Background: The non-steroidal aromatase inhibitors (nsAIs) have been mainly employed as adjuvant therapy or as early recurrent treatment for postmenopausal breast cancer with hormone receptor-positive tumor. When its treatment fails, it is unclear which endocrine therapy including selective estrogen receptor modulator (SERM), steroidal AI (sAI), or fulvestrant is the most appropriate. Methods: Open labeled multicenter randomized comparative trial of high-dose toremifene for nsAI resistant breast cancer compared to exmestane (Hi-FAIR ex; registry number UMIN000001841) was performed from Nov 2008 to Oct 2011. Toremifene 120 mg (TOR120) or exemestane 25mg (EXE) was administered once daily. The primary end point was clinical benefit rate (CBR). The secondary end points were objective response rate (ORR), progression-free survival (PFS), overall survival (OS) and toxicity. Results: A total of 91 women were randomly assigned to TOR120 (n=46) or EXE (n=45). Two of 46 cases allocated for TOR120 refused this trial before the administration. There was no difference of patients’ characteristics between both groups. In the analysis as of median 16.9 months for an observation period, TOR120 were superior to EXE in CBR (47.5% vs. 26.7%; p=0.046) and in PFS (Hazard ratio; 0.62, 95% CI; 0.38-0.99, p=0.047). There was, however, no statistical differences in ORR (11.6% vs. 2.2%; p=0.065) and in OS (hazard ratio; 0.57, 95% CI; 0.24-1.34, p=0.19). Both treatments were well-tolerated, with no severe adverse events, although 3 of 44 women treated by TOR120 were stopped in a couple of weeks because of nausea and general fatigue which were thought to be early endocrine-related symptoms. Conclusions: Our data suggested that toremifene 120mg, as a following therapy for postmenopausal metastatic breast cancer who have failed non-steroidal AI treatment, could be potentially useful than steroidal AI.