Survey of Mammaglobin Expression by Immunohistochemistry in Gynecologic Carcinomas
Ian S Hagemann, John D Pfeifer, Dengfeng Cao. Washington University School of Medicine, St. Louis, MO
Background: Mammaglobin-A (MGA) has been proposed as a sensitive and specific immunohistochemical marker for breast carcinoma. The extent to which MGA immunoreactivity differs between breast and gynecologic lesions is not known, although the differential diagnosis of breast carcinoma versus a gynecologic primary arises frequently.
Design: We performed a survey of MGA immunoreactivity in 221 gynecologic samples from 191 patients, including 27 benign gynecologic tissues (6 ectocervix, 9 endocervix, 12 endometrium), 13 endocervical adenocarcinomas, 72 endometrial adenocarcinomas, and 88 ovarian adenocarcinomas. Staining intensity was scored as weak (1+), moderate (2+), or strong (3+). The extent of staining was scored as focal (<10% of cells), patchy (10-50%), and diffuse (>50%).
Results: Gynecologic specimens often showed weak (1+) diffuse MGA staining, which we interpreted as being nonspecific. For analysis we thus considered only 2+ or 3+ staining as being significant. In benign tissues, MGA was detected in 0/6 ectocervices, 5/9 endocervices, and 8/12 endometria. Among malignancies, MGA was detected in 1 of 13 endocervical adenocarcinomas (8%; patchy distribution), 12 of 21 endometrial endometrioid carcinomas (57%; 4 focal, 8 patchy), but only 1 of 18 endometrial clear cell carcinomas (5%; focal) and 5 of 31 endometrial serous carcinomas (16%; focal). MGA was present in 6 of 15 ovarian endometrioid carcinomas (40%; 2 focal, 3 patchy, 1 diffuse), 4 of 19 ovarian clear cell carcinomas (21%; 2 focal, 2 diffuse), 1 of 16 ovarian mucinous carcinomas (6%; focal) and 12 of 36 ovarian serous carcinomas (33%; 5 focal, 6 patchy, 1 diffuse). The 3 tumors with the strongest MGA reactivity were all serous carcinomas (ovarian, focal 3+; endometrial, focal 3+; endometrial, patchy 3+). Considering all cases with 2+ or 3+ MGA reactivity, the distribution of this reactivity was focal or patchy (i.e., not diffuse) in 52 of 60 cases. There were no cases with diffuse 3+ MGA expression, as is seen in many breast carcinomas. On the other hand, diffuse 2+ MGA was seen in 4 cases including one endometrioid carcinoma of ovary, one serous carcinoma of ovary, and two clear cell carcinomas of ovary.
Conclusions: A significant proportion of gynecologic carcinomas are immunoreactive for MGA. MGA expression is not specific for breast carcinoma, and gynecologic carcinomas should be considered in the differential diagnosis of MGA-positive malignancies.
Category: Gynecologic & Obstetrics
Wednesday, March 21, 2012 9:30 AM
Poster Session V # 152, Wednesday Morning