Associations of body mass index with cancer incidence among populations, genders, and menopausal status: A systematic review and meta-analysis

In order to further reveal the differences of association between body mass index (BMI) and cancer incidence across populations, genders, and menopausal status, we performed comprehensive meta-analysis with eligible citations. The risk ratio (RR) of incidence at 10 different cancer sites (per 5 kg/m2 increase in BMI) were quantified separately by employing generalized least-squares to estimate trends, and combined by meta-analyses. We observed significantly stronger association between increased BMI and breast cancer incidence in the Asia⿿Pacific group (RR 1.18:1.11⿿1.26) than in European⿿Australian (1.05:1.00⿿1.09) and North-American group (1.06:1.03⿿1.08) (meta-regression p < 0.05). No association between increased BMI and pancreatic cancer incidence (0.94:0.71⿿1.24) was shown in the Asia⿿Pacific group (meta-regression p < 0.05), whereas positive associations were found in other two groups. A significantly higher RR in men was found for colorectal cancer in comparison with women (meta-regression p < 0.05). Compared with postmenopausal women, premenopausal women displayed significantly higher RR for ovarian cancer (pre- vs. post- = 1.10 vs. 1.01, meta-regression p < 0.05), but lower RR for breast cancer (pre- vs. post- = 0.99 vs. 1.11, meta-regression p < 0.0001). Our results indicate that overweight or obesity is a strong risk factor of cancer incidence at several cancer sites. Genders, populations, and menopausal status are important factors effecting the association between obesity and cancer incidence for certain cancer types.