CASE REPORTS
Low-Grade Endometrial Stromal Sarcoma Presenting as Vaginal Nodule Marcus V. N. Corpa, MD, Eduardo P. Serafini, MD, and Carlos E. Bacchi, MD Endometrial stromal sarcoma is a rare neoplasm of the uterus. Extrauterine locations of this neoplasm, excluding metastases or local extension, are even more unusual and are usually associated with the presence of endometriosis. The authors report a case of endometrial stromal sarcoma presenting as a vaginal wall nodule, without any sign of primary uterine tumor after extensive evaluation or presence of endometriosis. The morphology, immunohistochemical profile, differential diagnoses, and pathogenesis are discussed, as well as a review of the literature on this issue. Ann Diagn Pathol 8: 295-298, 2004. © 2004 Elsevier Inc. All rights reserved. Index Words: Endometrial stromal sarcoma, vagina, endometriosis, CD10
E
NDOMETRIAL stromal sarcoma (ESS) is a rare mesenchymal neoplasm of the uterus1-4 which belongs to a unique group of neoplasms that are composed of a spectrum from benign to highly malignant tumors. Occasionally, ESS arises primarily in extrauterine sites, such as pelvic cavity,5-7 ovary,5,6 abdominal cavity,5 fallopian tube,5 retroperitoneum,5 vagina,5,8,9 and vulva.10 In those cases, accompanying foci of endometriosis are frequently, but not invariably, present. The authors present a case of primary vaginal ESS without detectable association of endometriosis with review of literature. Clinical History A 40-year-old woman presented with periurethral vaginal nodule measuring 2 cm in diameter. Her past medical history was unremarkable. Pelvic ultrasonographic examination and uterine curettage were performed, revealing no abnormalities. The nodule was completely excised. Hysterectomy was not performed and no further treatment was given. There was no tumor recurrence, and no sign of uterine malignancy was detected during follow-up.
From Department of Pathology, State University of Sa˜o Paulo, Botucatu Campus, State of Sa˜o Paulo, Brazil. Address reprint requests to Carlos E. Bacchi, MD, Departamento de Patologia, Faculdade de Medicina de Botucatu-UNESP, Botucatu, SP 18618-000, Brazil. E-mail:
[email protected] © 2004 Elsevier Inc. All rights reserved. 1092-9134/04/0805-0007$30.00/0 doi:10.1016/j.anndiagpath.2004.07.007
The patient is alive and well, free of tumor 38 months after the diagnosis. Material and Methods The tissue samples of the tumor were fixed in 10% buffered formalin and embedded in paraffin. Four-micrometer-thick sections were cut from the paraffin blocks and stained with hematoxylin-eosin. Immunohistochemical study was performed on formalin-fixed tissue using avidin-biotin complex method with heat-induced antigen retrieval as previously described.11 Immunostains for CD10 (1:100; clone NCL-CD 1-270; Novocastra, New Castle upon Tyne, UK), estrogen receptor (1:2000; clone ID5; Dakocytomation, Carpinteria, CA), progesterone receptor (1:200; clone TGR636; Dakocytomation), desmin (1:250; clone D33; Dakocytomation), muscle actin (1:1000; HHF 35; Dakocytomation), cytokeratin 8 (1:50; clone CAM5.2; Becton & Dickinson, San Jose, CA), and S-100 protein (1:4000; polyclonal; Dakocytomation) were prepared.
Pathologic Findings The surgical specimen consisted of numerous irregular rubbery fragments of tissue, ranging from 0.5 to 2 cm in diameter. Cut surface was solid and tan. Light microscopy disclosed an infiltrative tumor composed of uniform spindle cells with illdefined eosinophylic cytoplasm. The nuclei were oval-shaped with finely granular, evenly distributed chromatin and inconspicuous nucleoli. In some areas, there were small, thin-walled vessels with focal whorling of neoplastic cells around them (Fig 1A and B). Cytologic atypia was unremarkable,
Annals of Diagnostic Pathology, Vol 8, No 5 (October), 2004: pp 295-298
295
296
Copra, Serafini, and Bacchi
Figure 1. (A) The neoplasm is composed of spindle cells disposed in an infiltrative fashion, sparing the superficial vaginal epithelium. (B) The spindle cells contain eosinophylic cytoplasm and mild nuclear atypia, with focal whorling around small arterioles. (C) Diffuse strong positivity of the neoplastic cells to CD10. (D) Diffuse strong positivity of the neoplastic cells to estrogen receptor.
necrosis was absent, and mitotic index was less then 2 per 10 high-power fields. The tumor cells showed diffuse and strong membranous positivity for CD10 (Fig 1C) and diffuse intense nuclear staining for estrogen (Fig 1D) and progesterone receptors. Desmin, muscle actin, S-100 protein, and cytokeratin stains were all negative. Discussion Endometrial stromal tumors are rare uterine neoplasms. Microscopically, these tumors display a proliferation of cells closely resembling those of normal proliferative endometrial stroma. Foci of epithelial, smooth muscle, and sex cord-like differentiation can be present in a subset of cases.14 In their seminal report, Norris and Taylor4 attempted to classify these tumors, creating three categories.
