Cost-effectiveness analysis provides information on the potential value of new cancer treatments, which is particularly pertinent for decision makers as demand for treatment grows while healthcare budgets remain fixed. A range of decision analytic modelling approaches can be used to estimate cost-effectiveness. This study summarises the key modelling approaches considered in oncology, alongside their advantages and limitations. A review was conducted to identify single technology appraisals (STAs) submitted to the National Institute for Health and Care Excellence (NICE) and published papers reporting full economic evaluations of cancer treatments published within the last 5 years. The review was supplemented with the existing methods literature discussing cancer modelling. In total, 100 NICE STAs and 124 published studies were included. Partitioned-survival analysis (n=54) and discrete-time state-transition structures (n=41) were the main structures submitted to NICE. Conversely, the published studies reported greater use of discrete-time state transition-models (n=102). Limited justification of model structure was provided by authors, despite an awareness in the existing literature that the model structure should be considered thoroughly and can greatly influence cost-effectiveness results. Justification for the choice of model structure was limited and studies would be improved with a thorough rationale for this choice. The strengths and weaknesses of each approach should be considered by future researchers. Alternative methods (such as multi-state modelling) are likely to be utilised more frequently in the future, and so justification of these more advanced methods is paramount to their acceptability to inform healthcare decision making.