ABSTRACT Aim: Background: Estrogen receptor positive (ER+) metastatic breast cancer patients progressing on nonsteroidal aromatase inhibitors are candidates for treatment with everolimus (EVE) plus exemestane (EXE) or with chemotherapy. The aim of the study was to conduct a cost-effectiveness analysis (CEA) of EVE plus EXE versus bevacizumab (BEV) plus paclitaxel (PACL) and BEV plus capecitabine (CAPE) for the management of postmenopausal women with ER+ breast cancer from the perspective of the Greek National Health System. Methods: A Markov model, consisting of three health states, was used to describe disease progression and evaluate the CE of comparators. The model evaluated the comparators over a lifetime horizon in the course of a 1-month cycle from a third-party payer perspective. Efficacy, safety and utility data considered in the model were extracted from randomized Phase III clinical trials. Direct medical costs used reflecting the year 2013. Probabilistic sensitivity analysis was also conducted. Primary outcomes were quality-adjusted life years (QALYs) and total lifetime costs. Results: Discounted survival and quality adjusted survival of EVE plus EXE treated patients was higher compared to BEV plus PACL by 0.28 life-years and 0.19 QALYs, respectively, and by 0.21 life-years and 0.16 QALYs when compared to BEV plus CAPE. EVE plus EXE was the less costly treatment in terms of drug acquisition, administration and concomitant medications. The total lifetime cost per patient for EVE plus EXE was estimated at e46,251, for BEV plus PACL at e53,422 and for BEV plus CAPE at 50,619 e. The higher drug acquisition and administration cost of BEV plus PACL (EVE plus EXE: 16,956 e vs. BEV plus PACL: 22,379 e) partially explained the higher total lifetime cost over EVE plus EXE. Moreover, the higher pre-progressed costs of the BEV plus CAPE arm accounted for the largest part of the difference in the total lifetime cost between BEV plus CAPE and EVE plus EXE. Probabilistic analysis confirmed the results. Conclusions: Results suggest that EVE plus EXE may be a dominant (more effective and less costly) alternative relative to BEV plus PACL and BEV plus CAPE in the treatment of ER+ mBC patients failing initial therapy with NSAIs. Disclosure: G. Kourlaba: received funding from Novartis Hellas SA in order to perform the current cost-effectiveness analysis; M. Chatzikou: is employee of Novartis Hellas SA. All other authors have declared no conflicts of interest.