OBJECTIVE: As the prognosis of non-small cell lung cancer (NSCLC) patients after first and second line treatment remains poor, new targeted strategies in third-line treatment are of high interest. This study estimates the cost-effectiveness of gefitinib compared to Best Supportive Care (BSC) in the Dutch health care setting. METHODS: A Markov model was designed to evaluate the lifetime clinical and economic outcomes of gefitinib treatment and BSC. The model was calibrated using clinical data from randomized clinical studies, a Delphi panel (n=10), patient chart analysis and literature for costs data. The analysis was performed from a societal perspective for a hypothetical cohort of advanced NSCLC patients, who have failed two chemotherapy regimens. Only direct costs related to the treatment of severe adverse events, radiotherapy, evaluation of disease progression and terminal care were considered. The time horizon related to mortality, estimated the costs from start of therapy until death. Both costs and effects were discounted at 4% pa. RESULTS: With an assumed difference in survival of 2.45 months between gefitinib and BSC, the model predicts survival of 0.573 life years (LY) for BSC and 0.790 LY for gefitinib. Total costs related to BSC and gefitinib treatment until death are 8,444€ and 15,272€ respectively. The average cost-effectiveness ratio of gefitinib is higher than BSC (19,326€/LY versus 14,745€/LY). The incremental cost-effectiveness ratio of gefitinib compared to BSC is 31,380€ per LYG. Applying the threshold proposed by the Institute for Medical Technology Assessment for disease with highest burden (45,000€/QALY), gefitinib is cost-effective in 73% of advanced NSCLC patients compared to BSC in third-line therapy. CONCLUSION: In addition to its convenient oral administration, its favorable tolerability profile, gefitinib is cost-effective compared to not only BSC but also compared to heart or liver transplantations. 'Iressa' is a trademark of the AstraZeneca group of companies