OBJECTIVE: To assess the cost-effectiveness of gefitinib ('Iressa') compared to BSC in patients with refractory advanced NSCLC in the UK National Health Service (NHS). METHODS: A probabilistic model was developed to assess the cost-effectiveness (and associated uncertainty) of gefitinib compared with BSC in patients with refractory advanced NSCLC in the UK NHS. Efficacy data were drawn from two independent sources: data for gefitinib were derived from IDEAL II (patients refractory to platinum and docetaxel) and data for BSC were derived from a literature review (BSC arm of a randomised controlled trial in second-line advanced NSCLC). Cost data were collected from the perspective of the UK NHS. In the absence of a UK price for gefitinib, the pre-approval sales price in France (1950€) was converted into UK prices (approximately £1300). Resource utilisation and cost data for gefitinib were derived from published data and expert clinical opinion. Conservative assumptions were made for gefitinib in the base-case analysis; namely that gefitinib patients consumed the same supportive care resources as patients treated with a combination of mitomycin, ifosfamide and cisplatin, which is approximately 20% higher than for BSC. RESULTS: Additional costs of gefitinib compared with BSC were estimated to be approximately £5000. The additional life expectancy was estimated to be approximately 3 months giving an incremental cost per life year gained (LYG) ratio of approximately £22k (based on mean of probabilistic simulations). The ratio falls to approximately £17k per LYG when equivalent palliative care costs are assumed. CONCLUSIONS: According to this model, the results show that gefitinib is likely to be a cost-effective strategy in the UK for the treatment of advanced NSCLC patients refractory to platinum and docetaxel compared with best supportive care. The model’s conservative assumptions would further support this conclusion. 'Iressa' is a trademark of the AstraZeneca group of companies.