostate cancer. MVD was determined by counting vessels in a blinded fashion after immunohistochemical staining with an antibody against CD31. The results were compared with pre- and postoperative clinical and pathological parameters and clinical follow-up data. RESULTS: MVD was higher in TMA spots containing cancer as compared to benign tissue (P <0.001). There was no significant correlation of MVD with preoperative parameters, but with pathological T-classification and Gleason score (P <0.001, each) of the prostatectomy specimens. Furthermore, a higher MVD correlated with tumor location in the peripheral zone (P = 0.01). Follow-up data were available for 1,521 patients. In a univariable analysis, an MVD of >/=36 per spot was a significant predictor of PSA recurrence after radical prostatectomy (P = 0.03). However, it failed to provide an independent prognostic factor when combined with standard predictors in a multivariable analysis. CONCLUSIONS: MVD in prostate cancer is closely related to other factors contributing to tumor aggressiveness. However, the implementation of this parameter into routinely performed pathological reports does not seem to be warranted.
PURPOSE: Angiogenesis is an important part in tumor progression and intratumoral microvessel density (MVD) has proven a prognostic factor in several solid tumors. However, its value in prostate cancer is still unclear, especially in small biopsy samples. We evaluated the prognostic potential of MVD in a large, homogeneous cohort of prostate cancers and its correlation with other pathologic parameters. METHODS: We used a tissue microarray (TMA) containing samples of 3,261 prostatectomy specimens from patients with prostate cancer. MVD was determined by counting vessels in a blinded fashion after immunohistochemical staining with an antibody against CD31. The results were compared with pre- and postoperative clinical and pathological parameters and clinical follow-up data. RESULTS: MVD was higher in TMA spots containing cancer as compared to benign tissue (P <0.001). There was no significant correlation of MVD with preoperative parameters, but with pathological T-classification and Gleason score (P <0.001, each) of the prostatectomy specimens. Furthermore, a higher MVD correlated with tumor location in the peripheral zone (P = 0.01). Follow-up data were available for 1,521 patients. In a univariable analysis, an MVD of >/=36 per spot was a significant predictor of PSA recurrence after radical prostatectomy (P = 0.03). However, it failed to provide an independent prognostic factor when combined with standard predictors in a multivariable analysis. CONCLUSIONS: MVD in prostate cancer is closely related to other factors contributing to tumor aggressiveness. However, the implementation of this parameter into routinely performed pathological reports does not seem to be warranted.