Visual estimation of tumour extent is not an independent predictor of prostate specific antigen recurrence

OBJECTIVE To analyse tumour extent as a predictor of cancer progression after radical prostatectomy (RP), using a multivariate Cox regression model, as several variables (e.g. Gleason grade and tumour stage) are well-established prognostic factors in prostate cancer but it is uncertain if the visual estimation of tumour extent (percentage of carcinoma) is an independent predictor for prostate cancer recurrence. PATIENTS AND METHODS Tumour extent was estimated in the RP specimens from 504 men with clinically localized prostate cancer; clinical follow-up data were available for 459 men. The mean (range) follow-up was 44.3 (1.5-144) months. Cancer progression was defined by the development of biochemical recurrence, local recurrence, or distant metastasis. Multivariate analysis was used to assess tumour extent as a predictor of cancer progression. RESULTS Of the 459 patients, 157 had cancer progression; the mean tumour extent was 36% and 24% in those with and without cancer progression, respectively (P < 0.001). Univariate analysis showed a significant association between the visual estimation of tumour extent and tumour stage, Gleason grade, surgical margins, extraprostatic extension, seminal vesicle invasion, lymph node metastasis, and preoperative serum prostate-specific antigen level (all P < 0.001). However, in a multivariate Cox regression model controlling for pathological stage, Gleason score, and surgical margin status, the visual estimation of tumour extent was no longer a significant predictor of cancer progression (P = 0.84). CONCLUSION The visual estimation of tumour extent was associated with various established prognostic factors for prostate cancer, and with cancer progression in a univariate analysis, but it was not a significant predictor of cancer progression in the multivariate analysis controlling for pathological stage, Gleason score, and surgical margin status.