Interaction of adolescent anthropometric characteristics and family history on breast cancer risk in a historical cohort study of 426 families (USA)

Objective: To determine whether the association of adolescent anthropometric characteristics with breast cancer is modified by a family history of the disease. Methods: These interactions were evaluated in a historical cohort of 426 families of breast cancer probands diagnosed between 1944 and 1952 at the University of Minnesota. The occurrence of breast cancer and the measurement of risk factors in sisters, daughters, granddaughters, nieces and marry-ins was determined through telephone interviews and mailed questionnaires conducted from 1991-1996. Cox proportional hazards regression, accounting for age, birth cohort, adult body mass index (BMI), and clustering within families, was used to estimate relative risks (RR) and 95% confidence intervals (CIs) of breast cancer. Results: Among 4632 women from 426 families available for analysis, there were 175 breast cancers. There was a strong interaction between degree of relationship to proband and relative weight at age 12 on breast cancer risk (p < 0.001). Among sisters and daughters of the probands, risk of breast cancer was slightly increased in those with below average weight at age 12 (RR= 1.55; 95% CI = 0.66-3.64), and strongly increased in those with above average weight (RR = 4.25; 95% CI = 1.71-10.5), compared to those with average weight. In contrast, among marry-ins, there was a weak positive association for those with below average weight at age 12 (RR = 1.61; 95% CI = 0.91-2.83), while there was an inverse association for above average weight (RR = 0.75; 95% CI = 0.26-2.16), compared to those with average weight. There were no significant interactions between degree of relationship to proband and height (p = 0.55), weight at age 18 (p = 0.22) and BMI at age 18 (p = 0.63) on breast cancer risk. Conclusions: Family history appears to modify the effect of obesity in early adolescence on subsequent breast cancer risk, and may identify differing etiologic pathways.