Fluorescence in situ hybridization (FISH) in the diagnosis of bladder and upper tract urothelial carcinoma: the largest single-institution experience to date.

INTRODUCTION We evaluated the UroVysion (Abbott Molecular, IL, USA) fluorescence in situ hybridization (FISH) assay for the diagnosis of urothelial cancer in patients diagnosed with or suspected to have bladder, upper tract urothelial carcinoma (UTUC), and combined upper and lower tract urothelial carcinoma (BC). MATERIALS AND METHODS A single institution retrospective analysis comparing sensitivity, specificity, positive predictive value, and negative predictive values for FISH and urinary cytology. FISH within 6 months of endoscopic evaluation were obtained from outpatient voided urine samples. Our institutional pathology department confirmed pathologic disease from specimens obtained during endoscopic evaluations for lower tract disease. For upper tract disease, disease was confirmed by retrograde ureteroscopy, biopsies of visual lesions, and site-specific upper tract cytology. RESULTS A total of 415 patients submitted FISH specimens. Overall, FISH was more sensitive than cytology 54.9% in comparison with cytology 42.2% (p = 0.01), specificity favored cytology 92.9% compared to 73.5% with FISH (p < 0.01). For BC only patients, the same significant finding of increased sensitivity and decreased specificity was identified, but for UTUC alone and combined UTUC and BC, there was no significant difference. Cytology had improved positive predictive value (PPV) over FISH, 76.9% in comparison to 64.6% (p = 0.02). Negative predictive value (NPV) also favored cytology 74.2% versus 64.9% (p = 0.02). When analyzing individual cohorts, cytology had improved PPV for BC alone patients. UTUC showed no difference for PPV and NPV. For both UTUC and BC, NPV was slightly favored for FISH over cytology 93.2% versus 91.2% (p = 0.03). CONCLUSIONS Voided urine FISH testing does offer a higher detection of urothelial carcinoma for BC compared to voided cytology; however, specificity was worse. FISH does not appear to improve detection of urothelial carcinoma in patients with either UTUC only or both BC and UTUC.