RESUMEN
Observational studies suggest that host factors are associated with breast cancer risk. The influence of obesity, vitamin-D status, insulin resistance, inflammation, and elevated adipocytokines in women at high risk of breast cancer is unknown. The NSABP-P1 trial population was used for a nested case-control study. Cases were drawn from those who developed invasive breast cancer and controls selected from unaffected participants (≤4 per case) matched for age, race, 5 year Gail score, and geographic location of clinical center as a surrogate for latitude. Fasting serum banked at trial enrolment was assayed for 25-hydroxy vitamin-D (25OHD), insulin, leptin (adipocytokine), and C-reactive protein (CRP, marker of inflammation). Logistic regression was used to test for associations between study variables and the risk of invasive breast cancer. Two hundred and thirty-one cases were matched with 856 controls. Mean age was 54, and 49% were premenopausal. There were negative correlations for 25OHD with body mass index (BMI), insulin, CRP, and leptin. BMI ≥ 25 kg/m(2) was associated with higher breast cancer risk (odds ratio [OR] 1.45, p = 0.02) and tamoxifen treatment was associated with lower risk (OR = 0.44, p < 0.001). Suboptimal 25OHD (<72 nmol/l) did not influence breast cancer risk (OR = 1.06, p = 0.76). When evaluated as continuous variables, 25OHD, insulin, CRP, and leptin levels were not associated with breast cancer risk (all p > 0.34). In this high risk population, higher BMI was associated with a greater breast cancer risk. Serum levels of 25OHD, insulin, CRP, and leptin were not independent predictors of either breast cancer risk or tamoxifen benefit.