Purpose: A significant proportion of oncological patients experiences lack of appetite. Precise measurement is relevant to improve the management of lack of appetite. The so-called computer-adaptive test (CAT) allows for adaptation of the questionnaire to the individual patient, thereby optimizing measurement precision. The EORTC Quality of Life Group is developing a CAT version of the widely used EORTC QLQ-C30 questionnaire. Here, we report on the development of the lack of appetite CAT. Methods: The EORTC approach to CAT development comprises four phases: literature search, operationalization, pre-testing, and field testing. Phases 1–3 are described in this paper. First, a list of items was retrieved from the literature. This was refined, deleting redundant and irrelevant items. Next, new items fitting the “QLQ-C30 item style” were created. These were evaluated by international samples of experts and cancer patients. Results: The literature search generated a list of 146 items. After a comprehensive item selection procedure, the list was reduced to 24 items. These formed the basis for 21 new items fitting the QLQ-C30 item style. Expert evaluations (n = 10) and patient interviews (n = 49) reduced the list to 12 lack of appetite items. Conclusions: Phases 1–3 resulted in 12 lack of appetite candidate items. Based on a field testing (phase 4), the psychometric characteristics of the items will be assessed and the final item bank will be generated. This CAT item bank is expected to provide precise and efficient measurement of lack of appetite while still being backward compatible to the original QLQ-C30 scale.