Presentation Authors: Xu Chen*, Jian-bing Fan, Guangzhou, China, People's Republic of, Darryl Luke Irwin, San Diego, CA, Ming Huang, Chanjuan Wang, Zeyu Jiang, Jingtong Zhang, Meng Yang, Xia Li, Guangzhou, China, People's Republic of, Yu Zhou, San Diego, CA, Tianxin Lin, Guangzhou, China, People's Republic of
Introduction: The gold standard method used to detect and monitor bladder cancer is cystoscopy, which leads to high diagnostic costs and a high patient burden. The currently available non-invasive approaches either show unsatisfying sensitivity in low-grade tumors or possess limited specificity. This study aimed to develop a new non-invasive strategy based on urine methylation biomarker to diagnose bladder cancer effectively.
Methods: To identify the bladder cancer specific markers, genome-wide methylation analysis was performed on 21 paired bladder tumors and normal tissues from The Cancer Genome Atlas (TCGA) cohort, and 20 paired bladder tumors and normal tissues from Sun Yat-sen Memorial Hospital (SYMH) cohort. Next, Time of Flight Mass Spectrometer (TOF-MS) analysis was performed to access markers efficiency in large cohort. We enrolled 423 patients diagnosed with bladder cancer, 82 healthy participants and 276 controls with benign diseases from SYMH cohort. DNA of 781 urine samples extracted from the urinary cells and bisulfite modificated, and methylation status was analyzed using TOF-MS. In training set, 314 urine samples were used to develop a gene classifier by LASSO regression. An additional 467 urine samples were analyzed using the 5-gene classifier for independent validation.
Results: We firstly identified 40 significant methylation markers in a combined analysis of TCGA cohort and SYMH cohort. 28 of 40 markers showed good concordance (87.0%) in the assessments of patient tumor tissues and urine. The TOF-MS based methylation classifier yield a good AUC of 0.915 (with a sensitivity of 89.3% and a specificity of 83.6%) in the training group. The validation group also showed an AUC of 0.903, with a sensitivity of 88.1% and a specificity of 82.4%. The methylation classifier also exhibited a significantly improved sensitivity compared to voided urine cytology and FISH. In addition, this approach could analyze 128 samples in max one time and provide clinical report less than 2 days, which facilitated doctor make decision on time.
Conclusions: TOF-MS based urine DNA methylation classifier showed high accuracy and strong diagnostic power, even in early-stage/low-grade tumor patients. Therefore, it may be used as a non-invasive fast and high-through approach for diagnosis and recurrence surveillance in bladder cancer prior to the use of cystoscopy, which would greatly reduce the burden on patients.
Source of Funding: This study was supported by the National Natural Science Foundation of China (Grant No. 81825016, 81702523)