Purpose: We aimed to investigate the methylation pattern in bladder cancer and assess the diagnostic potential of such epigenetic changes in urine.

Experimental design: The methylation status of 7 genes (RARbeta, DAPK, E-cadherin, p16, p15, GSTP1, and MGMT) in 98 cases of bladder transitional cell carcinoma and 4 cases of carcinoma in situ was analyzed by methylation-specific PCR. Twenty-two cases had paired voided urine samples for analysis.

Results: In transitional cell carcinoma tumor tissues, aberrant methylation was frequently detected in RARbeta (87.8%), DAPK (58.2%), E-cadherin (63.3%), and p16 (26.5%), whereas methylation of p15 (13.3%), GSTP1 (5.1%), and MGMT (5.1%) is not common. No association between methylation status and grading or muscle invasiveness was demonstrated. In 22 paired voided urine samples of bladder cancer, methylation of DAPK, RARbeta, E-cadherin, and p16 could be detected in 45.5%, 68.2%, 59.1%, and 13.6% of the cases, respectively. The sensitivity of methylation analysis (90.9%) was higher than that of urine cytology (45.5%) for cancer detection. Methylation of RARbeta(50%), DAPK (75%), and E-cadherin (50%) was also detected in carcinoma in situ. In 7 normal urothelium samples and 17 normal urine controls, no aberrant methylation was detected except for RARbeta methylation in 3 normal urothelium samples (42.9%) and 4 normal urine samples (23.5%), respectively.

Conclusions: Our results demonstrated a distinct methylation pattern in bladder cancer with frequent methylation of RARbeta, DAPK, E-cadherin, and p16. Detection of gene methylation in routine voided urine using selected markers appeared to be more sensitive than conventional urine cytology.