[show abstract][hide abstract] ABSTRACT: In men over 30 years old, serum levels of testosterone (T) decrease with age. A shorter polymorphic CAG repeat length in exon 1 of the androgen receptor (AR) gene is associated with higher transcription activation by the AR. We determined the number of CAG repeats for 882 men aged between 40 and 70 years from the Massachusetts Male Aging Study (MMAS). MMAS is a population-based random sample survey of men for whom baseline (1987-1989, mean age 53+/-8 years) and follow-up (1995-1997, mean age 61+/-8 years) serum hormone levels were available. Multiple linear regression was used to determine if CAG repeat length would be predictive of hormone levels at follow-up. Hormone levels measured included T, free T, albumin-bound T, dihydrotestosterone (DHT), sex hormone-binding globulin (SHBG) and luteinizing hormone (LH). The CAG repeat length was significantly associated with T (P=0.041), albumin-bound T (P=0.025) and free T (P=0.003) when controlled for age, baseline hormone levels and anthropometrics. Follow-up levels of T decreased by 0.74%+/-0.36 per CAG repeat decrement. Likewise, the percentages of free and albumin-bound T decreased by 0.93%+/-0.31 and 0.71%+/-0.32 per CAG repeat decrement respectively. These results suggest that androgen levels may be modulated by AR genotype.
Journal of Endocrinology 08/1999; 162(1):137-42. · 4.06 Impact Factor
[show abstract][hide abstract] ABSTRACT: Background: To determine whether six lifestyle factors (dietary fat, smoking, sleep, alcohol consumption, physical, and sexual activities) are associated with non-prostate cancer-related, elevated prostate specific antigen (PSA) levels.Methods: PSA levels were measured on frozen serum that was collected at baseline (time T1, 1987–1989) and at follow-up (time T2, 1995–1996) in the Massachusetts Male Aging Study, a population-based random sample survey of 1709 men aged 39–70 years old. Men with a PSA level >4.0 ng/ml at time T2 were offered a biopsy. Excluding men who had ever had a diagnosis of prostate cancer (N = 82) left 892 men with both time T1 and time T2 PSA measurements. Lifestyle factors assessed at time T1 were used for the analysis.Results: Of 866 men who had a time T1 PSA <4.0 ng/mL, 56 (6%) had a PSA level >4.0 ng/ml at time T2. In logistic regression models predicting crossing the threshold of 4.0 ng/ml, sexual activity was a significant predictor, but other lifestyle factors were not. Relative to little or no sexual activity, the age-adjusted odds ratio for sex 1–3 times/month was 4.18 (95% confidence interval (CI) = 1.4–12.3), and for sex ≥4 times/month was 2.54 (95% CI = 0.9–6.9).Conclusions: Men who were more sexually active at time T1 were at higher risk of a falsely elevated PSA level >4.0 ng/ml at time T2. Although interim knowledge about lifestyle is lacking, this finding may reflect the effect of ejaculation on serum PSA. Other lifestyle factors had no effect on time T2 PSA levels.