Abstract During the last decade, nearly 60 studies have addressed possible associations between various genetic sequence alterations and risk of adverse reactions after radiotherapy. We report here an overview of these studies with information on the genetic variants, tumour type, number of patients included, the endpoint studied, the mechanism(s) by which the candidate genes are involved in the pathogenesis of normal tissue toxicity, and odds ratios (ORs) for candidate variants. Though many positive results have been reported, inconsistent findings and non-replication of previous results have frequently occurred. This can presumably be attributed to certain methodological shortcomings including lack of statistical power to detect small effect sizes. Based on theoretical considerations and experiences from other scientific fields, we discuss how future studies should be designed in order to successfully unravel the genetics of normal tissue radiosensitivity. We propose a model of the allelic architecture that may underlie differences in normal tissue radiosensitivity. Genome wide association studies have proven a powerful tool to identify novel loci that affect various phenotypes. Nonetheless, genome wide association studies are extremely demanding in terms of sample size. Furthermore, certain limitations still relate to this kind of studies, emphasizing the need for international consortia such as the ESTRO GENEPI.