The widespread use of several new non-cytotoxic drugs and the significant improvements in functional imaging highlights a number of difficulties in monitoring, interpreting and predicting treatment response in clinical practice. Certain guidelines for disease assessment after therapy are already available: the traditional Response Evaluation Criteria in Solid Tumours guidelines based on tumour size variations using conventional imaging technologies, the recent combined method developed by Choi and colleagues in gastrointestinal stromal tumour treated with tyrosine kinase inhibitors based on tumour density variations using computed tomography (CT), and the European Organization for Research and Treatment of Cancer criteria based on tumour glucose metabolism variations using fluorodeoxyglucose (FDG) positron emission tomography (PET). At the moment combined PET/CT response criteria are still not available. A number of new PET compounds other than FDG are also currently being developed to visualize specific cellular and molecular tumour pathways but their role in assessment and prediction of cancer treatment response has not yet been thoroughly investigated in a large series. However, in clinical practice many oncologists treat cancer patients with targeted therapies or chemotherapy and evaluate the response using conventional or functional imaging without appropriate and standardized guidelines. The aim of this study was to present a selection of clinical cases that illustrate the usefulness of new PET tracers and efficacy evaluation of new drugs. In the era of molecular imaging and molecular therapies, these cases highlight the urgency to develop new criteria for treatment assessment and the exigency of correctly interpreting the biological information obtained from new technologies, and introduce new concepts that require further investigation in clinical trials.