Abstract P4-13-02: Long-term effects of weight loss on breast cancer biomarkers in postmenopausal women

Abstract

Abstract Background: Obesity increases risk for postmenopausal breast cancer, and is associated with elevated blood levels of inflammation-related marker C-reactive protein (CRP), insulin, glucose, and the angiogenesis-related biomarker vascular endothelial growth factor (VEGF). The long-term effect of weight loss on these breast cancer risk biomarkers is unknown. This study aim was to test the 30-month effect of weight loss on breast cancer biomarkers in postmenopausal overweight/obese women, who had completed a randomized control trial test of behavioral weight loss and/or exercise compared to control. Methods: From 1/2005-7/2008, women (N=438) were randomized to 1 of 4 arms: reduced calorie weight loss diet (D; N=118); moderate intensity aerobic exercise (E; 225 minutes/week) (N=117); both interventions (D+E; N=116); or control (C, no intervention; N=87). The D intervention goals were: 1200-2000 calories/day, dietary fat < 30% of calories, and > 10% loss of initial weight. The E goal was > 45 minutes of moderate-intensity exercise 5 days/week. The D+E interventions were identical to those for D and E. A total of 157 women returned for a 30-month blood draw (18 months after end-of-study). Analysis: We compared the average changes in outcomes from baseline to 30 months in weight loss (combining D and D+E groups) vs. no weight loss (combining C and E groups), according to the intention-to-treat principle, using generalized estimating equations (GEE). As a secondary, pre-planned analysis, we combined data for all women regardless of randomization group, and divided them into four categories based on change from baseline to 30 months: 1) gained weight or remained weight stable; 2) lost < 5% of baseline weight; 3) lost 5% - <10% of baseline weight; and 4) lost > 10% of baseline weight. We then compared 30-month analyte changes by these 30-month weight loss groups, with category 1 as referent. Results: Mean 30-month weight changes from baseline were: D+E, - 7.7%; D, - 6.3%; E, -1.9%; and C, -3%. CRP was 22.5% lower than baseline for women randomized to a diet group, compared with a 4% reduction from baseline in women in a non-diet group (p=0.07). Insulin was 33% lower in women randomized to diet, while it was 20% lower in women not randomized to a diet group (p=0.10). Women who lost 5% - <10% of baseline weight at 30-months experienced an 18% (p<0.001) decrease in CRP compared to baseline, while those who lost > 10% had a 56% lower CRP level (p<0.001). VEGF decreased to a similar amount (14% -15%) in women who lost <10% of baseline weight; and decreased by 26% in those whose 30-month weight decreased by > 10% (p<0.002). Compared with baseline levels, insulin decreased by 16% in those who lost < 5% of baseline weight. In contrast, insulin levels decreased by 31% and 44% (each p=0.004), in those whose 30-month weight loss was 5% - <10% lower than baseline and > 10% lower than baseline, respectively. Glucose levels increased in women whose weight loss was < 5%, but remained stable in those whose 30-month weight was > 5% lower than baseline (each p=0.04). Conclusions: Long-term weight loss of at least 5% may have biological effects relevant to breast cancer prevention in overweight or obese postmenopausal women. Citation Format: McTiernan A, Duggan C, Tapsoba JdD, Mason C, Wang C-Y. Long-term effects of weight loss on breast cancer biomarkers in postmenopausal women [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P4-13-02.