Background: The aim of this study was to evaluate the potential use of the UroVysion® fluorescent in situ hybridization test (U-FISH) to stratify the risk of urothelial carcinoma (UC) in patients with a diagnosis of "atypical urothelial cells" (AUC) in urinary tract cytology (UTCy).
Methods: Using a histologic diagnosis of UC and respectively of high grade UC (HGUC) within 12 months of the index UTCy as a reference standard, we determined the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy of U-FISH for patients with AUC diagnosed 2008 to 2014.
Results: Of the 377 patients with AUC, 62 (16.45%) were diagnosed with UC (29 low grade UC and 33 HGUC) within 12 months. U-FISH were uninformative in 45 (11.94%), positive in 63 (16.71%) and negative in 269 (71.35%). UC was diagnosed more frequently in patients with positive than in those with negative U- FISH results (31/63, 49.21% vs. 25/269, 9.29%, P < 0.0001). The sensitivity, specificity, PPV, NPV and accuracy of U-FISH in the setting of AUC were 44.64%, 81.82%, 47.17%, 80.25%, and 71.91% for UC and respectively 48.39%, 78.77%. 28.3%, 89.81%, and 74.29% for HGUC. U-FISH showed a high false positive rate (28/53, 52.83%) that remained high even after extended follow-up, arguing against "anticipatory positive" results.
Conclusions: U-FISH allows risk stratification in patients with AUC. However, its usefulness is diminished by the high false-positive rate, making it important to interpret U- FISH results in the patient's clinical context. Diagn. Cytopathol. 2017;45:481-500. © 2017 Wiley Periodicals, Inc.
Keywords: FISH; STARD; UroVysion®; sensitivity; specificity; urine cytology; urothelial carcinoma.
© 2017 Wiley Periodicals, Inc.