In this case, we once wrongly diagnosed SDUS due to the predominance of an undifferentiated carcinoma component, a high proportion of rhabdoid structure, and a stable microsatellite. Thus, we tried anti-PD-1 immunotherapy but got poor outcomes, which made us reflect on the importance of a clear pathological diagnosis in our work.The median age at diagnosis of UEC was 65 years old [2], slightly older than SDUS (mean age 36), and older than patients with small cell carcinoma of the ovary, hypercalcaemic type (SCCOHT) (mean age 29) [10]. Postmenopausal bleeding is the most common symptom (76.9%), and most patients are advanced at the time of diagnosis (64.7%). [11] Typical undifferentiated carcinoma lacks adenoid or squamous differentiation, has no obvious nest-like or trabecular structures, and the tumor cells are more uniform, mostly medium-sized, lack adherence, and have a solid sheet-like distribution with a morphology similar to lymphoma, high-grade endometrial mesenchymal sarcoma, or small cell carcinoma. SDUS typically presents with diffuse sheets of mediumto large-sized epithelioid cells with areas of transverse myeloid morphology, interstitial hyalinized cords, focal lobulated structures, and less frequently, nuclear pleomorphism. Immunohistochemistry SDUS shows loss of SMARCA4 (BG1) protein expression; negative or focally positive for the epithelial markers CK, wave protein, and Claudin-4; but focal or weak for CD10.Negative for WT-1 and myogenic markers.The SDUS has a microsatellite stable phenotype for mismatch repair proteins, a low mutational load, occasional mutations in the TP53, RB1, and CTNNB1 genes, and no other mutations, rearrangements, or fusions have been identified so far[12, 13].The tumor in this case study had 90% undifferentiated cancer. Undifferentiated endometrial cancer may be challenging to distinguish from undifferentiated uterine sarcoma if it displays a striated muscle appearance and is microsatellite stable. We ultimately came to the conclusion that it was undifferentiated endometrial cancer due to the prominent expression of epithelial markers.We propose that to distinguish SMARCA4-deficient dedifferentiated endometrioid carcinoma from mimics, such as undifferentiated carcinomas with rhabdoid morphology, immunohistochemistry for cyclin D1, claudin-4, SMARCA4 and mismatch repair can be employed in conjunction with histology in the right clinical situation if genomic studies are not available [12] .
Other differential diagnoses include high-grade endometrial mesenchymal sarcoma, epithelioid smooth muscle sarcoma, and lymphoma. Endometrial mesenchymal tumors show strong CD10 staining and no microsatellite instability[14]. Smooth muscle uterine tumors such as smooth muscle sarcoma express SMA, desmin, and MSA[15]. Smooth muscle markers are positive in epithelioid smooth muscle sarcoma and malignant PEComa, which both have complex chromosomal abnormalities. In addition, the uterus can also develop primary or secondary lymphomas. CD45, as a leukocyte common antigen, is a specific marker for lymphoma and helps in the diagnosis.[16].
The SWI/SNF complex is a tumor suppressor gene. The core members of the SWI/SNF complex are SMARCB1 (INI1), SMARCA2 (BRM), and SMARCA4 (BRG1). Mutations or loss of function in these genes have been reported in a variety of human cancers. [17, 18].Some of the more prevalent gynecologic malignancies, such as endometrioid and clear cell carcinoma of the ovary, have been found to carry mutations in non-core SWI/SNF complex components such as ARID1A.It is becoming more widely understood that the SWI/SNF complex's essential parts, such as SMARCB1 and SMARCA4, may be involved in the secondary dedifferentiation process of the gastrointestinal pancreatic cancer subpopulation, leading to rhabdomeric phenotypic metastasis (rhabdomeric dedifferentiation) with a poor prognosis. [19, 20]. Both SWI/SNF complex mutations and rhabdoid alterations are not present in all dedifferentiated carcinomas.15 of 30 (50%) DEC showed loss of both SMARCA2 and SMARCA4 (37%), or loss of both SMARCA2 and SMARCB1 (13%), in the undifferentiated component. Both SMARCB1 and SMARCA4 losses are mutually exclusive. This is also consistent with our case. These essential SWI/SNF proteins were completely expressed in each of the 31 grade 3 endometrioid carcinomas.In about 23% of SMARCA4-deficient instances, the morphology is rhabdoid[21–22].
