[422] Comparison of Urine Cytology and Fluorescence In Situ Hybridization in Upper Urothelial Tract Samples.

Jordan P Reynolds, Jesse S Voss, Benjamin R Kipp, RJeffery J Karnes, Aziza Nassar, Thomas J Sebo, Amy C Clayton, Jun Zhang, Michael R Henry, Kevin C Halling. Mayo Clinic, Rochester, MN

Background: Fluorescence in situ hybridization (FISH) with the Urovysion® probe set (Abbot Molecular Inc., Des Plaines, IL) has been shown to detect chromosomal abnormalities in urine cytology with high sensitivity and specificity for bladder urothelial carcinoma (UC). Surveillance of the upper urinary tract (UT) urothelium remains challenging. The aim of this study is to compare the performance of cytology and FISH in the UT.
Design: UT urine samples (n=112) were collected from 61 patients (48 male, 13 female, mean age 66.2 years) via cysto/ureteroscopy from 2003-2009. Cytologic interpretation and FISH were performed using probes to chromosomes 3, 7, 17 and 9q21. Positive cytologic diagnoses were considered as positive and all other diagnoses were considered as negative. FISH was abnormal when identifying tetrasomy (≥10 cells with 3-4 copies of each probe (TETRA)) or hypertetrasomy/polysomy (≥4 cells with gains in at least two probes with one showing ≥5 signals (HT)). Biopsy diagnosis of UT cancer (n=21) within 2 years of the cytology/FISH result was the gold standard. In cases without a biopsy, a negative urogram or ureteroscopy with at least 1 year of follow up was considered negative (n=91). The mean follow up time was 3.2 years.
Results:

Upper Tract Specimens

	All cases		Cases with bladder UC excluded
	Sens	Spec	Sens		Spec
Cytology	38% 8/21	89% 81/91	50% 7/14		95% 53/56
FISH HT	43% 9/21	84% 76/91	50% 7/14		89% 50/56
FISH HT & TETRA	67% 14/21	53% 48/91	79% 11/14		52% 29/56
FISH HT & Cytology	52% 11/21	77% 70/91	64% 9/14		84% 47/56

Cytology and FISH HT showed increased sensitivity over cytology alone (p=0.08), but had significantly decreased specificity (p=0.009). A majority of the false positives were caused by concomitant bladder UC. After excluding bladder UC, cytology and FISH HT had higher sensitivity (p=0.15) than cytology alone, but had significantly decreased specificity (p=0.01). Inclusion of TETRA as positive identified 3 additional cancers, but produced 28 additional false positives.
Conclusions: Compared to cytology alone, performing cytology with FISH on UT samples raises the sensitivity but significantly lowers the specificity. Specimens showing tetrasomy should not be reported as positive due to the high frequency of false positives. A large proportion of false positive results were likely due to bladder cancer contaminating the UT specimen during collection. We recommend that UT cytology and FISH should be interpreted with caution in patients with concomitant bladder cancer.
Category: Cytopathology

Wednesday, March 2, 2011 1:00 PM

Poster Session VI # 69, Wednesday Afternoon