OBJECTIVES: Treatment decisions for patients diagnosed with advanced non-small cell lung cancer (NSCLC) require assessment of the risks and benefits of treatment. We sought to understand the patient perspective when making these choices by estimating utilities (preference weights and relative importance weights) for different hypothetical NSCLC treatment profiles. METHODS: One hundred patients with NSCLC were recruited in the UK and completed a self-administered, web-based conjoint analysis questionnaire. The questionnaire presented patients with pair-wise choices of NSCLC treatment profiles which systematically varied the duration of progression-free survival (PFS), severity of disease symptoms, severity of the treatment-related adverse events (diarrhea, fatigue, fever/infection, nausea/vomiting and rash) and mode of treatment administration (intravenous versus oral). Preference weights were estimated using a random-parameters logit. Importance weights were calculated from the model coefficients. RESULTS: Eighty-nine patients (73% male) completed all choice tasks appropriately. The highest utility was associated with treatments that increased PFS and improved disease-related symptom severity from severe to mild (10.0; 95% CI: 6.1, 13.9). However, patients preferred a reduction in PFS if disease-related symptoms were severe. Utility was higher for treatments that had no fatigue (5.0; 95% CI: 2.7, 7.3), no diarrhea (2.8; 95% CI: 0.7, 4.9), no fever/infection (2.1; 95% CI: 0.2, 4.1), no nausea/vomiting (2.1; 95% CI: 0.1, 4.1), no rash (2.0; 95% CI: 0.2, 3.9) and for oral administration instead of infusion (1.8; 95% CI: 0.0, 3.6). Patients were found to be indifferent to treatments associated with mild diarrhea and mild nausea/vomiting. Avoiding moderate fatigue was half as important as increasing PFS by seven months with improvement in symptom severity from severe to mild. CONCLUSIONS: NSCLC patients attributed the highest utility to treatment efficacy. Treatments that increased PFS with low severity of disease related symptoms, no fatigue and oral administration were preferred.