Abstract P124: Network Meta-analysis of the Effects of Breast Cancer Hormone Therapy on Changes in Lipid Profiles

Abstract

Introduction: Adjuvant hormone therapy prolongs survival of patients with early-stage hormone receptor-positive breast cancer (BC). For postmenopausal patients, aromatase inhibitors (AIs) have been shown to improve disease free survival compared to tamoxifen, but the impact on overall survival has been inconsistent. A meta-analysis showed higher risk of cardiovascular diseases (CVDs) for patients taking AIs. Deteriorating lipids induced by AIs may contribute to this result. This analysis aims to compare the effects of hormone therapeutic options on changes in lipids from published randomized clinical trials (RCTs). Methods: RCTs evaluating effects of adjuvant hormone therapy on lipids (total cholesterol, high-density lipoprotein cholesterol (HDLc), low density lipoprotein cholesterol (LDLc), and/or triglycerides) in postmenopausal early-stage BC patients published in PubMed and Embase, prior to Jan. 31, 2016 were reviewed. Bayesian network meta-analysis was used to compare effects of placebo, selective estrogen receptor modulators (SERMs- tamoxifen and toremifene) and AIs (letrozole, anastrozole, and exemstane) on lipids across longitudinal time points. Heterogeneity was examined by meta-regressions adjusting for mean age, baseline lipid value, and prior tamoxifen use. An arm-based random effect model tested the consistency of the direct and indirect evidence of the drug effects. Results: We identified 17 articles from 13 RCTs for a total of 1,913 subjects. The Table summarizes the results, with statistically significant results bolded. Toremifene significantly improved all lipids and was the best choice regardless of covariate adjustment, while tamoxifen had weaker but significant LDLc lowering but opposite HDLc/triglyceride effects to toremifene. AIs generally had little effect on lipids. Conclusions: In general, AIs tend to have worse effects on lipids than SERMs, while only toremifene had beneficial effects on all lipid values. Patients on AIs with high risk of CVD should monitor their lipids.