Molecular Characterization of Ovarian Yolk Sac Tumor in a Postmenopausal Woman: Implications for Pathogenesis

A 59‐year‐old woman visited our hospital because of a mass in her lower abdomen. She had no familial cancer history and experienced menopause at 40 years old. The patient provided written informed consent to be part of the case report. A magnetic resonance imaging scan revealed a 16 × 12 × 10 cm tumor expanding beyond the umbilicus with both solid and cystic components. Ovarian cancer was suspected. A computed tomography scan revealed no metastasis but detected pulmonary thromboembolism (PE). Consequently, anticoagulant therapy was initiated. The first surgery, consisting of a hysterectomy, bilateral salpingo‐oophorectomy, and omentectomy, was performed. Pelvic and para‐aortic lymphadenectomy were not conducted because of the presence of PE. The tumor on the left ovary had ruptured and appeared macroscopically as a solid tumor suggestive of ovarian cancer. Pathological diagnosis confirmed a pure ovarian YST without any epithelial components in the left ovary. We prepared at least one histologic slide per centimeter of tumor size, yielding a total of 18 slides. Immunohistochemistry showed positive AFP, SALL4, and Glypican‐3 and negative PAX8, confirming germ cell tumor differentiation (Figure  1 ). After improvement of PE, additional staging surgery (pelvic and para‐aortic lymphadenectomy) was performed. A peritoneal dissemination lesion was identified in the peritoneum of the left ovarian vein and resected. Postoperative pathology showed no lymph node metastasis but revealed peritoneal metastasis. The patient was diagnosed with FIGO stage IIIB (pT3bN0M0). Histopathologic findings. The tumor showed a microcystic and endodermal sinus pattern with Schiller–Duval body (A, H&E, 20×). No epithelial ovarian components were observed. Immunohistochemistry showed positive staining for AFP (B, 20×), SALL4. (C, 20×), and Glypican‐3 (D, 20×). After staging laparotomy, TC therapy (paclitaxel 175 mg/m 2 and carboplatin area under the curve 6, every 3 weeks) was initiated. Serum AFP levels markedly decreased after surgery but began to rise during TC therapy. After five cycles, CT revealed liver metastasis and peritoneal dissemination. The regimen was changed to BEP therapy (bleomycin 30 000 IU on days 2, 9, and 16, etoposide 100 mg/m 2 on days 1–5, and cisplatin 20 mg/m 2 on days 1–5, every 3 weeks). However, serum AFP levels continued to rise. CT showed that the peritoneal dissemination worsened after three cycles of BEP. The treatment was changed to VeIP therapy (vinblastine 0.11 mg/kg on days 1 and 2, ifosfamide 1.2 g/m 2 on days 1–5, and cisplatin 20 mg/m 2 on days 1–5, every 3 weeks). The treatment was not effective (Figure  2 ). Because the tumor was resistant to these platinum‐based chemotherapies, the patient underwent CGP to assess responsiveness to molecular targeted therapy. FoundationOne CDx (Chugai Pharmaceutical) was used for CGP. This is a hybrid‐capture–based next‐generation sequencing assay covering the coding exons of 324 cancer‐related genes and selected introns of rearrangement‐prone genes. AFP trend following the initial surgery and during treatment. BEP, bleomycin, etoposide, and cisplatin; CGP, comprehensive genomic profiling; TC, paclitaxel and carboplatin; VeIP, vinblastine, ifosfamide, and cisplatin. The CGP results showed somatic mutations in ARID1A c.2632C>T (p.Q878*), ARID1A c.4847del (p.A1616Efs*10), CTNNB1 c.121A>G (p.T41A), NFE2L2 c.241G>A (p.G81S), PIK3CA c.3127A>G (p.M1043V), PIK3CA c.3140A>G (p.H1047R) (hotspot). These pathogenic variants are less commonly observed in ovarian malignant germ cell tumors and frequently associated with clear cell carcinoma ( ARID1A and PIK3CA ) and endometrioid carcinoma ( ARID1A , CTNNB1 , and PIK3CA ). No druggable variants were identified, and the patient died 13 months after the initial treatment.

