Allelic variants of aromatase and the androgen and estrogen receptors: Toward a multigenic model of prostate cancer risk

Purpose: The purpose of this study was to determine whether polymorphisms i n the CAG repeat in exon I of the androgen receptor (AR), two intronic rest riction sites in the estrogen receptor (ESRI XbaI and ESRI Null), and an Ar g264Cy5 substitution in the aromatase gene (CYP19) contribute to prostate c ancer risk. Experimental Design: A case-control study was performed with 88 Caucasian p rostate cancer patients and 241 Caucasian male controls. Logistic regressio n models were used to assess individual and joint contributions of genotype s to prostate cancer risk. Results: For single polymorphisms, only the AR repeat number was significan tly related to increased prostate cancer risk [age- and body mass index (BM I)-adjusted odds ratio (OR), 1.14; 95% confidence interval (CI), 1.04-1.25] , suggesting a 14% increase in risk for each missing CAG repeat. When subje cts were classified as either long (greater than or equal to 23 AR CAG repe ats) or short (< 23 repeats) carriers, a significant increase in risk was a lso observed (age- and BMI adjusted OR, 1.75; 95% Cl, 1.05-2.95; P = 0.04). The aromatase C/T was associated with an increase in risk of borderline si gnificance (age- and BMI-adjusted OR, 2.50; 95% Cl, 0.99-6.28). When examin ing the effects of two polymorphisms on prostate cancer risk, homozygosity for the ESRI XbaI restriction site together with a longer AR was more frequ ent among controls (32%) than cases (18%; age-and BMI-adjusted OR, 0.39; 95 % Cl, 0.19-0.78). The aromatase C/C genotype together with a longer AR was also more frequent among controls (55 %) than cases (41 %; age-and BMI-adju sted OR, 0.51; 95% Cl, 0.30-0.89). Conclusions: Estrogen and aromatase may play a role in prostate cancer. A m ultigenic model of prostate cancer susceptibility is also supported.