Polymorphic repeats in the androgen receptor gene in high-risk sibships

BACKGROUND. Genetic susceptibility may explain some familial clusters of pr ostate cancer. The polymorphic androgen receptor (AR) gene, which mediates androgen activity in the prostate, is a candidate gene that may influence p redisposition to the disease. METHODS. We analyzed the polymorphic (CAG)n and (GGN)n repeats within the A R gene in men from 51 high-risk prostate cancer sibships, which included at least one affected and one unaffected man (n = 210). We compared repeat le ngths of men with prostate cancer (n = 140) to their brothers (n = 70) with out disease, stratified by median age at diagnosis of affected men within e ach sibship. Conditional logistic regression was used to compute odds ratio s (OR) and 95% confidence intervals to evaluate associations between prosta te cancer and repeat length. RESULTS. The OR for prostate cancer associated with short (CAG)n repeats (< 22) compared to longer repeats (greater than or equal to 22) was 1.13 (95% CI 0.5-2.4) overall, but was higher in sibships with a median age of < 66 years at diagnosis (OR = 1.72, 95% CI 0.5-6.0). The (GGN)n array also was n ot associated with prostate cancer in general. However, in older men (great er than or equal to 66 years), there was a modest elevation in risk (OR = 1 .56, 95% CI 0.6-4.1) among those with short repeats (GGN of less than or eq ual to 16). Men with both a short (CAG)n (< 22) and a short (GGN)n (less th an or equal to 16) array were not at higher risk (OR = 1.06) compared to me n with two long repeats [((CAG)n greater than or equal to 22 and (GGN)n > 1 6)]. CONCLUSIONS. These results suggest that the (CAG)n and (GGN)n repeats in th e AR gene do not play a major role in familial prostate cancer.