Randomized clinical trials in medical oncology depend on an accurate assessment of tumor response to therapy. In the case of solid tumors, radiological imaging has assumed a dominant role in assessing response, because it can provide reliable and mostly objective measurements of tumor size. RECIST (Response Evaluation Criteria in Solid Tumors) and Lugano criteria (Response assessment in Lymphomas) are the most commonly used image-based methodologies for assessing tumor response to therapy in the field of clinical trials. These methodologies are based on a set of principles and rules, that are used to assign the category of response, in an reliable, most objective and reproducible fashion. In this talk I will review the principles and assumptions of these methods, as well as their limitations, with examples of daily challenges in practice.