Purpose: Androgen receptors are present in virtually all epithelial cells of the prostate, including benign epithelium, high grade prostatic intraepithelial neoplasia and cancer. However, there have been variable results regarding the clinical significance of cells expressing androgen receptors in prostate cancer. We evaluated the predictive accuracy of androgen receptor expression in prostatic intraepithelial neoplasia and cancer for clinical progression and survival in patients with organ confined prostate cancer treated with radical prostatectomy.
Materials and methods: The study consisted of 172 previously untreated patients who underwent radical prostatectomy at our clinic between 1987 and 1991 with intermediate to high grade (Gleason score 6 to 9), pathological stage T2 cancer and negative surgical margins. Mean followup was 7.4 years (range 1.2 to 10.1). Mouse monoclonal anti-human androgen receptor antibody was used for immunohistochemical studies on select tissue sections from each case. We counted 100 nuclei from 3 separate areas of benign epithelium, prostatic intraepithelial neoplasia and cancer (total 300 nuclei for each diagnostic category) for each case. Mean nuclear androgen receptor expression was determined from the mean of the individual cases for each diagnostic category. Intensity was also evaluated using a subjective scale from 0 (no staining) to 3 (strong staining). We determined the correlation of clinical progression and the number of androgen receptor immunoreactive prostatic intraepithelial neoplasia or cancer nuclei, and then performed multivariate analysis which included deoxyribonucleic acid ploidy, radical prostatectomy Gleason score and preoperative serum prostate specific antigen using the Cox proportional hazards model. Progression was defined as a positive biopsy, positive bone scan or biochemical progression (postoperative serum prostate specific antigen greater than 0.2 ng./ml.).
Results: Nuclear immunoreactivity for androgen receptors was observed in all cases. Mean percent of immunoreactive nuclei was higher in benign epithelium than in prostatic intraepithelial neoplasia and cancer (56.3, 46.1 and 53.6%, respectively, pairwise comparisons p <0.05 for each pair). With rare exceptions, basal cells in benign epithelium and prostatic intraepithelial neoplasia were negative. The most intense nuclear staining was observed in benign epithelium. Immunoreactivity was also faint but detectable in the cytoplasm in prostatic intraepithelial neoplasia but not in benign epithelium or cancer. Mean number of androgen receptor immunoreactive nuclei in prostatic intraepithelial neoplasia and cancer was not a significant univariate or multivariate predictor of clinical and/or biochemical progression, or all cause survival (all p >0.05).
Conclusions: Androgen receptor expression was present in all cases of benign epithelium, prostatic intraepithelial neoplasia and cancer. The greatest extent and intensity of expression were observed in benign epithelium, with about half of the nuclei showing intense immunoreactivity. The number of androgen receptor immunoreactive nuclei in prostatic intraepithelial neoplasia and cancer in patients with organ confined prostate cancer treated with radical prostatectomy was not predictive of progression or survival.