Urinary estrogen metabolites and breast cancer: differential pattern of risk found with pre- versus post-treatment collection

Introduction: The products of estrogen metabolism may affect breast carcinogenesis. The 16alpha-hydroxyestrone (16-OHE) metabolite has a higher affinity for the estrogen receptor (ER) than the 2-hydroxyestrone (2-OHE) metabolite, while conjugated 2-OHE metabolite may inhibit angiogenesis. We investigated the association between the relative concentrations of these metabolites in urine (2-OHE/16-OHE) and breast cancer in a case-control study of Chinese women living in Shanghai. Methods: Incident breast cancer cases between 25 and 65 years of age (n = 110) were identified from hospital or population tumor registries in Shanghai, China. Controls (n = I 10) were randomly selected from a complete registry of the Shanghai population, and individually matched to cases by menopausal status, age, and pre-treatment or post-treatment urine collection time. Urine samples were collected prior to any breast cancer treatment or surgery among 78 case-control pairs, while urine was collected after surgery, and perhaps other treatments, among 32 case-control pairs. A commercial enzyme-immunoassay kit was used to measure urinary estrogen metabolite concentrations. Conditional logistic regression was used to calculate odds ratios summarizing the 2-OHE/16-OHE and breast cancer association within subjects providing either pre-treatment or post-treatment urine samples. Results: Subjects with a higher urinary 2-OHE/16-OHE ratio were less likely to be diagnosed with breast cancer, but only when urine samples were collected prior to breast cancer treatment (ORTertile (3) ((T3)) (versus Tertile) (1) ((T1)) = 0.5, 95% CI (0.2, 1.1)). In contrast, a higher 2-OHE/16-OHE ratio was significantly associated with breast cancer among subjects providing urine specimens after treatment initiation (ORT3 (versus T1) = 8.7, 95% Cl (1.6, 47.1)). This observed cross-over modification occurred within both pre-menopausal and post-menopausal women, and independent of body mass index or recent dietary intake. Conclusion: Cross-study differences in urine collection protocols may explain observed inconsistencies in the 2-OHE/16-OHE and breast cancer association. Our case-control analysis using pre-treatment urine samples suggested that a lower 2-OHE/16-OHE ratio was associated with an increased risk of pre-menopausal and post-menopausal breast cancer diagnosis among Chinese women. (C) 2002 Elsevier Science Inc. All rights reserved.