1074 Background: Metastasis suppressor genes (MSGs) such as Nm23 are distinct from classical tumor suppressors in that transfection of MSGs has no effect on primary tumor size, but significantly reduces metastatic potential and growth at distant sites. Preclinical data suggest that in the absence of estrogen (ER) and progesterone receptors (PR) MPA binds to the glucocorticoid receptor, increasing Nm23, thrombospondin (TSP-1), and plasminogen activator inhibitor (PAI-1). Methods: Postmenopausal women with HR-MBC were eligible irrespective of number of prior therapies. Pts were treated with MPA 1,000 mg/day orally. MPA was increased to 1,500 mg/day in pts not achieving trough levels ≥ 50 ng/ml after 10-14 days. Plasma TSP-1, PAI-1 antigen, and PAI-1 activity were assessed serially. The primary endpoint was clinical benefit response (CBR) defined as objective response or stable disease ≥ 6 months using a serial two-stage optimal design to identify a CBR ≥ 20%. Results: 15 pts were enrolled. Median age was 58 (34-77). All pts were ER- and PR-; 14 had visceral disease. Patients were heavily pretreated with prior anthracyclines (n=14), taxane (n=13), bevacizumab (n=10), platinum (n=7), vinorelbine (n=6), capecitabine (n=5), and gemcitabine (n=4). Mean MPA trough concentration was 38.06 ± 47.58 ng/ml. All pts required dose escalation; only 7 pts ever achieved MPA trough concentrations > 50 ng/ml. Grade 3 toxicities reported by > 1 pt included dyspnea (n=3) and fatigue (n=2). Median progression-free survival was 55 days. Three pts with stable disease remained on treatment for 63, 111, and 504 days. PAI-1 antigen increased after 4 weeks (16.74 ng/ml vs. 21.78 ng/ml, p=0.0467) but there were no significant changes in PAI-1 activity or TSP-1. Conclusions: MPA had limited activity in this population; CBR did not meet the prespecified level to expand this cohort. Poor bioavailability limited exposure despite dose escalation. Nonetheless, the increase in PAI-1 antigen and prolonged stable disease in one pt is encouraging. Accrual of a second cohort evaluating MPA in combination with metronomic chemotherapy has been completed. No significant financial relationships to disclose.