We analyzed the potential prognostic significance of the immunohistochemical expression of androgen and growth factor receptors determined in prostatectomy specimens of patients with prostate cancer.Materials and Methods
A cohort of 211 patients with locally confined prostate cancer treated with radical prostatectomy with or without antiandrogen pretreatment between January 1, 1990 and August 31, 1996 was observed prospectively. Prostatectomy samples were processed immunohistochemically to visualize androgen and growth factor receptors, of which immunoreaction intensity was scored relative to that of positive control tissue. Clinical postoperative data were processed using the Kaplan-Meier method, log rank test, and univariate and multivariate explorative Cox modeling to evaluate the contribution to overall and relapse-free survival.Results
There were statistical dependencies between the androgen receptor and epidermal growth factor receptor staining indexes. Following data stratification according to the epidermal growth factor receptor staining index the prognosis associated with a low androgen receptor staining index was worse than that with a higher androgen receptor staining index. Cox regression analysis for relapse-free survival confirmed that the risk factors low androgen receptor and increased epidermal growth factor receptor staining were associated with significantly increased relative risk. Univariate Kaplan-Meier analysis showed that patients with grade 3 carcinoma had a worse prognosis than those with better differentiated carcinoma, whereas antiandrogen pretreatment had no influence on overall survival or relapse-free survival.Conclusions
Using a multivariate proportional hazards regression model for data on a cohort of 211 patients with 68 showing relapse/progress or death from disease a low intensity of androgen receptor staining indicated a poor prognosis.