Uterus

This case first appeared as Performance Improvement Program in Surgical Pathology (PIP) 2021, Case 18, and is a low-grade endometrial stromal sarcoma of the uterus.

The information provided in this case was accurate and correct at the time of publication in 2021.

Any changes in terminology since the time of publication may not be reflected in this case.

Microscopic examination shows tumor infiltrating adjacent myometrium in large islands or “tongues.” At higher power, the tumor consists of small, monotonous cells with round to oval nuclei, smooth nuclear contours, and scant cytoplasm, resembling proliferative-phase endometrial stroma. Mitotic activity is fewer than 5/10 high power fields (HPF). Extensive vascular invasion is seen. The overall findings of this case are consistent with the diagnosis of low-grade endometrial stromal sarcoma (LGESS).

LGESS is often recognized at the time of gross examination of a hysterectomy specimen because the tumor has a characteristic tongue-like penetration into the endometrium and may plug up numerous vessels, imparting a “wormlike” appearance macroscopically. Microscopically, LGESS have cytologic features typical of non-neoplastic proliferative-phase endometrial stroma. Specifically, the neoplastic cells are ovoid to fusiform and bland, with no or minimal cytologic atypia. There is conspicuous concentric growth around spiral arteriole-like vessels. Immunohistochemistry is not necessary for the diagnosis, but these tumors almost universally express CD10, estrogen receptor (ER), and progesterone receptor (PR); depending on variant histological differentiation, they can express immunohistochemical markers of smooth muscle differentiation (usually <10% of cells) or sex-cord differentiation. LGESS are negative for cyclin D1 and BCOR immunohistochemical stains. Molecular confirmation is often sought. Molecularly, LGESS are characterized by translocations resulting in gene fusions, commonly involving the JAZF1 gene (JAZF1::SUZ12; t(7;17)) (seen in approximately 50% of cases), as well as other fusions involving PHF1, MEAF6, EPC1/2, and BRD8.

The cytomorphology and immunohistochemical and molecular profile of LGESS are identical to that of ​​endometrial stromal nodule (ESN), a benign lesion. The distinction between the two is an architectural one: ESN macroscopically presents as a solitary, well-defined, yellow myometrial mass and microscopically should be well-circumscribed, while LGESS shows a tongue-like pattern of invasion and frequent vascular invasion. Even though ESN are well-defined, they are allowed to have minor marginal irregularity (up to three protrusions, each less than 3 mm beyond the main mass). Those with more, but without the widespread myometrial infiltration seen in LGESS, should still be regarded as potentially malignant, and the term “LGESS with limited infiltration” may be used. Limited evidence is available in these rare cases, but they appear to have malignant potential, unlike the benign ESN. Thus, sampling of the entire tumor-to-myometrial interface is imperative in the differential of ESN and LGESS. ESN often demonstrate translocations similar to LGESS, meaning molecular testing cannot reliably distinguish the two.

High-grade endometrial stromal sarcoma (HGESS) is a malignant tumor of endometrial stromal derivation with uniform, high-grade, roundand/or spindled cells. Compared to LGESS, these tumors show brisk mitotic activity of more than 10/10 HPF and tumor cell necrosis; they may exhibit a spectrum from tongue-like to more destructive myometrial invasion by high-grade round cells. There are three molecular subtypes of HGESS – YWHAE::NUTM2A/B fusion, BCOR::ZC3H7B fusion, and BCOR internal tandem duplication. Tumors with YWHAE::NUTM2A/B fusion are composed of round cells with eosinophilic cytoplasm and high-grade nuclei. In addition, in these tumors, a biphasic (low- and high-grade) pattern may be present, and the associated low-grade spindle cell component can have fibromyxoid features. Immunohistochemically, the high-grade component tends to be negative for CD10, ER, and PR, and strongly expresses cyclin-D1 and BCOR (expression not typically seen in LGESS). Importantly, the low-grade component of this subtype of HGESS will have the immunohistochemical profile of LGESS; thus, when performing immunohistochemistry, a block with a high-grade component should be selected for accurate diagnosis. Despite variable morphology, all areas have characteristic YWHAE rearrangements, which are exclusive to HGESS. Tumors with genetic rearrangements involving BCOR often have extensive myxoid stroma and variable spindled morphology. Immunohistochemically, tumors with BCOR::ZC3H7B fusion are positive for cyclin D1 and CD10, and are negative for ER and PR; BCOR immunohistochemical stain is positive in half of cases. Tumors with BCOR internal tandem duplication are positive for cyclinD1 and BCOR stains, and have variable CD10 and ER expression.

