THE RELATION OF REPORTED ALCOHOL INGESTION TO PLASMA-LEVELS OF ESTROGENS AND ANDROGENS IN PREMENOPAUSAL WOMEN (MARYLAND, UNITED-STATES)

We undertook a cross-sectional study in 107 premenopausal women in Mar yland (United States) of alcohol intake and hormonal status in order t o evaluate whether plasma hormone levels might mediate the reported po sitive relation between alcohol ingestion and breast cancer risk. Alco hol ingestion was estimated using a drinking pattern questionnaire, a food frequency questionnaire, and seven-day food records. Fasting bloo d specimens were collected on days 5-7, 12-15, and 21-23 of each parti cipant's menstrual cycle and pooled to create follicular, midcycle, an d luteal phase samples, respectively, for analysis. Estrone, estrone s ulfate, estradiol, androstenedione, and dehydroepiandrosterone sulfate (DHEAS) in plasma were measured by radioimmunoassay, and sex-hormone binding globulin (SHBG) was measured by an immunoradiometric assay. Af ter adjusting for age, weight, and total energy intake, alcohol ingest ion was not associated with plasma estrogens in the follicular, midcyc le, or luteal phases of the menstrual cycle, nor with the level of SHB G or DHEAS in plasma averaged from the three phases of the cycle. Alco hol, however, was significantly positively associated with the average level of plasma androstenedione. Based on these cross-sectional findi ngs among premenopausal women, the increased risk of breast cancer rel ated to alcohol ingestion does not appear to be mediated by elevated p lasma estrogen levels. Androstenedione, however, may mediate the alcoh ol/breast cancer-association.