Genetics of High-Grade Endometrioid Adenocarcinoma of the Ovary With Yolk Sac Differentiation: A Case Report.

A 49-year-old multiparous woman (G3P3) with a history of bilateral tubal ligation and multiple caesarean sections presented with a progressively enlarging pelvic mass and worsening pelvic pain over eight months. The patient’s mother had a history of an unknown gynaecological cancer. Abdominal examination demonstrated significant bilateral lower abdominal tenderness with pain and fullness palpated up to the supraumbilical region. Pelvic MRI without contrast revealed a 15 cm complex cystic left ovarian mass extending from the pelvis to the mid abdomen, with an inferior 5.5 cm solid component. The right ovary showed several cystic follicles and no evidence of malignancy. No lymphadenopathy was identified. Laboratory analysis revealed elevated serum levels of alpha-fetoprotein (AFP), cancer antigen 125 (CA-125), cancer antigen 19-9 (CA 19-9), and carcinoembryonic antigen (CEA) (Table 1 ). The patient subsequently underwent a hysterectomy with bilateral salpingo-oophorectomy, resection of regional lymph nodes, and staging (including pelvic washings and omental biopsies). Grossly, the left ovary was replaced by a multi-cystic mass with tan-brown solid areas and regions of hemorrhage and necrosis. The ovarian capsule was involved by a neoplasm. The omentum demonstrated multiple foci of tumor involvement. Microscopic examination revealed multiple morphological patterns that included variable glandular patterns, solid pattern, some squamoid-like foci with abundant cytoplasm, Schiller-Duval bodies with clear cell appearance, as well as primitive appearing cells. The tumor cells showed moderate to severe nuclear atypia, multiple mitosis, and extensive areas of necrosis and hemorrhage. On the periphery of the ovary, there was an area of endometrioid adenofibroma with squamous metaplasia (Figure  1 ). A. Solid and papillary pattern with squamoid, columnar, and clear cells (H&E, x40). B. Glandular and trabecular pattern with columnar cells (H&E, x100). C. Endodermal pattern with Schiller-Duval body (H&E, x200). D. Omental implant with multiple patterns (H&E, x100), necrotic changes and hemorrhage are seen (A-D). Immunohistochemical studies were performed on the formalin-fixed paraffin-embedded sections. All tumor cells were of epithelial origin and were reactive with AE1-AE3 and focally positive with PAX-8, predominantly in the cystadenoma. Tumor cells exhibited a yolk sac tumor profile that included strong reactivity with SALL-4 and focal/weak reactivity with AFP and glypican-3 (Figure 2 ). A. Strong SALL-4 reactivity of neoplastic cells (x100). B. Focal reactivity with AFP seen in the cells with clear cytoplasm (x200). C. Tumor cells demonstrate strong reactivity with glypican-3 (x100). D. Cells of adenofibroma demonstrate nuclear ER-reactivity (x40). The tumor cells were negative for vimentin, p16, ER, and PR receptors. A few cells were positive for GATA3. Immunohistochemical panels for clear cell carcinoma, serous carcinoma, embryonal carcinoma, and choriocarcinoma were negative. Mismatch repair protein testing by immunohistochemistry (MLH1, PMS2, MSH1, and MSH2) showed intact expression, ruling out Lynch syndrome. Microscopic examination of four regional lymph nodes was negative for neoplasia; multiple peritoneal implants, mesenteric, and omental implants showed metastases of tumor with yolk-sac differentiation. Solid next-generation exome sequencing (Gatorseq NGS Panel; 718 genes; University of Florida) utilizing Illumina NovaSeq showed few pathogenic mutations in CTTNB, ARID1A,  and PIK3CA (Table 2 ). The patient was clinically staged as stage IIIC ovarian high-grade endometrioid carcinoma with yolk-sac differentiation. The patient received three cycles of bleomycin, etoposide, and cisplatin (BEP), a germ cell tumor protocol with a favorable response. She tolerated the therapy, apart from developing anemia and nausea. AFP levels went down to 132 ng/ml, and the tumor burden was decreased.

Ovarian pure yolk sac tumors (YST) are the second most common germ cell pediatric neoplasms with well-known sensitivity to chemotherapy [ 1 , 2 ]. Ovarian adenocarcinoma with YST differentiation is an extremely rare entity [ 3 ]. Among women who are perimenopausal/postmenopausal, YSTs typically present in association with other epithelial tumors (endometrioid adenocarcinoma, serous carcinoma, clear cell carcinoma, or carcinosarcoma) and are characterized by variable and diverse histological patterns [ 4 ]. Here, we present a case of endometroid adenocarcinoma with yolk sac differentiation that occurred in a perimenopausal woman with unusual morphological, immunohistochemical, and genetic profiles. Pure YSTs can be challenging to diagnose due to the morphological variability owing to their primitive endodermal nature [ 5 ]. Ovarian carcinomas with yolk sac differentiation are rare, and understanding their pathogenesis is essential for optimization of clinical management [ 6 , 7 ].

