Background: Knowledge of androgen receptor (AR) content could help predict hormone response and disease course in prostate cancer. However, determination of AR by biochemical assay is difficult. An immunohistochemical assay (ICA) would solve most difficulties and be especially useful if it could be performed on paraffinized tissue.
Experimental design: AR was studied in paraffin sections from 90 men for whom endocrine response, survival (except one case), and/or biochemical AR was known. After Ag retrieval in a microwave oven, a polyclonal anti-AR Ab was used with the peroxidase antiperoxidase method. Results were semiquantified using a Histoscore (Hscore) and were correlated with biochemistry, endocrine response, and survival.
Results: Only 15 patients were AR-negative. AR-ICA did not correlate with biochemistry, Gleason score, stage, or ethnicity but did correlate with endocrine response and survival. The average Histoscore was significantly lower in patients with progressive disease (p < 0.05). In a Cox's regression analysis of survival (mean follow-up = 30 months) AR-ICA was a significant predictor (p = 0.015). Risk of death was 2.5 times greater for a patient with a negative assay compared with one with a positive result.
Conclusions: Our data indicate that AR status by ICA may be a useful predictor of survival and endocrine response in prostate cancer. Further studies are needed to confirm these results because the assay could impact significantly on management.