OBJECTIVES: To evaluate patient-reported impairment in health states from the EuroQol (EQ-5D) assessment and derive UK preference-weighted utilities during periods of disease progression (PROG), toxicity (TOX), and time without symptoms of progression or toxicity (TWiST), based on a recent phase 3, randomized, outpatient study of first-line treatment of poor-prognosis patients with advanced renal cell carcinoma (RCC) randomized to temsirolimus, interferon, or interferon+temsirolimus. METHODS: In the clinical study, the EQ-5D was assessed at baseline, week 12, and week 32, at each grade 3 or 4 adverse event (AE), at study discontinuation, and at study end. Assessments were classified by their timing into 3 distinct periods: TOX (during grade 3 or 4 AEs or AE-related study discontinuation), PROG (during progression/relapse), and TWiST (during neither progression nor toxicity). Published UK-population preference weights were applied to patient-reported impairments on the EQ-5D domains - mobility, self-care, usual activities, pain/discomfort, and anxiety/depression - to derive utility values, consistent with EQ-5D methodology (euroqol.org). Median utility scores were compared across the periods of TOX, PROG, and TWiST. The proportion of patients reporting at least one EQ-5D domain at the highest level of severity was compared across the 3 periods, using the chi-square test. RESULTS: EQ-5D response rates were 87% (260/300) during PROG, 40% (230/570) during TOX, and 99% (278/279) during TWiST. Median utility values were 0.585 and 0.587 during TOX and PROG respectively, compared with 0.689 for the TWiST period (p<0.0001). The proportion of patients with at least one of the EQ-5D domains at the highest level of severity was 34.2% (89/260) during PROG, 38.3% (88/230) during TOX, and 18.7% (52/278) during TWiST (p<0.0001). CONCLUSION: Patients with advanced RCC reported similar impairment of EQ-5D domains during periods of toxicity and progression, significantly greater than in the absence of both; translating to corresponding differences in UK preference-weighted utility levels.