Abstract A96: Association of leptin with sex steroid hormones, insulin, and CRP among adult men in NHANES III

Abstract

Abstract A96 Objective We previously observed an inverse association between leptin concentration and risk of prostate cancer, which was contrary to our hypothesis (AACR 2008). It is possible that the association between leptin - a hormone secreted by adipocytes - and prostate cancer or other cancers may be explained by confounding by other hormone and metabolic biomarkers. To address this issue, we evaluated the association of leptin with sex steroid hormones, insulin, and C-reactive protein (CRP) among a nationally representative sample of US men in the Third National Health and Nutrition Survey (NHANES III). Methods 1149 men aged 20+ years and who attended the morning examination session of Phase I of NHANES III (1988-1991) were included. We measured serum concentrations of testosterone, androstanediol glucuronide (AAG) (a metabolite of dihydrotestosterone), estradiol, and sex hormone binding globulin (SHBG) by immunoassay. Insulin, CRP, and leptin were measured previously. Free hormones were estimated from measured total hormones, albumin, and SHBG. Height, weight and waist circumference were measured. Percent body fat was calculated from bioelectrical impedance and anthropometric data. Data on alcohol, smoking, and physical activity were obtained by interview. Using linear regression, we estimated geometric mean concentrations of hormones, insulin, and CRP by quartiles of leptin, applying sampling weights and adjusting for age, alcohol, smoking, physical activity, waist circumference, percent body fat and race/ethnicity. Results Men who had higher leptin concentrations had lower testosterone (Q1: 5.7, Q2: 5.3, Q3: 4.8, Q4: 4.7 ng/mL; p-trend = 0.02). Although the trend was not statistically significant, men who had higher leptin had lower free testosterone (0.109, 0.106, 0.098, 0.096 ng/mL; p-trend = 0.10). Men who had higher leptin had higher estradiol (34.7, 34.8, 36.4, 38.5 pg/mL; p-trend = 0.03), free estradiol (0.86, 0.89, 0.94, 1.01 pg/mL; p-trend = 0.007), and insulin (43.7, 50.1, 56.3, 60.0 pmol/L; p-trend = 0.0006) levels. Men in the lowest quartile of leptin had the highest SHBG; levels were the same in the upper 3 quartiles (37.7, 34.0, 33.1, 32.5 nmol/mL; p-trend = 0.14). Leptin did not appear to be associated with AAG or CRP. These patterns were similar within age strata. Conclusions Even after taking into account extent of body fat and other factors, leptin appears to be inversely associated with total and free testosterone but positively associated with estradiol, free estradiol, and insulin. In analysis of the independent association of leptin and prostate and other cancers, whether overall or by stage/grade, in addition to adiposity, adjustment for other hormonal and metabolic correlates of leptin may also be necessary. Citation Information: Cancer Prev Res 2008;1(7 Suppl):A96.