Mechanisms underlying the effect of adiposity on risk of post-menopausal oestrogen receptor-positive breast cancer: an interventional mediation analysis.

BACKGROUND

Adiposity increases the risk of post-menopausal oestrogen receptor (ER)-positive breast cancer. Inflammation, insulin, and insulin-like growth factors and sex-steroid hormones may explain this effect. We performed interventional mediation analysis to estimate the effects of hypothetical interventions targeting these pathways in females with obesity on reducing their excess risk of post-menopausal ER-positive breast cancer relative to females with normal weight.

METHODS

The mediation analysis included 1260 post-menopausal females (352 with ER-positive breast cancer) from a case-cohort within the Melbourne Collaborative Cohort Study. A Monte Carlo g-computation approach with non-parametric bootstrapping was used to estimate the risk differences (RDs) and 95% confidence intervals (CIs) for interventional direct and indirect effects of hypothetical interventions that shift the joint distribution of inflammation, insulin, and sex-steroid hormone biomarkers in females with obesity to the levels observed in females with normal weight.

RESULTS

The estimated RD for the total effect of obesity relative to normal weight on post-menopausal ER-positive breast cancer was 16.1 (95% CI: 3.3, 30.4) per 1000 females. The RDs for indirect effects through inflammation, insulin, and sex-steroid hormones were 16.4 (95% CI: 4.3, 27.2), -8.4 (95% CI: -23.4, 1.2), and 13.0 (95% CI: 3.3, 23.7) per 1000 females, respectively. The - RD for the direct effect not via any of these three pathways was -5.3 (95% CI: -17.4, 10.6) per 1000 females.

CONCLUSION

Inflammation and sex-steroid hormones, but not insulin, contributed to the detrimental effect of adiposity on the risk of post-menopausal ER-positive breast cancer. Interventions targeting these pathways may reduce the risk for females with obesity.