Limited evidence suggests that hyperinsulinemia may contribute to the risk of breast, endometrial, and, possibly, ovarian cancer. The aim of this study was to assess the association of serum glucose and insulin with risk of these cancers in postmenopausal women, while taking into account potential confounding and modifying factors. We studied 21 103 women with fasting baseline insulin and glucose measurements in a subsample of the Women's Health Initiative. The subsample was composed of four studies within Women's Health Initiative with different selection and sampling strategies. Over a mean of 14.7 years of follow-up, 1185 breast cancer cases, 156 endometrial cancer cases, and 130 ovarian cancer cases were diagnosed. We used Cox proportional hazards models to estimate hazard ratios (HRs) and 95% confidence intervals (95% CIs) by quartile of glucose or insulin. Serum insulin was positively associated with breast cancer risk (multivariable-adjusted HR for highest vs. lowest quartile 1.41, 95% CI: 1.16-1.72, Ptrend<0.0003), and glucose and insulin were associated with roughly a doubling of endometrial cancer risk (for glucose: HR: 2.00, 95% CI: 1.203.35, Ptrend=0.01; for insulin: HR: 2.39, 95% CI: 1.32-4.33, Ptrend=0.008). These associations remained unchanged or were slightly attenuated after mutual adjustment, adjustment for serum lipids, and assessment of possible reverse causation. Glucose and insulin showed no association with ovarian cancer. Our findings provide support for a role of insulin-related pathways in the etiology of cancers of the breast and endometrium. However, because of the unrepresentative nature of the sample, our results need confirmation in other populations.