Content area
Abstract
Background. Cigarette smoking and arsenic exposure are established risk factors for bladder cancer, but the mechanisms by which these exposures induce or promote bladder carcinogenesis are unclear. TP53 alterations are common molecular alterations observed in bladder tumors thought to arise from tobacco-related carcinogens and perhaps arsenic exposure. Thus, we examined the prevalence, type, and location of TP53 alterations in bladder tumors in relation to cigarette smoking and cumulative arsenic exposure from drinking water. We also examined potential etiologic heterogeneity with respect to these exposures between TP53 -altered and TP53-wild type bladder cancer.
Methods. We conducted a population-based study of 509 urothelial cell carcinoma cases and 992 controls from Maine and Vermont, enrolled between 2001 and 2004. Mutations were identified by direct sequencing of exons 5–8 of TP53 using DNA from tumor tissue, and P53 protein expression was assessed through immunohistochemical staining of tumor tissue. Polytomous regression was used to examine associations of cigarette smoking and arsenic exposure with TP53-altered and TP53-wild type bladder cancer cases compared to controls.
Results. The prevalence of TP53 mutations and P53 protein overexpression did not significantly differ between never and ever smokers, or between those with low and high (≥ 37.04 mg) cumulative lifetime arsenic exposure. G:C→T:A transition mutations at CpG sites were more frequent among never smokers, while non-CpG transitions were more frequent among smokers. We confirmed the presence of hotspot mutation sites at codons 273, 280, and 285 in bladder cancer overall, and observed a novel site at codon 132 only among smokers. After restricting the analyses to the subset of cases positive for both mutations and P53 overexpression, G→T transversion mutations occurred only among smokers. There was no etiologic heterogeneity for the associations of either smoking or arsenic exposure by bladder tumor molecular subtype, based on the presence or absence of either mutations or protein overexpression. However, when further stratifying the analyses by combinations of tumor stage and grade thought to reflect distinct bladder cancer phenotypes, cigarette smoking was more strongly associated with mutation-positive than mutation-negative bladder cancer in the subgroup of non-invasive, high-grade tumors.
Conclusions. These findings suggest that TP53 mutational patterns may differ between smokers and non-smokers, and that cigarette smoking may influence the occurrence of TP53 mutations in the histopathologic subgroup of noninvasive yet high-grade bladder tumors.
