Bladder cancer is an increasingly important international public health problem. As a multifactorial disease, both environmental and genetic factors are involved in its development and progression. This neoplasm is a paradigm for the participation of low-penetrance genetic variants (GSTM1-null and NAT2-slow) and provides the best established gene-environment interaction in cancer (NAT2-slow * tobacco). Genetic variants in nucleotide excision and double strand break DNA repair pathways have provided promising results, ERCC2-XPD rs238406 being the most consistent variant associated with an increased of bladder cancer risk, by itself and by interacting with tobacco. Variants in other pathways such as cell-cycle control, 1-C metabolism and inflammation have been studied, although the results are inconsistent. Three very large whole genome association studies are being undertaken using the same genotyping platform. Their results will be available soon. Genetic variants have seldom been considered as markers of prognosis or response to therapy in this tumour. The results of these studies are inconclusive. Other issues that need to be addressed are the role of genetic variants in different population subgroups--defined by ethnicity, gender and age, among others--and the association with bladder cancer subphenotypes according to clinical, pathological and molecular characteristics of the tumour. This endeavour can only be achieved by integrating multidisciplinary tools and information. Can this information be applied better to identify high-risk populations? Can the information be used to better assess prognosis or predict response to therapy? These questions require large, well-designed, multicentre studies to be conducted. Funding agencies should be aware of these needs.