Anthropometrics, body shape over 12 years and risk of cancer events in pre- and post-menopausal women.

Studies of anthropometry and cancer have focused on body mass index (BMI). Relations between weight, waist (WC) and hip circumferences (HC), birth length and adult height with cancer are less well studied. Women from the French E3N study, born between 1925 and 1950, were followed biennially from 1995 until 2008. Body shape was classed into four groups based on median WC and HC at baseline. Hazard ratios (HRs) were estimated by Cox proportional hazards regression models. Over the 12 years of follow-up, 7,247 of 63,798 women developed cancer. As WC increased, we found a trend for decreasing cancer risk in pre-menopausal women, which reversed to an increasing risk in post-menopausal women. This remained unchanged after further adjustment for HC /or height [HR: 0.72 (0.52-1.00) before menopause and 1.17 (1.04-1.31) in the 5th vs. 1st quintile of HC], and were similar after exclusion of breast cancer. We showed that large body shape decreased cancer risk before menopause and increased it after [HR: 0.87 (0.73-1.02) and 1.11 (1.04-1.17), respectively, in women with large waist and hips compared to small waist and hips]. Adult height was associated with an non-significant increase in cancer in pre-menopause and a significant cancer risk in menopause, independent of other anthropometric characteristics [5th vs. 1st quintile [HR: 1.24 (0.98-1.56) and 1.20 (1.10-1.30)], respectively as was long birth length in post-menopausal women [HR: 1.18 (1.07-1.30) compared to medium birth length]. These results suggest independent roles of height and WC on cancer risk, through different pathways.