Dedifferentiated liposarcoma (DDLPS) is a term that was first introduced by Evans in 1979 to describe liposarcomas containing a mixture of atypical lipoma (ALT)/WDLPS and a high-grade nonlipogenic sarcomatous component, usually with an abrupt transition between the two components[2
]. Dedifferentiated areas exhibit a wide morphological spectrum and often show a high-grade sarcoma resembling a malignant fibrous histiocytoma (MFH) or low-grade spindle cell sarcoma[3
]. DDLPS has a less aggressive clinical course than other types of high-grade sarcoma, although the underlying mechanism is unclear. Approximately 40% of DDLPSs will recur locally, 17% will metastasize and 28% of patients will ultimately die as a result of the tumor[1
DDLPS usually occurs in the extremities and retroperitoneum. The GI tract is not the typical location for primary liposarcoma and liposarcomas primarily involving GI tract are extremely uncommon. Documented cases of primary liposarcoma have been reported in different parts of the GI tract, such as stomach, small intestine and colon[5
]. Secondary involvement of the GI tract in cases of retroperitoneal liposarcoma is also extremely uncommon. Five cases of DDLPS have been described from the small bowel mesentery[4
]. The occurrence of these tumors in the small intestine often present as a submucosal polyp[6
] or even intussusceptions[7
]. In our case, a well-differentiated component was located in the small bowel mesentery and a dedifferentiated component transmurally invaded the bowel wall and presented as a submucosal polyp. Most of the DDLPSs consist of well-differentiated areas of liposarcoma admixed with areas of morphologically high-grade nonlipogenic sarcoma. However, in our case, the interface between the two components was sharply demarcated grossly. Because dedifferentiation occurring in ALT/WDLPS is considered as time dependent and WDLPS is absent in the small bowel mass, the tumor appeared to originate from the small bowel mesentery as secondary involvement of the small bowel wall.
For our case, taking the radiological information and the location of the lesion into account, the possibility of GIST and sarcoma infiltrating fat was considered. GISTs typically stain with CD117 and do not consist of a WDLPS component. DDLPSs need to be distinguished from other high-grade sarcomas such as MFH because these high-grade sarcomas have a much worse prognosis. The distinction between sarcoma infiltrating fat and DDLPS is the existence of ALT/WDLPS. Immunohistochemically, DDLPS is usually negative for CD117 and CD34 in the dedifferentiated component and positive for S-100 protein in the well-differentiated component[8
The distinction of DDLPS from other high-grade sarcomas can be problematic on needle biopsy as the samples are small and the component of DDLPS can be identical to other high-grade sarcoma. Consequently, liposarcomas should be considered in the differential diagnosis when a submucosal mass is associated with a radiographical mass lesion, especially when the mass is retroperitoneal.