Aims: In the female genital tract CD10 has been used to assist in the evaluation of mesenchymal tumours of the uterus and in determining whether endometrial stroma is present. CD10 positivity has also been shown in cervical mesonephric remnants and this antibody has been suggested as a useful immunohistochemical marker of mesonephric lesions in the female genital tract. Calretinin has also been shown to be positive in mesonephric lesions. In this study the specificity of these two antibodies in evaluating cervical and uterine glandular lesions and the value of CD10 in determining whether stroma is endometriotic or not were investigated.
Methods and results: Cases of cervical tubo-endometrial metaplasia (TEM) (n=11), microglandular hyperplasia (MGH) (n=10), endometriosis (n=8), mesonephric remnants/hyperplasia (n=12), endocervical adenocarcinoma, usual type (n=15), mucinous variant of minimal deviation adenocarcinoma (MDA) (n=7) and mesonephric adenocarcinoma (n=3) were stained with antibodies against CD10 and calretinin. Nine cases of endometrial adenocarcinoma of endometrioid type were also stained. In all the cervical cases normal endocervical glands were negative with both antibodies except for one case with strong positive luminal staining with CD10. All cases of TEM, MGH and endometriosis were negative with CD10 and calretinin except for focal staining with CD10 in one case each of MGH (cytoplasmic staining) and endometriosis (luminal staining). Most usual endocervical adenocarcinomas were negative with both antibodies, although one exhibited focal cytoplasmic staining with calretinin and five exhibited limited luminal positivity with CD10. All MDAs were negative with both antibodies. Ten of 12 mesonephric remnants/hyperplasia showed luminal positivity with CD10 and one exhibited cytoplasmic and nuclear staining with calretinin. Two of three mesonephric adenocarcinomas showed luminal positivity with CD10 and nuclear and cytoplasmic positivity with calretinin. Seven of nine endometrial adenocarcinomas were positive with CD10 (four cytoplasmic, two membranous and cytoplasmic, one luminal and cytoplasmic) and three with calretinin (two cytoplasmic, one nuclear and cytoplasmic). Positive staining of endometriotic stroma with CD10 was present in all endometriosis cases but normal cervical stroma was also strongly positive, especially around glands. Endometriotic stroma and cervical stroma were negative with calretinin.
Conclusions: We conclude that most endocervical glandular lesions, including mesonephric remnants/hyperplasia, are negative with calretinin. However, the focal nuclear and cytoplasmic positivity with calretinin in two of three mesonephric adenocarcinomas suggests that this may be a useful indicator of a mesonephric origin of a cervical adenocarcinoma. Most mesonephric remnants/hyperplasias exhibit luminal positivity with CD10, although this is not invariable and staining is usually focal. Positive luminal staining of a benign endocervical glandular lesion with CD10 may help confirm mesonephric remnants. Although positive staining with CD10 was found in two of three mesonephric adenocarcinomas, the observed immunoreactivity of several conventional cervical adenocarcinomas limits the diagnostic value of CD10 in confirming a mesonephric origin for an adenocarcinoma. Since all cervical MDAs were negative with CD10, positivity with this antibody may be of value in distinguishing mesonephric hyperplasia from MDA, although this distinction rarely necessitates immunohistochemistry. Most endometrial adenocarcinomas of endometrioid type stain with CD10 and thus positivity with this antibody is not specific for a mesonephric origin of an endometrial adenocarcinoma. Positivity of normal cervical stroma limits the value of CD10 staining in confirming a diagnosis of cervical endometriosis.