Mihiro Dejima: conceptualization, writing – original draft, methodology, investigation. Koji Matsumoto: formal analysis, supervision. Mamiko Onuki: formal analysis, supervision. Minoru Nagashima: writing – review and editing, writing – original draft, conceptualization, methodology, investigation. Miki Kushima: data curation, investigation. Akihiko Sekizawa: formal analysis, supervision.

This study was approved by the Institutional Review Board of Showa Medical University.

The authors have nothing to report.

In the current patient, the tumor was a pure YST without epithelial components. Somatic variants in ARID1A , CTNNB1 , NFE2L2 , and PIK3CA were identified by CGP. Whole‐exome sequencing analysis of the YST sample revealed several mutated genes, including mutations in KRAS and KIT and a low frequency of TP53 mutations [ 7 ]. Copy number alterations in driver genes were detected, including deletions in ARID1A and PARK2 and amplifications in ZNF217 , CDKN1B , and KRAS . The genetic mutations of germ cell tumors are different from those observed in this case. In ovarian endometrioid carcinoma, which is often associated with postmenopausal YST, the common genetic mutations include mutations in KRAS , PIK3CA , PTEN , CTNNB1 , ARID1A , and TP53 [ 8 ]. The common genetic mutations of ovarian clear cell carcinoma include mutations in ARID1A , PIK3CA, TERTp, KRAS, TP53, POLE , and ATM [ 9 ]. In endometriosis, somatic mutations of KRAS and PIK3CA are commonly observed [ 10 ]. While the tumor in the current case lacked epithelial components, the somatic mutations were more similar to those found in endometrioid carcinoma or clear cell carcinoma than those in malignant germ cell tumors. This is supported by the observation that ARID1A , PIK3CA , and CTNNB1 alterations are frequently detected in ovarian clear cell and endometrioid carcinomas but are uncommon in malignant germ cell tumors [ 7 , 8 , 9 ]. These findings support the hypothesis that pure YSTs in postmenopausal women may originate from epithelial ovarian cancer. In support of this concept, a recent study found that some pure YSTs in older women harbored diverse genomic alterations, including DICER1 , PIK3R1 , PTPRT , PMS1 , TP53 , PTEN , ARID1A , ARID1B , FGFR2 , and CTNNB1 [ 11 ]. Hall et al. identified somatic mutations in CTNNB1 , PIK3R1 , and PTEN in a case of YST with associated endometrioid carcinoma [ 4 ]. The authors reported that these mutational patterns are similar to those seen in pure epithelial counterparts, suggesting somatic derivation of the germ cell component. Skala et al. described six cases of YST with epithelial ovarian tumors and endometriosis or lesions suspicious for endometriosis involvement [ 12 ]. Somatic mutations were identified in CTNNB1 , FBXW7 , FGFR2 , RB1 , PIK3CA , PTEN , and TP53 . These findings support the notion that the germ cell tumor component of these tumors is often somatically derived. These two cases, like the present case, harbored the PIK3CA c.3140A>G (p.H1047R), a hotspot mutation. BEP is generally considered the standard chemotherapy regimen for ovarian YST, although the number of cycles varies according to residual disease and treatment response [ 13 ]. Nevertheless, treatment for YST in postmenopausal women is particularly challenging because of the rarity of these tumors and the lack of established therapeutic protocols. Many patients undergo platinum‐based chemotherapy such as TC or BEP, but the response rate is low and some cases show aggressive progression [ 2 ]. Serum AFP levels are elevated in most cases and useful for monitoring treatment response and disease recurrence. Among the 14 previously reported cases of pure YST in postmenopausal patients with known clinical outcomes, 11 experienced recurrence or death, indicating a very poor prognosis (Table  1 ) [ 2 , 3 , 5 , 14 , 15 , 16 , 17 , 18 , 19 ]. Of the 10 patients who received chemotherapy, 6 were treated with BEP and 2 with cisplatin and etoposide. While pure YST without epithelial components is usually treated by germ cell tumor protocols, the overall prognosis remains poor. For the present case, TC therapy, a standard treatment for epithelial tumors, was chosen as the initial treatment. However, the disease progressed. The regimen was changed to BEP and then VeIP, but these germ cell tumor‐based treatments were completely ineffective. Therefore, treatment based on epithelial tumor protocols may have been worth considering. Larger studies are needed to identify the optimal treatment approach. In addition to their potential role in tumor pathogenesis, these genomic alterations may have therapeutic implications. ARID1A deficiency and PIK3CA mutations have been associated with potential sensitivity to immune checkpoint inhibitors [ 20 ] and PI3K/AKT/mTOR pathway‐targeted therapies in ovarian clear cell carcinoma [ 21 ]. Although the clinical utility of these approaches in postmenopausal YST remains unknown, CGP may provide useful information for future therapeutic development in this rare disease. Summary of the literature on post‐menopausal patients with pure yolk sac tumors. Abbreviations: ACE, actinomycin D, cyclophosphamide, and etoposide; AWD, alive with disease; BEP, bleomycin, etoposide, and cisplatin; CG, carboplatin and gemcitabine; DDP, cisplatin, adriamycin, and 5‐fluorouracil; DOD, dead of disease; EP, cisplatin and etoposide; FP, 5‐fluorouracil and cisplatin; m, months; NED, no evidence of disease; POMB, methotrexate, vincristine, cisplatin, and bleomycin; TC, paclitaxel and carboplatin; TG, paclitaxel and gemcitabine. These findings suggest that YSTs in postmenopausal women may represent epithelial carcinomas with YST‐like features. Although they are pathologically similar to YSTs in younger patients, their genetic profiles and resistance to platinum‐based chemotherapy indicate a closer resemblance to clear cell carcinoma.

