Abstract Objective: Hot flashes in breast cancer survivors (BCS) receiving adjuvant aromatase inhibitor (AI) therapy are common, but risk factors for these symptoms are not well-defined. This study tested if body size is associated with hot flashes in BCS on AI therapy.Materials and methods: We performed a cross-sectional analysis of a cohort of postmenopausal women with Stages 0-III breast cancer receiving adjuvant AI therapy at a university hospital-based outpatient breast oncology clinic. Subjects provided patient-reported outcomes on hot flashes, current and pre-breast cancer weight and height, and additional demographic and treatment data.The primary outcome was occurrence of hot flashes. Secondary outcomes included hot flash severity and frequency. The exposures of interest were 1) current weight and 2) weight change since breast cancer diagnosis, categorized as weight loss (lost 10 pounds or more), maintenance (+/- 10 pounds), or gain (gained 10 pounds or more). Multivariable logistic regression models examined the independent association between hot flash outcomes and current weight or weight change while controlling for confounders.Results: 300 participants were enrolled at a mean age of 61 years (range 33-86) after mean duration of AI exposure of 23 months (range 1 month-9 years). 177 (59%) reported hot flashes overall, 96 (32%) reported moderate to very severe hot flashes, and 60 (20%) had at least 4 hot flashes daily. 182 (61%) experienced weight maintenance, while 81 (27%) had weight gain and 34 (11%) had weight loss. In a multivariable model, weight gain was independently associated with hot flash occurrence (OR 2.1, 95% CI 1.1-4.4) and hot flash severity (OR 2.4, 95% CI 1.4-4.9), but not hot flash frequency, after adjusting for current weight and height, age, smoking, race, prior chemotherapy and prior tamoxifen therapy. Hot flash occurrence was also significantly associated with current smoking (OR 5.1, 95% CI 1.1-24.0) and age (OR for each additional year of age 0.93, 95% CI 0.90-0.97). Current weight, alcohol use, AI type, and AI duration were not associated with any of the three outcomes.Conclusions: In an outpatient BCS population on AI therapy, weight gain is a risk factor for both hot flash occurrence and severity. Women who gained at least ten pounds since breast cancer diagnosis were two times more likely to have hot flashes than women who maintained or lost weight. These results support the thermoregulatory model of hot flashes and argue against a protective effect of body fat in this population.Support: ACS MRSG-08-110-01-CCE, ACS CCDA#-08-107-01, Pennsylvania Department of Aging Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 1059.