Uterine sarcomas: a review.

OBJECTIVE

Uterine sarcomas are rare tumors that account for 3% of uterine cancers. Their histopathologic classification was revised by the World Health Organization (WHO) in 2003. A new staging system has been recently designed by the International Federation of Gynecology and Obstetrics (FIGO). Currently, there is no consensus on risk factors for adverse outcome. This review summarizes the available clinicopathological data on uterine sarcomas classified by the WHO diagnostic criteria.

METHODS

Medline was searched between 1976 and 2009 for all publications in English where the studied population included women diagnosed of uterine sarcomas.

RESULTS

Since carcinosarcomas (malignant mixed mesodermal tumors or MMMT) are currently classified as metaplastic carcinomas, leiomyosarcomas remain the most common uterine sarcomas. Exclusion of several histologic variants of leiomyoma, as well as "smooth muscle tumors of uncertain malignant potential," frequently misdiagnosed as sarcomas, has made apparent that leiomyosarcomas are associated with poor prognosis even when seemingly confined to the uterus. Endometrial stromal sarcomas are indolent tumors associated with long-term survival. Undifferentiated endometrial sarcomas exhibiting nuclear pleomorphism behave more aggressively than tumors showing nuclear uniformity. Adenosarcomas have a favorable prognosis except for tumors showing myometrial invasion or sarcomatous overgrowth. Adenofibromas may represent well-differentiated adenosarcomas. The prognosis of carcinosarcomas (which are considered here in a post-script fashion) is usually worse than that of grade 3 endometrial carcinomas. Immunohistochemical expression of Ki67, p53, and p16 is significantly higher in leiomyosarcomas and undifferentiated endometrial sarcomas than in endometrial stromal sarcomas.

CONCLUSIONS

Evaluation of H&E stained sections has been equivocal in the prediction of behavior of uterine sarcomas. Immunohistochemical studies of oncoproteins as well as molecular analysis of non-random translocations will undoubtedly lead to an accurate and prognostically relevant classification of these rare tumors.