CD10 is a sensitive and diagnostically useful immunohistochemical marker of normal endometrial stroma and of endometrial stromal neoplasms.

AIMS

The CD10 antigen is expressed in acute lymphoblastic leukaemia and follicle centre cell lymphoma. A recent study investigating the expression of CD10 in a wide range of non-haematopoietic neoplasms found positive staining in a small number of endometrial stromal sarcomas as well as in normal endometrial stroma. The present study aimed to ascertain whether CD10 positivity is indeed found in normal endometrial stroma and endometrial stromal neoplasms. Staining of a range of tumours which can be confused morphologically with endometrial stromal neoplasms was also undertaken to ascertain whether antibodies against CD10 are of value in a diagnostic sense.

METHODS AND RESULTS

Neoplasms included in the study were endometrial stromal nodule (n=1), low-grade endometrial stromal sarcoma (ESS) (n=13), high-grade ESS (n=6), mixed endometrial stromal-smooth muscle tumour (n=1), uterine cellular leiomyoma (n=10), uterine leiomyosarcoma (n=5), adult granulosa cell tumour (AGCT) (n=10), undifferentiated endometrial carcinoma (n=6), uterine carcinosarcoma with an endometrial stromal component (n=1) and type II uterine mesenchymal tumour with sex cord-like elements (n=1). Cases of proliferative (n=5), secretory (n=5) and atrophic (n=3) endometrium were also stained. There was positive staining of stroma but not of glands in all cases of non-tumorous endometrium. There was positive staining of the endometrial stromal nodule and of all low-grade ESS. Staining in these varied but was often diffuse and of moderate to strong intensity. There was positive staining of four of six high-grade ESS, but this was usually focal. There was also positive staining of the endometrial stromal component in the mixed endometrial stromal-smooth muscle tumour and in the uterine carcinosarcoma. Most cellular leiomyomas were completely negative although three exhibited weak positivity. There was some positivity, usually focal or weak, of three of five leiomyosarcomas. Most AGCT and undifferentiated carcinomas were completely negative although one case of each exhibited focal staining. There was focal staining of the type II uterine mesenchymal tumour with sex cord-like elements.

CONCLUSION

CD10 is a reliable and sensitive immunohistochemical marker of normal endometrial stroma. Positivity, which is often strong and/or diffuse is found in endometrial stromal nodules and low-grade ESS. Positive staining with CD10, when strong and diffuse, may be useful in distinguishing these tumours from histological mimics, especially cellular leiomyoma and AGCT which are generally negative. In this situation, CD10 should be used as part of a panel which might include desmin and alpha-inhibin depending on the differential diagnosis considered. Positive staining with CD10 in a high-grade uterine sarcoma which is negative with muscle markers might indicate endometrial stromal differentiation and identify a group of neoplasms which it is correct to diagnose as high-grade ESS rather than undifferentiated uterine sarcoma.