Abstract PD4-08: Post-diagnosis physical activity and comorbidities, not BMI, explain mortality risk in the after breast cancer pooling project

Abstract

Abstract Background: In its 2014 position statement, ASCO concluded that obesity is associated with worse prognosis after cancer diagnosis. However in the same year, a comprehensive review by the World Cancer Research Fund concluded that there was limited evidence that greater body fatness increases risk of overall or breast cancer mortality, indicating that further investigation into lifestyle factors are needed. The After Breast Cancer (ABC) Pooling Project has reported, separately, significant mortality effects of pre-diagnosis BMI and of post-diagnosis physical activity (PA). We investigate whether the effect of BMI can be limited to subgroups characterized by comorbidities and physical activity. Methods: Data are from the three US cohorts that were harmonized in the ABCPP (n=9513) including: the Women's Healthy Eating and Living (WHEL), Life After Cancer Epidemiology (LACE), and Nurses' Health (NHS) studies. Stepwise delayed entry Cox proportional hazards models examined each lifestyle predictor (BMI, PA, and comorbidities assessed after diagnosis) sequentially and together in multivariate models for breast cancer and all-cause mortality. Results: In multivariate models without the other two target variables, PA was significantly associated with a 17% decrease in the risk of breast cancer mortality among women in the highest quartile of PA (MET hr/wk > 21.4), compared to the lowest quartile (MET hr/wk < 2.7) (HR=0.81,95% CI= 0.67,0.97). In the model with major comorbidities, there was a significant 40% increase in the risk of breast cancer mortality among women diagnosed with both diabetes and hypertension (HR=1.40, 95% CI= 1.01,1.93). In the model with BMI, there was no significant association with risk of breast cancer mortality. These results were essentially unchanged with all variables in a single model. For all-cause mortality, the PA-only model showed a significant PA effect with the hazard decreasing from 20% to 40% across quartiles (Q2 HR=0.80, 95% CI=0.71,0.90, Q4 HR=0.62, 95% CI=0.54,0.71). In the comorbidity-only model, both diabetes and hypertension significantly increased hazard of all-cause mortality 80% and 33%, respectively. Having both diagnoses was associated with a significant, 2.3 fold increase in all-cause mortality (HR=2.34, 95% CI= 1.95,2.81). In the BMI-only model, being underweight was associated with a significant 2.4 fold increase in risk of all-cause mortality, and there was a 20 and 37% increase in risk associated with being categorized as obese I or II (Obese I HR=1.23, 95% CI=1.07,1.40, Obese II HR=1.37, 95% CI=1.16,1.61). With all three variables in the model, the risk associated with being obese decreased and became non-significant (Obese I HR=1.06, Obese II HR=1.05), while the significance, strength, and direction of the association of comorbidities and PA with all-cause mortality remained constant. Conclusion: These data suggest that post-diagnosis comorbidities and lack of physical activity, rather than high BMI , are the important risk factors for all-cause and breast cancer specific mortality. While needing further validation, these suggest that physical activity interventions and monitoring treatment for comorbidities should become standard of care for breast cancer survivors. Citation Format: Nelson SH, Marinac CR, Patterson RE, Nachuta SJ, Caan BJ, Chen WY, Shu X-O, Pierce JP. Post-diagnosis physical activity and comorbidities, not BMI, explain mortality risk in the after breast cancer pooling project. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr PD4-08.