Risk factors associated with estrogen and progesterone receptor-defined breast tumors

Abstract

1082

Accumulating evidence suggests that the etiology of hormone receptor-defined breast cancers may be heterogeneous, with hormonal factors more strongly associated with hormone receptor-positive than receptor-negative cancer risk (Althuis et al., Ann Epidemiol. 2003 Sept: 13(8): 559). This study compares risk factor profiles for breast cancers stratified by estrogen (ER) and progesterone receptor status (PR) among pre and postmenopausal women from a population-based case-control study in Poland. Incident breast cancer cases, aged 25-74 years diagnosed during 2000-2003, were identified by participating hospitals (>90% of cases) and cancer registries. Controls were randomly selected from population lists and frequency matched to cases within 5 year age groups. Seventy-nine percent of eligible cases and 68% of eligible controls agreed to participate in the study and completed a detailed questionnaire. This report includes 1,955 controls and 1,265 breast cancer cases, representing all cases available at the time of analysis who were not treated prior to surgery (89% of cancers) and with information on ER and PR status (79% of invasive and 29% of in situ cancers) determined by either immunohistochemistry (>95%) or biochemical methods at the treating hospital. Multivariate polytomous logistic regression was used to estimate odds ratios (OR) and 95% confidence intervals (CI) for ER+/PR+ and ER-/PR- breast cancer risk. Among 243 premenopausal women, increased risk of ER+/PR+ but not ER-/PR- tumors was associated with a later age at first birth (30+ vs. <20 years: OR_ER+/PR+=3.2 (CI, 1.0-9.5); OR_ER-/PR-=0.8 (0.2-2.3); P-value for interaction = 0.03), nulliparity (vs. parous: OR_ER+/PR+=2.9 (1.0-5.2); OR_ER-/PR-=0.9 (0.2-2.1); P = 0.11), and a lower body mass index (lowest vs. highest quartile: OR_ER+/PR+=3.1 (1.4-5.0); OR_ER-/PR-=1.3 (0.6-2.5); P = 0.05). Among 717 postmenopausal women, ER+/PR+ but not ER-/PR- tumors were associated with increasing body mass index (highest vs. lowest quartile: OR_ER+/PR+=1.3 (1.0-1.8); OR_ER-/PR-=0.8 (0.5-1.1); P = 0.02). Risks associated with age at menarche, age at menopause, alcohol consumption, oral contraceptive use, family history, and cigarette smoking did not differ by receptor status for either pre or postmenopausal women. Expansion of this work to the remaining breast cancer cases (∼880) and improvement of measures of ER and PR expression using quantitative immunostaining methods should further clarify these relationships and may help elucidate relationships obscured by currently employed techniques.