Macroscopically, tan-brown solid mass on cut-surface revealed large hemorrhagic and necrotic areas. Histologically, the mass revealed a malignant stroma and benign müllerian-type (endometrial, mucinous and squamous) glands. Glandular part was minimum and scattered in the stroma (Figure 1A). Stromal component was overtly sarcomatous. Mitotic activity showed differences in different areas and heterologous components (1-6 per 10 high-power field). Ki-67 proliferating index also showed differences (Figure 1B). Lipomatous, leiomyomatous and cartilaginous areas -heterologous elements- were found (Figure 1C). There were foci of endometriosis in uterine serosa and ovary; adenomyosis in myometrium was present. All submitted lymph nodes were reactive.

Fig 1: (A,B,C). High-grade adenosarcoma (case 1). (A): A fascicular proliferation of stromal spindle cells and benign endometrial-type glands (H&E;, x200), (B): sarcomatous areas with higher Ki 67 proliferation index (immunohistochemically, x200), (C): Chondrosarcomatous component (H&E;, x200)

Case 2
A 37 year-old woman (gravida 1, para 1) was admitted to the hospital with abdominal pain. She did not have any gynecologic history. A pelvic mass was diagnosed on pelvic examination and confirmed on USG as vascular solid mass with cystic areas.

Macroscopically, the solid mass was well-circumscribed with a maximum diameter of 13 cm. On cut-surface, the neoplasm was fleshy solid containing numerous cystic structures (Figure 2A). Histologically, it had epithelial and stromal components. Glands were lined by benign-appearing endometrioid cells that were columnar and pseudostratified. Most of the glands were cystic but did not have “leaf-like” configuration. Some of the glands revealed focally simple or complex hyperplastic features. The stroma in large areas appeared loose cellular and looked like endometrial stroma. Based on these features, the tumor looked like an adenomyoma (Figures 2B&C;). On frozen section, it had been reported as a benign condition but exact diagnosis could not been given. We made several samplings from different areas. Stromal cells were bland and had low mitotic activity (1 per 10 high-power field). Focal smooth muscle bundles were also present. Stromal cellularity showed condensation around the glands. In these areas the stromal cells didn’t showed atypia or high mitotic rate. In two slides on microscopic foci there were overt sarcomatous component (but <25% of all areas) (Figure 2D). The stromal cells were focally but strongly positive for estrogen receptor and CD10. There were also adenosarcomatous foci in left tuba uterina and the ovary. All submitted lymph nodes were reactive.

Fig 2: (A,B,C,D). Low-grade adenosarcoma (case 2). (A): cut surface of primary pelvic solid mass with cysts, (B&C;): hyperplastic endometrial-type glands with periglandular stromal condensation (H&E;, x40), (D): sarcomatous area with nuclear pleomorphism and mitotic figure (H&E;, x200)

Our patients are peri-menopausal (case 1) and at reproductive age (case 2). Laboratory data were normal. Summary of clinical and pathological characteristics is seen in table 1.

Table 1: Summary of reported cases

In these cases, the glands were lined by endometrial type epithelium and in case 1 glands were lined also by mucinous and squamous epithelium. Carcinomatous component was not observed. In Case 1, heterologous elements were found in high grade sarcomatous areas. We know that Müllerian adenosarcomas differ from malignant mixed müllerian tumors by their less malignant behavior[2,3]. So these tumors should be adequately sampled to investigate whether there is a carcinomatous component.

Adenosarcomas may have large areas of low-grade sarcomatous component. Glands are dilated and cystic, and may be hyperplastic with minor atypia. Most adenosarcomas have focally hypercellular stroma, especially in periglandular areas. Periglandular stromal cuffing is a critical diagnostic feature of adenosarcoma[1,2]. Stromal cells have bland nuclei so they can be misdiagnosed as endometriosis or as adenofibroma like our case 2. But we were lucky, there were small foci of overt sarcoma that contains high nuclear pleomorphism and atypical mitosis. Stromal cellularity, high mitotic rate and nuclear atypia, if present, are useful criteria in distinguishing from benign conditions. The stromal mitotic counts that have been recommended by Clement and Scully, is 2 or more per 10 high power field[2]. Low grade adenosarcomas arising in uterine cervix especially in young patients have favorable prognosis so they should be distinguished from rhabdomyosarcoma. Rhabdomyosarcomas may have loose connective stroma and periglandular rhabdomyoblastic accumulation resembling a low grade adenosarcoma.

Stromal component may contain heterologous elements especially rhabdomyoblasts. When the stroma is sarcomatous, at least 25% of total area, described as “sarcomatous overgrowth” and cases with this feature make up a minority[2]. In this group due to the presence of sarcomatous overgrowth, the prognosis is adversely affected.

Immunohistochemical findings of adenosarcomas have been reported previously in a few articles, so we found that estrogen receptors can be positive in stromal cells 8. This finding may be helpful in the treatment of these tumors[3,4,6].

In Case 1, it is difficult to tell that tumor has arisen in foci of endometriosis. We could not show any of Scully’s criteria in lots of samples made[5]. According to the literature, we must keep on our mind that persistent and recurrent endometriosis, especially extrauterine forms, may transform to adenosarcoma or be underdiagnosed as “endometriosis”[5,6]. In case 1, we couldn’t review prior pathology slides for endometriosis.

In conclusion, these cases taught that, diagnosis of adenosarcomas on frozen section has difficulties, underdiagnosis can be a pitfall. To achieve the correct diagnosis of these tumors we must make adequate sampling. Numerous samplings can reveal high-grade sarcomatous areas or carcinomatous focus. So we can explain the difference between pathological appearance and clinical behavior.

1) Clement PB, Scully RE. Extrauterine mesodermal (Müllerian) adenosarcoma. A clinicopathologic analysis of five cases. Am J Clin Pathol 1978;69:276-83.

2) Eichhorn JH, Young RH, Clement PB, et al. Mesodermal (Müllerian) adenosarcoma of the ovary. A clinicopathologic analysis of 40 cases and a review of the literature. Am J Surg Pathol 2002;26:1243-57.

3) Fukunaga M, Nomura K, Endo Y, et al. Ovarian adenosarcoma. Histopathology 1997;30:283-7.

4) Visvalingam S, Jaworski R, Blumenthal N, et al. Primary peritoneal mesodermal adenosarcoma: Report of a case and review of the literature. Gynecol Oncol 2001;81:500-5.

5) Judson PL, Temple AM, Fowler WC, et al. Vaginal adenosarcoma arising from endometriosis. Gynecol Oncol 2000;76:123-5.

6) Liu L, Davidson S, Singh M. Müllerian adenosarcoma of vagina arising in persistent endometriosis: report of a case and review of the literature. Gynecol Oncol 2003;90:486-90.