To address this area of clinical need, a recent article published by van der Heijden et al. in the journal Clinical Epigenetics introduced a non-invasive, sensitive method to predict bladder cancer recurrence using the combination of a three-gene (CFTR, SALL3, and TWIST1) methylation classifier and urine cytology. The researchers collected freshly voided urine samples from bladder cancer (BC) patients, controls, and patients in follow-up for bladder cancer (PFBC). The authors selected a total of seven DNA methylation markers that are known to play a role in bladder cancer (CDH13, CFTR, NID2, SALL3, TMEFF2, TWIST1, and VIM2) for PCR. A logistic regression model showed that the combination of CFTR, SALL3, and TWIST1 hypermethylation classifier has the highest accuracy of detecting bladder cancer from urine samples.
The investigators conducted a comparison between the three-gene methylation classifier and urine cytology showing that the three-gene methylation classifier has higher sensitivity, higher negative predictive value compared to urine cytology only. In the training set, the three-gene methylation classifier achieved an AUC 0.874 and an AUC 0.741 was achieved in the testing set. The combination of the three-gene methylation classifier with urine cytology in the validation set showed significantly improved the performance with an AUC of 0.86 with a sensitivity of 96%, a specificity of 40%, and a positive and negative predictive value of 56 and 92%, respectively.
This study defines a urine-based test for surveillance of NMIBC. Non-invasive liquid biopsy approaches are expected to play a bigger role in the future.
Written by Bishoy M. Faltas, MD, Weill Cornell Medicine, New York, NY
Read the Abstract: Urine cell-based DNA methylation classifier for monitoring bladder cancer