OBJECTIVES: To develop multi-attribute utility instruments (MAUIs) based on two widely used cancer-specific quality of life (QOL) instruments, the EORTC QLQ-C30 and FACT-G. METHODS: To determine the health state classification systems (HSCSs) for the two MAUIs, we conducted secondary analyses on pooled data from international sources (QLQ-C30, n=2616; FACT-G, n=6912), using established criteria to select a subset of dimensions and items for the HSCSs. These were then valued using discrete choice experiments (DCEs) administered to population-based online panels in Australia (QLU-C10D and FACT-8D), Germany and France (QLU-C10D), with quota-sampling to achieve representativeness by age and sex.  Conditional logit models, allowing for intra-individual correlation, were estimated to generate country-specific value sets.  RESULTS: Based on a suite of psychometric analyses and input from our multidisciplinary research team, HSCSs were devised for the QLQ-C30 (QLU-C10D, 10 dimensions: long/short walk, work limitations, depression, family/social, tiredness, nausea/vomiting, pain, sleep, appetite, constipation/diarrhoea) and the FACT-G (FACT-8D, 8 dimensions: nausea, pain, fatigue, sleep, work, worry condition will get worse, sad, family support). Four valuation surveys have been conducted in Australia (QLU-C10D, n=1791 and FACT-8D n=1471), Germany (QLU-C10D, n= 1000) and France (QLU-C10D, n=1000).  The results generally reflect the intended monotonic structure, extra years of life are viewed favourably, and movements away from no problems in each dimension are generally valued negatively.  Results across countries are generally similar, but different enough to warrant country-specific value sets. CONCLUSIONS: An advantage of adapting existing QOL instruments into MAUIs is that this reduces patient burden prospectively and allows retrospective conduct of cost-utility analysis, based on previously collected QLQ-C30 or FACT-G data. These two new MAUIs allow direct valuation from widely used cancer-specific QOL instruments, and may be able to capture health states that are uniquely important in oncology – an assertion that must now be tested in head-to-head comparisons with generic MAUIs.