YST typically arises in premenopausal women and is rare in postmenopausal women [ 3 ]. While most YSTs originate from germ cells, rare cases can arise from somatic tumor cells, particularly epithelial carcinomas [ 7 ]. In the current case, a perimenopausal woman presented with endometrioid adenocarcinoma of the ovary with YST differentiation. These tumors often present diagnostic challenges due to the overlapping morphologic features with high-grade serous carcinomas and other germ cell tumors [ 6 ]. A case series evaluating ovarian tumors with YST differentiation found that only one out of seventeen tumors was an endometrioid carcinoma with YST differentiation, underscoring the rarity of this tumor [ 2 ]. Immunohistochemistry helps in distinguishing such tumors, especially those that are mixed. The patient’s tumor lacked WT1, p16, and p53 overexpression, which helped exclude high-grade serous carcinoma [ 8 ]. A diagnosis of high-grade endometrioid adenocarcinoma with YST differentiation was rendered based on the morphologic features and immunopositivity for SALL-4, glypican-3, and AFP (Figure 2 ). These three immunohistochemical markers are highly sensitive for YST; however, not entirely specific [ 8 ]. SALL-4 is a germ cell marker that is typically positive in dysgerminoma and is usually used to differentiate germ cell tumors from epithelial neoplasms [ 8 ]. Tumor cell positivity for glypican-3 and AFP supported yolk sac differentiation [ 9 ]. Glypican-3 is considered highly specific for YST [ 6 ]. The tumor morphology and negative α-inhibin helped exclude sex cord stromal tumor [ 6 ]. GATA3 is a useful marker in highlighting the reticular-microcystic patterns of primary YST but not the glandular, hepatoid, or solid patterns [ 5 ]. In our case, GATA3 staining was focal and is consistent with prior observations [ 5 ]. CK7 and EMA are traditionally negative in pure YSTs [ 10 ]. The staining of the tumor with CK7 and EMA, in addition to the YST markers highlighted previously, thus further supports the diagnosis of endometrioid adenocarcinoma with YST differentiation [ 8 ]. The explanation of the variability of morphology and immunohistochemical stains could be that this tumor either originated from a common early precursor or through divergent/aberrant differentiation [ 8 , 11 ]. Aberrant differentiation is a phenomenon in which a new cell phenotype arises through a process of transformation of an original somatic precursor neoplasm [ 2 , 7 , 10 ]. While many ovarian divergent differentiation cases demonstrate distinct germ cell and epithelial carcinoma areas [ 2 , 10 ], there is a possibility that the germ cell component transitioned from and obscured the epithelial component, as evident in the current case [ 2 ]. Of particular interest in this case is the tumor’s genetic profile: the identification of pathogenic/likely pathogenic mutations in CTNNB1 (β-catenin), ARID1A , and PIK3CA strongly supports a diagnosis of endometrioid adenocarcinoma with somatic YST differentiation rather than a primary germ cell yolk sac tumor because CTNNB1 and PIK3CA mutations are endometrioid carcinoma-specific mutations [ 11 , 12 ]. These mutations are frequently observed in endometrioid ovarian carcinomas and are associated with tumors arising from endometriosis [ 13 ]. In contrast, pure ovarian YST of germ cell origin typically lack these mutations with some cases may show KRAS, ARID1A, KIT mutations; most cases often harbor distinct germ cell-related alterations, such as chromosomal 12p abnormalities [ 13 , 14 ]. Recent molecular analyses have demonstrated that in mixed tumors, the YST and carcinoma components can harbor shared somatic mutations, including CTNNB1 and ARID1A , implying a common clonal origin and supporting the theory of somatic derivation [ 11 , 15 ]. In the current case, the concurrent presence of CTNNB1, ARID1A, and PIK3CA mutations, which are well-described in endometrioid carcinomas, as well as the absence of germ cell tumor-specific genetic alterations, provides strong molecular evidence that the YST phenotype observed was somatically derived. This genetic profile reinforces the classification of the tumor in this case as an endometrioid adenocarcinoma with YST differentiation and emphasizes the value of molecular profiling in distinguishing between primary germ cell tumors and somatic carcinomas with germ cell-like features [ 13 , 15 ].

Ovarian tumors that undergo germ cell differentiation may be less responsive to chemotherapy directed solely at germ cell tumors. In these cases, platinum-based regimens designed to treat both epithelial ovarian neoplasms and germ cell tumors may offer better clinical outcomes. In conclusion, this case highlights the complex diagnostic challenges that pathologists may face when diagnosing ovarian tumors. YSTs are generally hard to diagnose due to the morphological and immunohistochemical overlap with other tumors. Another challenge in diagnosis is to exclude mixed tumors, which typically requires careful examination and adequate sampling. Genetic profiling can confirm somatic derivation and guide appropriate multimodal therapy.