SERUM ANDROGENS AND SEX HORMONE-BINDING GLOBULINS IN RELATION TO LIFE-STYLE FACTORS IN OLDER AFRICAN-AMERICAN, WHITE, AND ASIAN MEN IN THE UNITED-STATES AND CANADA
Ah. Wu et al., SERUM ANDROGENS AND SEX HORMONE-BINDING GLOBULINS IN RELATION TO LIFE-STYLE FACTORS IN OLDER AFRICAN-AMERICAN, WHITE, AND ASIAN MEN IN THE UNITED-STATES AND CANADA, Cancer epidemiology, biomarkers & prevention, 4(7), 1995, pp. 735-741
Differences in endogenous androgen levels have been hypothesized to ex
plain ethnic differences in prostate cancer risk. To examine this hypo
thesis, we gathered data on serum concentrations of androgens and sex
hormone-binding globulin (SHBG) in healthy older men from four ethnic
groups at different levels of prostate cancer risk. As part of a popul
ation-based case-control study of prostate cancer we conducted in Cali
fornia, Hawaii, and Vancouver, Canada, 1127 African-American, white, C
hinese-American, and Japanese-American control men, mostly ages 60 yea
rs or older (mean age, 69.9 years) provided information on various lif
estyle factors and donated an early morning fasting blood sample betwe
en March 1990 and March 1992, We used these data to examine the distri
butions of serum androgens I:testosterone (total, free, and bioavailab
le), dihydrotestosterone (DHT)I, the ratio of DHT to total testosteron
e (DHT:testosterone ratio), and SHBG in these four ethnic groups. We a
lso assessed correlations between concentrations of these measures wit
h age, body size, physical activity, and other personal characteristic
s, and we evaluated ethnic differences in concentrations of androgens
and SHBG after adjusting for these characteristics. In each of the fou
r ethnic groups, concentrations of free and bioavailable testosterone
declined with age, whereas SHBG concentrations,increased with age. Age
-adjusted concentrations of all androgen measures and SHBG decreased w
ith increasing levels of Quetelet's index. After adjustment for age; a
nd Quetelet's index, androgens and SHBG showed no clear and consistent
relationships to physical activity, alcohol consumption, or tobacco u
se. DHT:testosterone ratio was higher in men reporting a history of be
nign prostate disease than in men without such a history, and higher i
n vasectomized men than in nonvasectomized men. SHBG concentrations we
re higher in men reporting one or more first-degree relatives with pro
state cancer than in men without such a family history. After adjustme
nt for age and Quetelet's index, the levels of total and bioavailable
testosterone were highest in Asian-Americans, intermediate in African-
Americans, and lowest in whites. However, the DHT:testosterone ratio w
as highest in African-Americans, intermediate in whites, and lowest in
Asian-Americans, corresponding to the respective incidence rates in t
hese groups and providing indirect evidence for ethnic differences in
5 alpha-reductase enzyme activity.