[1091] The Detection of Endometrial Carcinoma Using Fluorescence In Situ Hybridization (FISH) and Routine Cytology on Endometrial Brushing Specimens

Emily G Barr Fritcher, Jesse S Voss, Benjamin R Kipp, Michael B Campion, Trynda N Oberg, Ekaterina V Pestova, Amy C Clayton, Kevin C Halling. Mayo Clinic, Rochester; Abbott Molecular, Des Plaines

Background: Endometrial carcinoma is the most common gynecologic malignancy in the USA; however, there is currently no screening test. A fluorescence in situ hybridization (FISH) probe set has been developed for the detection of endometrial carcinoma using endometrial brushings. The study aims were to establish thresholds for FISH positivity and to evaluate the performance of FISH and routine cytology for the detection of malignancy.
Design: Tao brush samplers (Cook OB/GYN, Spencer, IN) were utilized to collect cells from 97 hysterectomy specimens. A ThinPrep (Hologic, Bedford, MA) slide was prepared for cytology and categorized as nondiagnostic, negative, atypical, or positive for malignancy by consensus of 2 pathologists. A second slide was hybridized with FISH probes directed to 1q25, 8p11, 8q24, and 20q13. Receiver operator curves were generated to determine thresholds for FISH positivity. The hysterectomy histologic result was the gold standard.
Results: Histology, cytology and FISH results are shown in Table 1. Optimal FISH cut-off values were ≥4 cells with polysomy, ≥14 cells with gain of 1q25, and ≥10 cells with gain of 8p11, 8q24, or 20q13. The sensitivity and specificity of a positive cytology result were 57% and 100%, respectively, while including an atypical cytology diagnosis resulted in significantly (P<0.0001) increased sensitivity (95%) with decreased specificity (67%). The sensitivity of FISH was significantly higher than positive routine cytology (84% vs. 57%; P<0.0003) with similar specificity (95% vs. 100%).

Table 1: Comparison of Cytology and FISH Results by Histologic Category

	Cytology Positive (%)	Cytology Atypical + Positive (%)	FISH (%)	Cytology Positive + FISH (%)
Benign (n=39)	0 (0)	13 (33)	2 (5)	2 (5)
Complex Hyperplasia (n=3)	0 (0)	3 (100)	1 (33)	1 (33)
Grade 1 Endometrioid Carcinoma (EMC) (n=26)	10 (38)	23 (88)	20 (77)	22 (85)
Grade 2 EMC (n=12)	8 (67)	12 (100)	12 (100)	12 (100)
Grade 3 EMC (n=11)	10 (91)	11 (100)	10 (91)	10 (91)
Non-endometrioid Carcinoma (n=6)	5 (83)	6 (100)	6 (100)	6 (100)
Overall Sensitivity*	57%	95%	84%	88%
Specificity	100%	67%	95%	95%

EMC = Endometrioid Carcinoma, *Endometrial polyp and simple hyperplasia considered benign and complex hyperplasia categorized with carcinoma


Conclusions: This study suggests that FISH and routine cytology can detect endometrial carcinoma using Tao brush samplers. Additional studies are needed to validate the FISH thresholds determined in the current study and to further define a testing algorithm, such as routine cytology as a screening mechanism with reflex to FISH.
Category: Gynecologic & Obstetrics

Wednesday, March 21, 2012 1:00 PM

Poster Session VI # 199, Wednesday Afternoon