Abstract P2-11-09: Weight Change and Risk of Incident Diabetes After Breast Cancer

Weight gain after breast cancer treatment is common and may increase risk of diabetes. This analysis examined the association between weight change pattern and incident diabetes, adjusting for pre- and post-diagnosis BMI in breast cancer survivors (n = 1,617) participating in the Women9s Healthy Eating & Living (WHEL) Study. Weight at 1 year pre-diagnosis and ∼2 years post-diagnosis were used to calculate pre- to post-diagnosis body weight percentage change (categorized as: loss (>5%), stable (± 5%), moderate gain (>5% to In this cohort of 1617 breast cancer survivors, 54 (3.3%) women developed type 2 diabetes mellitus over 6 years of follow-up (approximately 8 years post-cancer diagnosis), with 19 (35.2%) self-reporting diabetes and the remaining identified by year 6 HbA1C level ≥ 6.5%. Less education, lower levels of physical activity, and higher BMI were statistically significantly associated with incident diabetes (p ≤ 0.05 for all). Overall, 43.8% of women had stable weight from pre- to post-diagnosis of breast cancer, followed by the categories of major weight gain (23.5%), moderate weight gain (19.1%), and weight loss (13.6%). After adjustment for pre-diagnosis BMI, both weight loss and major weight gain showed an approximate 2-fold increase in risk of incident diabetes, compared to stable weight (p 2 was associated with a much higher in risk of developing diabetes (OR 5.4, 95% CI 2.6, 10.9). As indicated above, we also constructed a weight change model that adjusted for pre-diagnosis and post-diagnosis BMI because of evidence indicating that “attained” BMI is a strong risk factor for diabetes. In this model, only weight loss was significantly associated with the development of diabetes (OR 3.2, 95% CI 1.5–7.0). In this model, pre-diagnosis BMI was not associated with risk of developing diabetes; however, having a post-diagnosis BMI of ≥ 30 kg/m 2 was associated with a large increase in risk (OR 5.4, 95% CI 1.5, 19.7). To explore the unexpected finding that pre- to post-diagnosis weight loss was associated with increased risk of diabetes, we graphed long-term weight outcomes. On average, women in the weight loss category had the highest pre-diagnosis weight (mean BMI= 27.1, SD = 5.3 kg/m 2 ), and returned to that weight by the year 6 follow-up visit. Treatment-associated weight change is considered an important risk factor for breast cancer survivors, both in terms of cancer prognosis and comorbidities. Our results indicate that weight gain around the time of cancer diagnosis tended to be maintained over time (i.e., is permanent), while weight loss appeared to be relatively transient. Therefore, the increased risk of diabetes associated with pre- to post-diagnosis weight loss appears to be biologically explained by subsequent weight regain. These data indicate that focusing on weight and weight change in a relatively narrow time frame around diagnosis may not provide a complete picture regarding the risk of comorbidities or breast cancer outcomes associated with breast cancer treatment. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P2-11-09.