Endometriosis-associated skeletal muscle regeneration: a hitherto undescribed entity and a potential diagnostic...

Skeletal muscle undergoes regeneration generally after an injury and in some cases it may mimic a malignant process. We observed these aspects in association with abdominal wall endometriosis and as no similar conditions were found in the literature this prompted us to study the main clinicopathologic and immunohistochemical profile of this phenomenon. Thirteen cases of abdominal wall endometriosis were retrieved from the files of our Institute. All original slides were reviewed to reveal the presence of skeletal muscle and 8 cases were enrolled for morphologic and immunohistochemical studies as follows: vimentin, desmin, myoglobin, myogenin, myoD1, CD56, S100, and p21. Histologically, in 4 of the 8 cases in the skeletal muscle adjacent to the endometriotic foci there was a proliferation of round cells with the typical appearance of maturing myoblasts. More peripherally, myotubes and early myocytes were present. This proliferation was florid in 1 case and focal in 3 cases. At immunohistochemical investigation, the less differentiated cells reacted with vimentin, desmin, S100, CD56, myoD1, and myogenin but not with myoglobin or p21. On the contrary, intermediately differentiated cells showed a progressive loss of vimentin, CD56, and myoD1 whereas they were positive for desmin, S100, myogenin, myoglobin, and p21. Terminally differentiated cells reacted only with desmin and myoglobin. This peculiar immunohistochemical profile was consistent with the immunophenotype of maturing myoblasts, confirmed the regenerative nature of the phenomenon and allowed differential diagnosis with other proliferations sharing a similar morphology. The expression of early differentiation markers was greatest in the islands of cells nearest to endometriosis, whereas in the more distant areas the markers of late differentiation prevailed. This gradient of expression suggests that muscle cells are stimulated by growth factors or other signals produced by the cycling endometrioid foci. In conclusion, we report a hitherto undescribed entity that may mimic a malignant process, especially when the reaction is florid or when endometriotic glands and stroma are not clearly evident, as during the examination of small biopsies, frozen sections, or cytology samples. Therefore, although the histologic diagnosis of endometriosis is usually straightforward, pathologists should be aware of the concomitant regenerative effects on skeletal muscle, which may represent a possible diagnostic pitfall.