OBJECTIVES: Despite demonstrated efficacy of anthrayclines in metastatic breast cancer (MBC), strong evidence links these agents to development of reduced left ventricular ejection fraction (LVEF) and congestive heart failure (CHF). We examined the cost-effectiveness of using pegylated lipisomal doxorubicin (PLD) to reduce cardiotoxicity. METHODS: A Markov model was developed to simulate the use of PLD vs. conventional doxorubicin (DOX) in the first 48 weeks after initiation of therapy. Our reference case was a 58 year old female newly diagnosed with MBC. Data from a large-scale clinical trial was used to model the incidence of CHF. The perspective was the provincial authority in Ontario, Canada. Costs included direct costs of chemotherapy, oncology and hospital services for managing MBC, as well as medical management for treating cardiotoxicity. Since trial data showed that patients with prior anthracycline exposure (15%) had a relative risk of 2.8 for LVEF, and 4.1 for progression to CHF, we adjusted for prior anthracycline exposure which is greater in the adjuvant setting (80% in Ontario). A previously developed cardiotoxicity risk model was used to identify high-risk patients. Analyses were performed for: 1) all patients, 2) subset of high-risk patients. RESULTS: For DOX, 40% experienced cardiotoxicity, with 18% developing CHF. These rates were 54% and 33% in high-risk patients. For PLD, no patients developed CHF, but 9% overall and 14% of high-risk patients developed LVEF. Using PLD instead of DOX was associated with an incremental cost of CAD$75,513 per patient spared CHF, $49,952 in high-risk patients. CONCLUSION: In the first treatment year, the greatest cost-effectiveness gain was avoidance of CHF in high-risk patients. In Ontario, where the vast majority of patients are high-risk, a large percentage of patients would be expected to avoid CHF by using pegylated lipisomal doxorubicin instead of conventional doxorubicin.