The Diagnostic Performance of a Four-Gene Digital Droplet PCR Panel for Urine Liquid Biopsy in Urothelial Bladder Cancer.

Background: Urothelial bladder carcinoma (UBC) is a disease that lacks robust non-invasive laboratory biomarkers. Recently, urine liquid biopsy has emerged as a promising tool for diagnosis and surveillance of patients with these tumors. The aim of this study was to evaluate the diagnostic potential of a urinary tumor DNA detection panel, which included eight common point mutations in TERT, GPR126, FGFR3, and PIK3CA genes, in UBC. Methods: The study included patients with histologically confirmed UBC (n = 88) and patients with cystitis, bladder leiomyomas, or other non-malignant conditions (control group; n = 72). DNA was extracted from whole urine specimens. ddPCR analysis was performed using the Bio-Rad QX200 AutoDG ddPCR system. Results: Urinary tumor DNA detection panel demonstrated a sensitivity of 78.4% and a specificity of 100% (AUC-ROC = 0.892). Detection rates for the analyzed mutations were the following: 54.5%, 37.5%, 28.4%, and 38.6% for TERT, GPR126, FGFR3, and PIK3CA, respectively. Pairwise comparisons of mutant allele fractions (MAFs) for samples simultaneously positive for ≥2 mutations revealed an absence of significant differences (p > 0.05), except for the pair of FGFR3 vs. PIK3CA (p = 0.03). MAFs were not associated with any clinical and demographic features (p > 0.05), with the only exception being the tumor size: patients with tumors larger than 2.16 cm3 had higher MAFs than the rest (23.4 [1.8; 46.3] vs. 1.6 [0; 24.6] %, respectively, p = 0.02). Conclusions: Upon further validation, the presented tumor DNA detection panel for ddPCR might become a useful tool for diagnostic purposes in UBC.