Urinary bladder cancer is the ninth most common cancer worldwide (Roth et al. 2012; Golka et al. 2011). The strongest known risk factors are cigarette smoking, occupational exposure to bladder carcinogens and male gender (Golka et al. 2012a, b; Ovsiannikov et al. 2012). It is well established that slow acetylation (NAT2) and a deletion variant of glutathione S-transferase M1 (GSTM1) are associated with increased bladder cancer risk (Schwender et al. 2012; Bell et al. 1993; Cartwright et al. 1984; Golka et al. 1996; Kempkes et al. 1996; Hengstler et al. 1998). Recently, genome-wide association studies have identified further polymorphisms, and their influence was confirmed in independent case–control series (Kiemeney et al. 2008, 2010; Rothman et al. 2010; Rafnar et al. 2009, 2011; Wu et al. 2009; Golka et al. 2009; Lehmann et al. 2010; Selinski et al. 2011; Binder et al. 2012; Bolt 2012). All the information currently known for the 13 polymorphisms has recently been summarized (Selinski 2012). Although it is highly likely that the most influential individual polymorphisms have already been identified, it remains unknown whether these polymorphism interact with one another leading to a higher risk of bladder cancer, in the case where an individual carries several high-risk alleles.

In the current issue of PLOS one, Holger Schwender and colleagues from TU Dortmund University established a technique to calculate whether there is interaction among the high-risk alleles and how this influences bladder cancer risk (Schwender et al. 2012). The authors used six of the previously identified single nucleotide polymorphisms (SNPs) in a bladder cancer case–control series of 1,595 bladder cancer cases and 1,760 controls. The authors tested all possible SNP combinations and their stability by using a bootstrap technique. The take home messages of this study are as follows:

  • Certain high-risk alleles interact; however, their interaction is less than additive.

  • Different SNP combinations are relevant in smokers and non-smokers.

  • The top scoring ‘smoker variants’ include variants of GSTM1 and UGT1A, both involved in detoxification of cigarette-smoke carcinogens.

  • The top scoring ‘non-smoker variants’ include APOBEC3A and cMYC, both involved in the proliferation and maintenance of DNA integrity.

  • The highest stable combination effect results in an odds ratio of approximately 2.0 (combined influence of the high-risk alleles of rs 9642880, rs 710521 and rs 1014971 in non-smokers).

  • However, the degree to which genetic disposition contributes to cancer risk still seems to be smaller than that of environmental factors. The odds ratio of cigarette smoking is approximately 3.5 and is therefore still higher than the combined influence of the high-risk alleles.

The study of Schwender et al. represents an important milestone on the path to calculate the cancer risk associated with entire genomes. A limitation of the current study is the relatively low number of cases (1,595), which will at most allow the analysis of three-way interactions. Therefore, an important next step in this field of research will be to conduct a SNP interaction study that includes all cell cases and controls available worldwide.