[956] Role of GLUT1 as a Marker for Endometrial Malignancy
PR Cohen, DE Burstein, PW Schlosshauer. The Mount Sinai School of Medicine, New York, NY
Background: Among cases of endometrial hyperplasia, complex atypical hyperplasia carries the highest risk of being associated with or progressing to endometrioid carcinoma (20-30%). However, the diagnosis of atypia is fraught with considerable inter- and intraobserver variability. Also, up to 3% of cases without atypia may progress to carcinoma. The transmembranous glucose transporter-1 protein (GLUT1) is overexpressed in many carcinomas. We hypothesized that GLUT1 might be a marker for high risk of concomitant or subsequent endometrial carcinoma in biopsy material that shows hyperplasia but no morphologically diagnostic carcinoma. We correlated immunohistochemical GLUT1 expression in biopsies of hyperplastic endometrium with the presence of endometrial carcinoma in subsequent hysterectomy specimens. Design: We analyzed 17 uteri removed for a diagnosis of endometrial hyperplasia based on a preceding biopsy. The uterus had endometrial carcinoma in 7 cases, while 10 cases showed either persistence or resolution of hyperplasia but no carcinoma. All biopsy material and selected hysterectomy sections were examined for GLUT1 immunoreactivity. Staining intensity was graded from weak (1+) to strong (3+), and the percentage of positive cells was estimated. Membranous GLUT1 expression in the biopsy was correlated with the presence or absence of carcinoma in the uterus. Fisher's Exact Test was used for statistical analysis. Results: 5 of the 7 endometrial biopsies associated with carcinoma in the subsequent hysterectomy showed positivity and of these 4 had at least focal 2+ staining. Of the 10 biopsies not associated with carcinoma, only 2 had any staining which was weak and present in less than 10% of the specimen. The presence of any GLUT1 staining in the biopsy was associated with a high risk of carcinoma in the uterus, almost reaching statistical significance (p = 0.058). Moderate staining (>1+) in biopsies was statistically significant (p value = 0.0147). Conclusions: GLUT1 was expressed in the majority of biopsies with endometrial hyperplasia whose subsequent hysterectomy specimen contained endometrial carcinoma. It was only infrequently expressed in hyperplastic biopsies when carcinoma was absent from the hysterectomy specimen. The presence of moderate staining in biopsies is a statistically significant predictor of carcinoma in subsequent hysterectomies. Our results suggest that GLUT1 may be a marker of high risk for concomitant or subsequent endometrial carcinoma in biopsies showing endometrial hyperplasia but no morphologically diagnostic carcinoma. Category: Gynecologic
Wednesday, March 11, 2009 1:00 PM
Poster Session VI # 141, Wednesday Afternoon
|