Androgen receptor in prostate carcinoma: immunohistochemical and ligand saturation analyses

Status of androgen receptor (AR) in prostate carcinoma is biologically important. Therefore, more methods assessing AR abnormalities are warranted. Immunohistochemical (IHC) and ligand saturation (LSA) assays were not compared in details. AR in 53 cases were tested by monoclonal antibody (Ab) F39.4.1 (Biogenex), polyclonal Ab N-20 (Santa Cruz) and by ligand saturation analysis with (3)H-methyltrienolon. Statistical analyses were performed with Spearman's nonparametric rank test including neoadjuvant therapy subgroups treated by antiandrogens, combined androgen blockade (22 cases; flutamide with gosereline) or without therapy. By using monoclonal Ab we found AR positive tumor nuclei in 46 cases. Mean of positives was 64%, median 75%. The polyclonal Ab was not sufficiently specific. With LSA AR were found in 43 cases. Mean level was 6.6 fmol/mg, median 5.5 fmol/mg. Comparing IHC with LSA, we noted correlation trend only for the monoclonal Ab (r=0.35; p=0.02). With thresholds 70% positive nuclei for IHC and 6 fmol/mg for LSA, there were 66% and 43% cases positive with IHC and LSA, respectively. The LSA and IHC positives did not show significant agreement, concordance level being 58% only. We found significant IHC-LSA correlation (r=0.68; p=0.004) solely in combined androgen blockade subgroup with 82% level concordance. Our study has demonstrated that AR IHC and LSA are independent complementary methods. Significant correlation between LSA and IHC show only cases treated with combined androgen blockade. An explanation hypothesis is discussed concerning LH-RH influence on free AR capable of ligand binding. IHC as well as LSA have specific biologic significance and may be useful for prostate cancer diagnostic and therapy.