[998] Morphometric Analysis of Normal Endocervical Glands with Emphasis on Applicability To Help Distinguish Adenocarcinoma In Situ from Early Invasion

SR Jannapureddy, S Logani. Emory University Hospital, Atlanta, GA

Background: Distinguishing cervical adenocarcinoma in-situ (AIS) from early invasive adenocarcinoma (EIA) of the cervix can be challenging because conventional criteria for invasion such as single cell infiltration or desmoplastic response are often lacking in a significant number of cases. Additional criteria proposed, but not uniformly agreed upon include 1) architecturally complex glands with confluent growth pattern; 2) cribriform growth pattern of malignant epithelium; 3) presence of small glands below the deep margin of normal glands and 4) presence of malignant appearing glands in close proximity to thick walled blood vessels (TV). If these criteria are to be relied upon to establish invasion, knowledge of the normal endocervical glandular epithelium relative to depth and association to large blood vessels is essential.
Design: We analyzed specific characteristics of normal endocervical glands with respect to their distribution and association to TV in the cervical stroma. Sections of the transformation zone and the lower endocervical canal from 29 women without glandular neoplasia were reviewed. The following parameters were recorded: 1) perpendicular depth of the deepest endocervical glands from the surface; 2) distance of the deepest endocervical gland from the nearest thick walled blood vessel (defined as blood vessel with a discernable muscle wall); 3) depth of this blood vessel from the surface and 4) the distance of other glands from its nearest thick walled vessel if less than the distance of the deepest gland from its vessel was also recorded.
Results: The age of the 29 patients ranged from 33 to 75 yrs (average 47 yrs). The deepest endocervical gland was recorded at a depth of 5.62 mm (range: 1.57-5.62 mm; average: 3.36 mm). The distance of the deepest gland from the nearest TV ranged from .08 to 4.5 mm (average: 0.94 mm), with the majority being less than 2.0 mm (27/29; 93%). In cases where an additional TV distance was recorded, the vessel was noted to be 0.06 to 0.7 mm away (average: 0.23 mm).
Conclusions: Our data suggest that normal endocervical glands can be observed at a considerable depth from the surface and provide a reference point to help distinguish AIS from EIA. Glands appearing to be involved by AIS deeper than 3.36 mm and certainly beyond 5.6 mm should strongly suggest the possibility of invasion. The close proximity of normal endocervical glands to TV suggests that glands lined by AIS type epithelium should not be construed as de facto evidence of early invasion.
Category: Gynecologic

Monday, March 9, 2009 1:00 PM

Poster Session II # 143, Monday Afternoon