Endometrial stromal nodule was defined as a circumscribed, solitary lesion with pushing borders. Low-grade ESS displayed infiltrative growth pattern, and less than 10 mitoses/10 high-power fields. High-grade ESS was distinguished from the lowgrade variant for possessing 10 or more mitoses/10 high-power fields. However, recent studies have provided strong evidence that mitotic index did not correlate with prognosis of ESS.15,16 This concept was endorsed in a recent review by Oliva et al.14 According to the recent evidence, it is more appropriate to consider ESS as a single entity: “low-grade ESS,” retaining the modifier “low-grade” to emphasize the biological nature of the neoplasm.14 The designation “poorly differentiated endometrial sarcoma” should be reserved for sarcomas originated from the endometrium but lacking endometrial stromal
Vaginal Endometrial Stromal Sarcoma
differentiation. Foci of smooth muscle and endometrioid glandular differentiation can be present in endometrial stromal neoplasms with no prognostic implication. Awareness of this characteristic, however, is an important consideration in the differential diagnosis as discussed below. Cases of primary ESS in extrauterine locations have been widely reported. The sites of occurrence include pelvic cavity,5-7 ovary,5,6 abdominal cavity,5 fallopian tube,5 retroperitoneum,5 vulva,10 and vagina.5,8,9 In the majority of the cases, foci of endometriosis are found in the vicinity of the neoplasm, which would be a probable source of the tumors in these cases. However, in several reports,5,6,17 including our case, no foci of endometriosis were found in association with extrauterine ESS. The secondary mu¨ llerian system, postulated by Lauchlan,18,19 could explain the occurrence of these tumors in extrauterine sites such as ovaries, fallopian tube, and pelvic peritoneum, but cannot be applied to cases occurring in sites such as vulva and vagina. According to this hypothesis, the mesenchymal cells present in tissues derived from the celomic epithelium would have the potential to differentiate into mu¨llerian-type epithelium and stroma. There is, to the best of our knowledge, only one previous report of primary ESS of the vagina. In the case reported by Berkowitz et al,9 a 57-year-old white woman presented with complaint of intermittent vaginal bleeding. At physical examination, a vaginal apex nodule was detected, which was shown to be a low-grade ESS, with subsequent vaginectomy. Microscopically, the tumor displayed typical features of ESS, with concurrent endometriosis. The patient had had a total abdominal hysterectomy and left salpingo-ooforectomy for ovarian endometriosis 15 years before, and the pathologic examination of the surgical specimen at that time did not reveal neoplasia. Differently from the case reported by Berkowitz et al,9 in our case no foci of endometriosis were found in association with the neoplasm. Although hysterectomy was not performed, the long follow-up with negative uterine curettage and serial ultrasound examinations allowed us to exclude the possibility of a vaginal metastasis from a uterine ESS. The mechanism by which this tumor occurred cannot be determined with accuracy. The Lauchlan’s hypothesis, as stated above, can only be applied to those cases occurring in anatomic sites
297
lined by coelomic-type of epithelium. In our case, we cannot exclude the possibility that this tumor arose in focus of endometriosis, which could have been obscured by tumor overgrowth. Immunohistochemically, the cells in ESS are positive for vimentin and estrogen receptor, with focal reactivity for desmin and muscle actin in varying proportions.14 Recent studies have described the unique expression of CD10 in endometrial stromal cells as well as in ESS, which was proven diagnostically useful in the distinction of ESS from smooth muscle neoplasms like low-grade leiomyosarcoma.20,21 The main histopathologic differential diagnoses of ESS are smooth muscle tumors, hemangiopericytoma, adenosarcoma, and carcinosarcoma. Smooth muscle neoplasms contain cells with eosinophilic vacuolated cytoplasm, arranged in long sweeping fascicles, lacking the whorling pattern present in ESS. In addition, as stated previously, immunohistochemistry can be decisive in the distinction between these tumors being ESS. Smooth muscle tumors are strongly positive for muscle markers such as desmin and muscle actin and consistently negative for CD10. On the other hand, ESS is either negative or only focally reactive for muscle markers, even in those cases with identifiable muscular differentiation, and typically reactive for CD10. Hemangiopericytoma can be differentiated from ESS by its unique branching and “staghorn” vascular pattern, which contrasts with the arteriolar pattern present in ESS, as well by its immunohistochemical profile. Adenosarcoma must be excluded when epithelial-like differentiation is present in ESS. The epithelial component in ESS usually resembles that of endometrial glands, with benign appearance and irregular distribution. Adenosarcomas demonstrate a regular distribution of the glandular component within the stroma of the tumor and lack the local invasiveness seen in ESS. Carcinosarcoma can be readily distinguished by the presence of unequivocal cytologic atypia in the epithelial component. In summary, we present a case of low-grade endometrial stromal sarcoma without unequivocal association of endometriosis presenting as vaginal tumor. This report stresses the importance of the inclusion of ESS in the repertoire of differential diagnoses when dealing with mesenchymal neoplasms of abdominal and pelvic cavities, vulva, and vagina.
