OBJECTIVES: To facilitate the comparison of net benefits of cancer treatments in clinical trials by developing and validating a system to convert data from a QoL instrument into precise and optimally weighted global QoL measures and utilities. METHODS: Two-hundred cancer patients completed the Utility-Based Questionnaire-Cancer (UBQ-C), a validated 34-item cancer-specific instrument which includes scales of health status, overall QoL, disabilities and distresses. Patients were interviewed to elicit time trade-off (TTO) utilities for their own health states. A global QoL measure was derived from a weighted combination of the UBQ-C scales based on linear regression of health status on the individual scales. An equation to convert global QoL into utility was derived. Validity was examined using baseline data from a RCT of advanced breast cancer chemotherapy (n=290). RESULTS: The weighted global QoL measure was more precise than the single-item QoL scales. Median scores (IQR) were much lower for the weighted global QoL measure: 0.77 (0.65,0.85) than for the direct TTO utility: 0.98 (0.85,1.0). The best model to predict utility from weighted global QoL was a power transformation: TTO=1-(1-global QoL)2.1. The measures discriminated between RCT subjects with good and poor performance status: mean (95% CI) derived utility scores for ECOG 0-1=0.91 (0.89,0.92), ECOG 2-3=0.75 (0.68,0.81), p<0.0001. Subjects with lower scores had worse survival: HR 1.6 (95%CI 1.2,2.0), p<0.0002. CONCLUSION: QoL measures underestimate utilities. The weighted global QoL and utility scores had discriminative and predictive validity in advanced cancer. Our work enables QoL data obtained with a simple questionnaire to be converted into optimally weighted measures that can be used in clinical trials to: describe the net effect of cancer treatments on QoL; evaluate trade-offs between quality and quantity of life using quality-adjusted survival analysis; and do cost-effectiveness analyses based on cancer patients' preferences.