Health technology assessment (HTA) has become the key health policy instrument for managing the introduction and use of new oncology drugs in Europe. While the methodology of technology assessment, including calculations of cost-effectiveness, is applicable in principle also to oncology, the implementation in practice has its specific problems and consequences. Most of them are linked to the specific need to do the assessment early in the development, with limited data on outcome in clinical practice. Technology assessments ask for estimates of gains in mean survival, whereas trials are powered to study differences in progression-free or overall median survival. The development of targeted therapies and personalized cancer medicine offers opportunities but also increases the complexity of the assessment. Joint assessment of a diagnostic and a new treatment increases the number intervention strategies that must be considered, and thus the need for data. The translation from efficacy in trials to relative effectiveness in clinical practice must also be considered. The close link between pricing of new oncology drugs and their cost-effectiveness makes the use of technology assessment for policy decisions complicated for all stakeholders involved. But without an obvious alternative that is better, the likely future is that HTA will play an increasing role in informing policy decisions aimed at evidence-based cancer care.