RISK FACTORS FOR BREAST CANCER: POOLED RESULTS FROM THREE ITALIAN CASE-CONTROL STUDIES

Publisher Website
Google Scholar
Abstract
Negri, E. (“Mario Negri” Institute for Pharmacological Research, Via Eritrea, 62, 20157 Milan, Italy), C. La Vecchia, P. Bruzzi, G. Dardanoni, A. Decarli, D. Palli, F. Parazzini, and M. Rosselli del Turco. Risk factors for breast cancer pooled results from three Italian case-control studies. Am J Epidemiol 1988; 128:1207–15. The role of menstrual and reproductive factors, family history, and body weight in the epidemiology of breast cancer has been reassessed in a meta-analysis of three large case-control studies of breast cancer from several Italian regions, for a total data set of 4,072 cases and 4,099 controls. Multiple logistic regression equations were used to obtain relative risks adjusted for study, center, age, and various combinations of risk factors considered. Relative to women with men-arche at age 15 or over, those with earlier menarche had a 20–30% higher breast cancer risk. However, there was no tendency for the risk to increase with lower age at menarche, and the association with menarche was stronger at younger age. The risk of breast cancer was directly related to age at menopause (relative risk (RR)=0.7 for 50 years), age at first live birth (RR =1.8 for >28 years vs. <22 years), and family history of breast cancer in first-degree relatives (RR=2.0). The effect of these factors was similar in various age strata. After allowance for age at first live birth, the risk of breast cancer did not differ among women with one to four live births, but it was significantly below unity (RR=0.6) for those with five or more live births. Furthermore, there was a clear modifying effect of age at diagnosis on parity-related risk, since parous women had elevated breast cancer risk below age 35 and reduced risk above age 40. The risk of breast cancer was higher in overweight women; this positive association, however, was confined to postmenopausal subjects (RR=1.6 for obese women vs. leaner women). In terms of the multistage theory of carcinogenesis, the observation that the relative risk for family history and age at first live birth are of similar magnitude in subsequent age groups would indicate that the effect of these factors is on one of the earlier stages of the process; the decreasing association with advancing age for menarche would suggest the existence of some short-term (hormonal) correlate of early menarche; and the complex pattern of risk for parity is consistent with a possible differential action of this factor (or its hormonal correlates) on breast carcinogenesis, in terms of accelerated promotion but delayed initiation.