Immunotherapy of cancer is still in the early stages of development although almost a century has passed since initial attempts were made to stimulate the immune system in order to destroy malignant cells. Historically, a variety of specific and non-specific immunostimulatory strategies have been administered with only modest clinical success. However, recent advances in tumour immunology, most notably the identification of new tumour antigens and the better understanding of antigen processing and presentation to avoid or break immune tolerance, have paved the way for the development of a variety of novel and specific vaccine approaches. The most important and widely used are whole-cell vaccines, dendritic-cell-based immunotherapy and peptide vaccines. The first wave of clinical trials has revealed that, in general, such vaccination strategies are safe. However, clear examples of clinical responses, especially in conjunction with vaccine-induced immune responses, are still rare. Most clinical trials are too small to allow for comments on the efficacy, and the cohort of patients studied is too heterogeneous with regard to immune status. Therefore, standardised techniques for the accurate assessment of the individual immune phenotype before and during the trial are needed to allow for the identification of the sub-group of patients who will respond favourably to treatment. The precise definition of immune parameters in these patients will then lead the way for optimised treatment procedures that might even be beneficial for a larger group of cancer patients.