A meta-analysis of endocrine therapy trials in early breast cancer (BC) evaluating the impact of obesity: Are aromatase inhibitors (AIs) optimal therapy in obese ER+ BC?

Breast Cancer - HER2/ER
Session Type and Session Title: 
General Poster Session, Breast Cancer - HER2/ER
Abstract Number: 



J Clin Oncol 31, 2013 (suppl; abstr 575)


Lina Pugliano, Dimitrios Zardavas, Marianne Paesmans, Ivana Sestak, Richard D. Gelber, Jack M. Cuzick, Mitchell Dowsett, Ahmad Awada, Martine J. Piccart-Gebhart, Evandro De Azambuja; Jules Bordet Institute, Breast International Group, Brussels, Belgium; Institut Jules Bordet, Brussels, Université Libre de Bruxelles, Ixelles, Belgium, Brussels, Belgium; Institut Jules Bordet, Université Libre de Bruxelles, Ixelles, Belgium, Brussels, Belgium; Queen Mary, University of London, London, United Kingdom; Department of Biostatistics and Computational Biology, Dana-Farber Cancer Institute, IBCSG, Frontier Science and Technology Research Foundation, Boston, MA; Centre for Cancer Prevention, Wolfson Institute of Preventive Medicine, London, United Kingdom; The Royal Marsden Hospital NHS Foundation Trust, London, United Kingdom; Medical Oncology Clinic, Jules Bordet Institute, Brussels, Belgium

Abstracts that were granted an exception in accordance with ASCO's Conflict of Interest Policy are designated with a caret symbol (^).

Abstract Disclosures


Background: Obesity is an adverse prognostic factor in BC. Mixed results are reported for the relative efficacy of AIs compared to tamoxifen (T) in obese ER+ BC patients. Our purpose was to conduct a meta-analysis of adjuvant randomised trials of AIs vs T assessing the impact of body mass index (BMI). Methods: We identified four studies evaluating BMI and endocrine therapy. Of these, 3 were randomised (non-steroidal AIs vs T) and were evaluable for the aggregation of results for DFS and OS in our meta-analysis. We extracted published data from ATAC, ABCSG-12 and BIG01-98, analyzed according to standard meta-analytic techniques. Results: A total of 11,383 patients were included in our study. BMI>25 is associated with reduced disease free survival (DFS) and a trend towards worse overall survival (OS) (Table 1). A significantly shorter DFS was seen for patients with BMI>25 treated with an AI while a trend was seen for OS. Reduced relative efficacy was seen for DFS for AIs compared to T for BMI<25 (HR=0.78; 95%CI 0.66- 0.91; p=0.002) and a trend for BMI>25 (HR=0.85; 95%CI 0.70- 1.02; p=0.08). The test for interaction was not significant (p=0.48), with similar results for OS for BMI<25 (HR=0.79; 95%CI 0.63-0.9; p=0.009) and BMI>25 (HR=0.98; 95%CI 0.61-1.60; p=0.95). The test for interaction was not significant (p=0.37). Notably, significant heterogeneity in patients treated with anastrozole and a BMI>25 did not allow a comparison between anastrozole and letrozole. Conclusions: BMI>25 has a negative prognostic effect in BC. AIs demonstrate improved outcomes in normal weight BC patients (BMI<25). Obesity was associated with observed relative reduced efficacy of AIs; however, we were not able to detect a significant interaction between BMI and treatment effect. Further analyses into the differing impact of type of AIs on BC outcomes in obese patients are warranted.

HR 95%CI P value
DFS DMI>25 vs <25 1.09 1.00-1.19 0.04
OS BMI>25 vs <25 1.12 0.97-1.29 0.12
DFS AI treated patients BMI>25 vs <25 1.21 1.07-1.39 0.003
OS AI treated patients BMI>25 vs <25 1.44 0.76-2.74 0.26
DFS T treated patients BMI>25 vs <25 1.01 0.90-1.14 0.84
OS T treated patients BMI>25 vs <25 1.08 0.88-1.33 0.45