Cellular leiomyoma can also be difficult to distinguish from LGESS due to its densely uniform cellularity, which can be significantly more cellular than background myometrium and appear blue on low-power examination. The presence of thick-walled blood vessels, prominent clefted spaces, focal conventional fascicular architecture, and cigar-shaped nuclei are helpful features supportive of smooth muscle differentiation. Mitotic figures are rare in cellular leiomyomas. In most cases, cellular leiomyomas are usually either well-circumscribed or have slightly irregular tumor borders and are more easily confused with ESN than with LGESS. The differential can become more challenging, however, in the case of intravenous leiomyomatosis, which is defined as intravascular growth of cytologically bland smooth muscle outside the confines of a leiomyoma. These tumors can show multiple convoluted wormlike extensions that infiltrate myometrial and/or parametrial vessels. LGESS permeate the myometrium but typically do not have extensive grossly observable vascular invasion. To resolve the differential of a smooth muscle tumor vs LGESS, an immunohistochemical panel including CD10 and multiple smooth muscle markers (smooth muscle actin, desmin, h-caldesmon) can be used. It is critical to consider the morphologic appearance when interpreting immunohistochemistry, given altered smooth muscle differentiation in LGESS, as well as the fact that highly cellular leiomyomas may express CD10 and lack strong smooth muscle marker expression. ER and PR staining can be similar in both cellular leiomyomas and LGESS.

​​​​​​​​​An additional diagnostic consideration before rendering a diagnosis of LGESS is uterine tumor resembling ovarian sex cord stromal tumor (UTROSCT). UTROSCT is a uterine tumor exclusively composed of sex cord stromal-like elements. In LGESS, there should be typical areas of overt endometrial stromal differentiation and the classical pattern of invasion, whereas in UTROSCT, a clear endometrial stromal component is lacking or minimal; thus, the differential can be especially difficult in small biopsies. Immunohistochemistry should be interpreted with caution, as both UTROSCT and LGESS can express both sex cord stromal markers as well as smooth muscle markers. Unlike LGESS, UTROSCT lacks JAZF1 and PHF1 rearrangements.

1. Which of the following aids in differentiation between low-grade endometrial stromal sarcoma (LGESS) and endometrial stromal nodule?
   1. ​JAZF1::SUZ12 translocation by fluorescence in situ hybridization
   2. Mitotic activity
   3. Morphologic appearance of the cells
   4. Tumor expression of CD10, estrogen receptor (ER), and progesterone receptor (PR)
   5. Widespread myometrial and vascular invasion
2. Which of the following features is most helpful in distinguishing a cellular leiomyoma from LGESS?
   
   
   1. Blue appearance on low-power examination
   2. Densely uniform cellularity
   3. Mitotic activity
   4. Tumor expression of CD10 and desmin
   5. Tumor expression of ER and PR
3. Which of the following immunohistochemical findings would one most likely see in LGESS?
   
   
   1. Diffuse staining for smooth muscle actin
   2. Negative staining for ER
   3. Negative staining for PR
   4. Strong staining for CD10
   5. Strong staining for Cyclin-D1

1. David B. Chapel, Marisa R. Nucci, An update in pathologic diagnosis of uterine mesenchymal tumours, Diagnostic Histopathology, Volume 27, Issue 12, 2021, Pages 506-518.
2. Devereaux KA, Schoolmeester JK. Smooth muscle tumors of the female genital tract. Surg Pathol Clin. 2019;12(2):397-455.
3. Hoang L, Chiang S, Lee CH. Endometrial stromal sarcomas and related neoplasms: new developments and diagnostic considerations. Pathology. 2018;50(2):162-177.
4. Lee CH, Nucci MR. Endometrial stromal sarcoma--the new genetic paradigm. Histopathology. 2015;67(1):1-19
5. Oliva E, Chiang S, Lee CH, et al. High grade endometrial stromal sarcoma. In: WHO Classification of Tumours Editorial Board. Female genital tumours [Internet]. Lyon (France): International Agency for Research on Cancer; 2020 [cited 2024/02/09]. (WHO classification of tumours series, 5th ed.; vol. 4). Available from: https://tumourclassification.iarc.who.int/chapters/34.
6. Oliva E, Lee CH Chiang S, et al. Low grade endometrial stromal sarcoma. In: WHO Classification of Tumours Editorial Board. Female genital tumours [Internet]. Lyon (France): International Agency for Research on Cancer; 2020 [cited 2024/02/09]. (WHO classification of tumours series, 5th ed.; vol. 4). Available from: https://tumourclassification.iarc.who.int/chapters/34.
7. Moore M, McCluggage WG. Uterine endometrial stromal tumors with limited infiltration: first report of a case series indicating potential for malignant behavior. Int J Gynecol Pathol. 2020;39(3):221-226.
8. Parra-Herran C, Howitt BE. Uterine Mesenchymal Tumors: Update on Classification, Staging, and Molecular Features Surg Pathol Clin. 2019;12(2):363-396.

1. Widespread myometrial and vascular invasion (e)
2. Tumor expression of CD10 and desmin (d)
3. Strong staining for CD10 (d)