298
Copra, Serafini, and Bacchi
References 1. Chang KL, Crabtree GS, Lim-Tam SK, et al: Primary uterine stromal neoplasms. A clinicopathologic study of 117 cases. Am J Surg Pathol 1990;14:415-438 2. Fekete PS, Vellios F: The clinical and pathological spectrum of endometrial stromal neoplasms. A report of 41 cases. Int J Gynecol Pathol 1984;3:198-212 3. Kahanpaa KV, Mahlstrom T, Grohn P, et al: Sarcomas of the uterus: A clinicopathologic study of 119 patients. Obstet Gynecol 1977;49:393-403 4. Norris HJ, Taylor HB: Mesenchymal tumors of the uterus. A clinical and pathologic study of 53 stromal tumors. Cancer 1966;19:755-766 5. Chang KL, Crabtree GS, Lim-Tam SK, et al: Primary extrauterine endometrial stromal neoplasms: A clinicopathologic study of 20 cases and review of the literature. Int J Gynecol Pathol 1993;12:282-296 6. Fukunaga M, Ishihara A, Ushigome S: Extrauterine endometrial stromal sarcoma: Report of three cases. Pathol Int 1998;48:297-302 7. McClugagge WG, Bailie C, Weir P, et al: Endometrial stromal sarcoma arising in pelvic endometriosis in a patient receiving unoposed oestrogen therapy. Br J Obstet Gynecol 1996;103:1252-1254 8. Kondhi-Paphtis A, Smirniotis B, Liapis A, et al: Stromal sarcoma arising on endometriosis. A clinicopathological and immunohistochemical study of 4 cases. Eur J Gynaecol Oncol 1998;19:588-590 9. Berkowitz RS, Ehrmann RL, Knapp RC: Endometrial stromal sarcoma arising from vaginal endometriosis. Obstet Gynecol 1978;51:34-37 10. Irvin W, Pelkey T, Rice L, et al: Endometrial stromal sarcoma of the vulva arising in extraovarian endometriosis: A
case report and literature review. Gynecol Oncol 1998;71:313316 11. Gown AM, de Wever N, Battifora H: Microwave-based antigenic unmasking: A revolutionary new technique for routine immunohistochemistry. Appl Immunohistochem 1993;1:256-266 12. Aaro LA, Synmonds RE, Dockerty MB: Sarcoma of the uterus. A clinical and pathologic study of 177 cases. Am J Obstet Gynecol 1966;94:101-109 13. Koss LG, Spiro RH, Brunschwig A: Endometrial stromal sarcoma. Surg Gynecol Obstet 1965;121:531-537 14. Oliva E, Clement PB, Young RH: Endometrial stromal tumors: An update on a group of tumors with a protean phenotype. Adv Anat Pathol 2000;7:257-281 15. Evans HL: Endometrial stromal sarcoma and poorly differentiated endometrial sarcoma. Cancer 1982;50:2170-2182 16. De Fusco PA, Gaffey TA, Malkasian GD Jr, et al: Endometrial stromal sarcoma: Review of Mayo Clinic experience, 19451980. Gynecol Oncol 1989;35:8-14 17. Young RH, Prat J, Scully RE: Endometrioid stromal sarcomas of the ovary: A clinicopathologic analysis of 23 cases. Cancer 1984;53:1143-1155 18. Lauchlan SC: The secondary mullerian system. Obstet Gynecol Survey 1972;27:133-146 19. Lauchlan SC: The secondary mullerian system revisited. Int J Gynecol Pathol 1994;13:73-79 20. Chu P, Arber DA: Paraffin-section detection of CD 10 in 505 nonhematopoietic neoplasms: Frequent expression in renal cell carcinoma and endometrial stromal sarcoma. Am J Clin Pathol 2000;113:374-382 21. Chu P, Arber DA, Weiss LM, et al: Utility of CD 10 in distinguishing between endometrial stromal sarcoma and uterine smooth muscle tumors: An immunohistochemical comparison of 34 cases. Mod Pathol 2001;14:465-471
