Background: Research suggests that physical activity can lower the risk of relapse and mortality from CC. However, it is unclear if such benefits are mediated through an effect on weight. Our aims were to (1) evaluate the impact of pre-treatment BMI on recurrence and survival in early stage CC, (2) examine how changes in weight from baseline may affect outcomes, and (3) explore if the effects of different body compositions are modified by adjuvant therapy (AT). Methods: Patients (pts) diagnosed with stage III CC from 2006 to 2008, evaluated at any 1 of 5 cancer centers in British Columbia, and who had their weights measured serially throughout their AT period were reviewed. Using Cox proportional hazards models, we compared outcomes among (a) BMI<25 [normal] vs. 25-30 [overweight] vs. >30 [obese] and (b) weight increase or decrease of < vs. >/=5% and < vs. >/=10%, while controlling for confounders. Results: A total of 821 pts were included: median age was 65 years, 50% were men, and 83% were ECOG 0/1. At baseline, 43, 37, and 20% were normal, overweight, and obese, respectively. Compared to pts with normal BMI, those who were overweight or obese did not show an increased risk of recurrence (HR 0.81, 95%CI 0.42-1.58 and HR 1.09, 95%CI 0.48-2.51, respectively, p trend=0.74) or death (HR 0.54, 95%CI 0.26-1.13 and HR 0.62, 95%CI 0.22-1.74, respectively, p trend=0.24). A number of pts experienced weight gain or loss during their AT period, but none of these weight changes affected outcomes (Table). Receipt of AT did not modify the effects of BMI or weight change on outcomes. In Cox models, advanced age and poor performance status were the main factors that consistently correlated with an increased risk of recurrence and death (p<0.05). Conclusions: Neither baseline BMI nor short-term weight changes during AT appear to affect outcomes in stage III CC. Physical activity likely exerts its benefit on relapse and survival by influencing long-term weight changes or by alternative mechanisms.