Many articles have reported the caveolin-1 gene to be down-regulated thus suggesting that it might be a candidate tumor suppressor gene in many tumors. However, its involvement in bladder cancer is not clear and may be depending on pathological grade. In this case-control study, the association of Cav-1 polymorphisms with bladder cancer risk in a central Taiwanese population was investigated. Three hundred and seventy-five patients with bladder cancer and the same number of age- and gender-matched healthy controls were genotyped. There were significant differences between bladder cancer and control groups in the distributions of their genotypes (P = 1.0 x 10(-12) and 0.299) and allelic frequencies (P = 1.4 x 10(-14) and 6.2 x 10(-3)) in the Cav-1 G14713A (rs3807987) and T29107A (rs7804372) polymorphisms, respectively. As for haplotype analysis, subjects who had GG/AT or GG/AA at Cav-1 G14713A/T29107A showed a decreased risk of bladder cancer compared to subjects with GG/TT, while those of any other combinations were of increased risk. There were joint effects of Cav-1 G14713A and T29107A genotypes with smoking status on individual bladder cancer susceptibility. This is the first report providing evidence that Cav-1 was involved in bladder cancer in that the A allele of the Cav-1 G14713A is risky, the A allele of the Cav-1 T29107A is protective, and AA/TT on these two polymorphisms may be the most risky haplotype for the development of bladder cancer and may be novel useful genomic markers for early detection of bladder cancer.