The absence of MMR protein in both endometrioid and undifferentiated carcinomas characterizes the vast majority of dedifferentiated endometrial carcinomas. Rosa-Rosa et al. compared the mutation sites of common MMR genes in undifferentiated carcinoma components and endometrioid adenocarcinoma components and discovered that the overlap rate was more than 90%, lending credence to the theory that dedifferentiated or undifferentiated endometrial cancer may be the result of dedifferentiated endometrial cancer[23].What's interesting is that in our case, SMARCA4 loss was only present in the undifferentiated carcinoma, which manifested as MMRp (microsatellite stability). As well, the endometrioid adenocarcinoma component had intact SMARCB1 and SMARCA4 expression, which manifested as MMRd (microsatellite instability). That is to say, an endometrioid carcinoma with the genetic background of MMRd has a SMARCA4 mutation, which leads to the emergence of undifferentiated subclones (undifferentiated cancer cells with rhabdomyoid morphology caused by the SMARCA4 mutation), and undifferentiated cancer cells acquire stemness. During the evolution of stemness, some cells regain MMR protein expression. It is speculated that the loss of core components of the SWI/SNF chromatin remodeling complex drives the dedifferentiation process and the emergence of the striated muscle phenotype in EC. This patient ultimately had a very poor prognosis because of a mutation in SMARCA4, which resulted in an undifferentiated component of the tumor that was ultimately fatal. It's a pity that this patient declined genetic testing.
Tessier-Coutier et al. [24] discovered that SWI/SNF complex protein deletion was found in 56 (68.3%) of 82 cases of dedifferentiated or undifferentiated endometrial cancer. Stage I cancers lacking SWI or SNF had a 2-year disease-specific survival rate of 65%, compared to 100% for tumors with SWI or SNF intact. The median survival for patients with stage III and stage IV disease was 4 months for SWI/SNF-deficient tumors and 36 months for SWI/SNF-complete tumors. In addition, standard platinum-based chemotherapy is ineffective against SWI/SNF-deficient malignancies.SMARCA4 mutations also contribute to the very aggressive monogenic malignancy known as small cell carcinoma of the ovary, hypercalcemic type (SCCOHT). As evidence of the paradox between low mutational load and high immunogenicity, SCCOHT responds well to anti-PD1 immunotherapy. This study also raises the possibility that the high immunogenicity of SMARCA4-deficient tumors may be related to this result. Some researchers have also suggested that inhibitors of the H3K27 histone methyltransferase EZH2 can specifically kill malignant rhabdoid tumors that lack the SWI/SNF core component. This approach might be therapeutic[25, 26]. Whether immunotherapy is effective in the treatment of undifferentiated endometrial cancer requires further data.A lack of somatic mutations or epigenetic alterations in the MMR gene results in MMRd (e.g., MLH1 promoter methylation is deficient in MLH1 and PMS2). Germline mutations in the MMR gene (Lynch syndrome) are comparatively uncommon [27]. Endometrial cancer will not benefit from immunotherapy if the methylation of the MLHI gene promoter is altered [28]. We nonetheless tried PD1 treatment on this patient despite the lack of genetic testing and the low percentage of MMR-positive phenotypic cells, but the outcome was not good.
Undifferentiated endometrial tumors were divided into BRG1/INI1-deficient and BRG1/INI1-intact kinds by Hoang et al. [11]. In every instance of BRG1/INI1-deficient tumors, they discovered a complete lack of PAX8 and ER staining in the undifferentiated component [29]. Endometrial dedifferentiated cancer only displayed more aggressive behavior and a poorer prognosis when the proportion of undifferentiated components reached 20% [30]. The prognosis of endometrial cancer improves with a higher rate of ER and PR positivity, and ER and PR in endometrial cancer could be employed as independent prognostic variables [31]. 90% of the instances in the current instance were accounted for by the rhabdoid component, and both the ER and PR were negative. Rapid disease development in the patient and a very bad prognosis once more proved that the SWI/SNF complex gene deficiency identifies a class of exceptionally aggressive, undifferentiated malignancies that are challenging to treat with traditional platinum/paclitaxel-based chemotherapy. More clinical trials are needed to confirm that the more mutations in this complex, the worse the prognosis for dedifferentiated endometrial cancer.