We assessed breast cancer risk in relation to weight at birth and adolescence. In-person interviews were completed with the biological mothers of women aged 45 years and younger who participated in the Shanghai Breast Cancer Study in 1996 -98 (288 cases, 350 controls). After adjustment for confounding, women who were 4000 g or more at birth were not at increased risk of breast cancer (odds ratio=0.7; 95% confidence interval 0.4 -1.4) relative to women whose birth weight was 2500 -2999 g. Compared with women of average perceived weight at age 15 years, no relation was apparent for heavier than average weight based on maternal report (odds ratio=0.7; 95% confidence interval 0.5 -1.2) or self-report (odds ratio=1.0; 95% confidence interval 0.7 -1.6). Perceived adolescent weight and height did not modify the association of birth weight with breast cancer risk. These results suggest that weight early in life is not related to premenopausal breast cancer risk in this low-risk population. (Ekbom et al, 1992;Innes et al, 2000;Michels et al, 1996;Sanderson et al, 1996). Conversely, high adolescent (Coates et al, 1999;Hislop et al, 1986;Le Marchand et al, 1988a), early adult (Coates et al, 1999;Huang et al, 1997;Trentham-Dietz et al, 1997) and adult weight or body mass index (Brinton and Swanson, 1992;Huang et al, 1997;Swanson et al, 1996;Ursin et al, 1995;van den Brandt et al, 2000) appear to be protective against premenopausal breast cancer. Several studies have investigated the association between breast cancer and weight at birth (De Stavola et al, 2000;Ekbom et al, 1992Ekbom et al, , 1997Innes et al, 2000;Le Marchand et al, 1988b;Michels et al, 1996;Sanderson et al, 1996Sanderson et al, , 1998a or weight at adolescence (Brinton and Swanson, 1992;Choi et al, 1978;Coates et al, 1999;Franceschi et al, 1996;Hislop et al, 1986;Le Marchand et al, 1988a;Pryor et al, 1989) with inconsistent findings. Possible limitations of these studies related to exposure measurement and age at diagnosis of breast cancer.Since self-report of body size in early life is prone to misclassification, maternal report may be less subjective. Maternal report was available for two of the studies investigating breast cancer risk associated with birth weight (Michels et al, 1996;Sanderson et al, 1998a), but none of the studies of adolescent weight. The present analysis was conducted to assess whether birth weight and adolescent weight as reported by subjects' mothers were related to premenopausal breast cancer risk. In addition, we investigated whether perceived adolescent weight and height modified the association of birth weight with breast cancer risk.
MATERIALS AND METHODSDetailed methods of this population-based case -control study appear elsewhere (Gao et al, 2000). Briefly, all women aged 25 -64 years who were permanent residents of urban Shanghai at the time of diagnosis of first primary invasive breast cancer (August 1996 through March 1998) were eligible for the study. Two senior pathologists histologically confirmed all diagnoses. We used...