In recent years, we have witnessed growing interest in the methodology of clinical trials with molecular-targeted agents. In phase I studies, alternative end points to toxicity have been proposed to define the optimal biological dose: the identification of a 'target effect', the measurement of 'surrogates' for biological activity and the assessment of drug plasma levels. However, these end points are not routinely incorporated into the study design and have rarely formed the primary basis for dose selection. In phase II studies, response rate remains the preferred end point in the early evaluation of new drugs. However, this approach might lead to rejection of potentially useful drugs when significant tumor shrinkage cannot be demonstrated. Therefore, a number of alternative end points have been proposed for agents that are not expected to cause a major tumor regression: time to progression, progression-free survival, overall survival, early progression rate and growth modulation index. In phase III trials, where efficacy in terms of survival remains the most important goal of the research, the major issues are the adequate selection of patients and the optimal clinical setting of evaluation of drugs. In conclusion, many important questions regarding the methodology of clinical research with target-based agents remain open and need to be defined by research in the near future.