Written informed consent for publication was obtained from the patient.

Malignant germ cell tumors differ from epithelial ovarian cancers in several respects, including younger age at onset, rapid growth, predominantly unilateral localization, and a generally favorable prognosis [ 1 ]. Ovarian yolk sac tumors (YSTs) are a subset of malignant germ cell tumors that are most commonly diagnosed in young women. In postmenopausal women, YSTs are rare, and these cases exhibit a poorer prognosis compared with those in young women [ 2 , 3 ]. YSTs in postmenopausal women are unlikely to arise from germ cells because germ cells are not observed in ovaries after menopause. Thus, YSTs may arise from different biological pathways in young and postmenopausal women. YSTs resemble the morphology of the primitive yolk sac and exhibit a variety of histological patterns, including microcystic/reticular, endodermal sinus (festoon), solid, alveolar‐glandular, parietal, papillary, polyvesicular vitelline, hepatoid, and myxomatous patterns [ 4 ]. In contrast, most tumors in postmenopausal women with YSTs present with coexisting epithelial components. The most commonly reported epithelial components include endometrioid carcinoma (43%), high‐grade serous carcinoma (22%), clear cell carcinoma (13%), mucinous carcinoma (7%), and neuroendocrine tumor (7%) [ 5 ]. Approximately 20% of patients have multiple epithelial components and 31% have endometriosis, endometrioid cysts, or atypical endometriosis. The presence of epithelial components in YSTs in postmenopausal women indicates different molecular pathways compared with germ cell tumors in younger patients [ 6 ]. Genetic analysis of YST with epithelial components has shown results similar to those of the coexisting epithelial components. The origin of pure YST without epithelial components remains unclear. Here, we describe a postmenopausal patient with pure YST who underwent comprehensive genomic profiling (CGP). The origin of the YST was inferred from the CGP results. To the best of our knowledge, this case is the first report of CGP findings in a pure YST in a postmenopausal patient.

The authors declare no conflicts of interest.