Associations of erythrocyte ω-3 fatty acids with biomarkers of ω-3 fatty acids and inflammation in breast tissue.

Roy S, Brasky TM, Belury MA, Krishnan S, Cole RM, Marian C, Yee LD, Llanos AA, Freudenheim JL, Shields PG
Int J Cancer 137 2934-46 12/15/2015

Abstract

There is increasing evidence that chronic inflammation is associated with increased breast cancer risk. Long-chain omega-3 polyunsaturated fatty acids (LCω-3PUFA) may reduce circulating biomarkers of inflammation; however associations of blood LCω-3PUFA with breast tissue LCω-3PUFA and breast tissue biomarkers of inflammation are not well understood. We conducted a cross-sectional analysis of breast tissue and blood samples from n = 85 women with no history of breast cancer, who underwent breast reduction surgery. Fatty acids of erythrocytes and undissected breast tissues were analyzed by gas chromatography; C-reactive protein (CRP), interleukin (IL)-6 and IL-8 in plasma and tissue were measured by ELISA. Multivariable-adjusted regression models were used to estimate associations between erythrocyte LCω-3PUFA and breast tissue biomarkers. Women in the highest erythrocyte LCω-3PUFA tertile had LCω-3PUFA concentrations in the breast 73% (95% CI: 31-128%; p trend < 0.0001) higher than women in the lowest tertile. Associations for each individual LCω-3PUFA were similar in magnitude. No significant association was found for the shorter ω-3 PUFA, α-linolenic acid. Although compatible with no association, women in the highest tertile of erythrocyte eicosapentaenoic acid had a nonsignificant 32% (95% CI: -23 to 62%) reduced breast tissue CRP. No correlation was observed between erythrocyte ω-3 PUFA and tissue IL-6 or IL-8 concentrations. Our findings provide evidence that erythrocyte ω-3 fatty acids are valid measures of breast tissue concentrations, and limited evidence that inverse associations from prospective epidemiologic studies of blood LCω-3PUFA and breast cancer risk may be partly explained by reductions in breast tissue inflammation; however, these findings require replication.

Full Text