Kaposi sarcoma (KS) prior to 1980 was a rare indolent disorder seen in elderly men of Eastern European descent. Today it is the most common neoplasm seen in patients with AIDS and has been estimated to contribute to 25% of the deaths. Chemotherapy is the mainstay treatment of KS in patients with extensive or progressive disease involving the visceral organs and lymph nodes. Two agents, liposomal daunorubicin (Daunoxome(r)) and liposomal doxorubicin (Doxil(r)), have replaced treatment with the toxic combination of ABV. OBJECTIVES: Develop a decision analysis model comparing the cost-effectiveness of Daunoxome( and Doxil( in severe KS. METHODS Probabilities used to determine patient responses and complication rates were obtained from clinical trials and data reflecting our KS population. The model assumed treatment of severe KS (CD4 (150) patients not previously treated with chemotherapy. It includes direct cost associated with drug acquisition, physician monitoring, administration of product, drug administration cost and treatment of complications. Probabilities and direct cost data for nausea and vomiting, peripheral neuropathy, neutropenia, alopecia and serious opportunistic infections were included in the model. Drug cost reflect dosing regimens currently in use, Daunoxome( 40mg/m2 every 2 weeks or Doxil 20mg/m2 every 2 weeks for 6 cycles. Effectiveness was determined using the median duration of response to therapy reported in clinical trials. RESULTS: The cost effectiveness ratio for Daunoxome( is $24,263/median remission year versus $32,767 for Doxil(. One way sensitivity analysis did not substantially alter the results. The model also demonstrates Daunoxome(r) to have a marginal cost advantage of $2,176 (20.9%) over Doxil(. CONCLUSION: The model identifies Daunoxome( as the most cost effective and least expensive agent in the treatment of KS.