In a neoplasm, dedifferentiation is characterised by the presence of a high-grade neoplasm which can occur de novo, be juxtaposed to, or arise as a recurrence of a previously well-differentiated tumour. Usually, this occurrence results in mesenchymal neoplasms. In epithelial malignant neoplasms, dedifferentiation has been observed in salivary gland carcinomas including adenoid cystic carcinoma, mucoepidermoid carcinoma, myoepithelial carcinoma, and acinic cell carcinoma. In addition, dedifferentiated carcinomas have been reported in the pancreas and in the gastrointestinal and urinary tracts. In the female genital tract, dedifferentiated carcinoma have been described in the endometrium and ovary. Histologically, this entity is characterised by both low-grade endometrioid carcinoma and a solid undifferentiated component.
Molecular alterations of EDC and impact on prognosis and therapy
To improve risk assessment for each endometrial carcinoma it is important to integrate molecular and clinicopathological data using such feasible techniques as immunohistochemistry and Sanger or next-generation sequencing analysis [50][51][52][53][68].
The combining of pathological classification and the surrogate markers suggested by TCGA demonstrates that immunohistochemistry plays an important role in the molecular classification of endometrial carcinomas both to establish a prognosis and therapeutic strategies. Thus, it is especially important that this ancillary technique is performed accurately without any problems due to bad fixation and bad antigen preservation [69], just as it is important to consider any intra-tumoral heterogeneity that the neoplasm might present.
Molecular alterations of EDC and impact on prognosis and therapy
In the Cancer Genome Atlas, EDC was reported as a heterogeneous neoplasm that had a higher number of mutations, so that it could be considered a high-risk endometrial carcinoma [22][53][70][71][72][73]. Because of its rarity, poor prognosis, and association with an advanced stage at diagnosis as well as presence of higher rates of gene mutations than other high-risk endometrial cancers, there is no established treatment. In addition, due to its morphological and genetic heterogeneity, to delineate a prognosis and treatment, it is important to evaluate for each case the genomic profiling [70][71][72][73]. Many Authors have reported associations between the prognosis and certain gene mutations, such as mismatch repair (MMR) gene mutations, POLE, SWI/SNF complex [72][73][74][75] and Tp53 mutations [70][76].
Mismatch repair (MMR) gene mutations
EDC is generally associated with MMR deficiency and is more frequent than endometrial carcinoma [12][20][77][78][79]. Although the number of cases of EDC that have been evaluated by some Authors were small [12][20][77][78][79], the data observed demonstrate that they have impact the prognosis and therapy in this rare malignancy. In fact, it has been demonstrated that MSI tumours are more immunogenic, and that immune checkpoint inhibitors (anti PD-1/PD-L1 antibodies) are effective for some tumours, such as colorectal carcinoma, melanoma, renal and lung carcinomas [80][81].
In addition, regarding EDC, Yokomizo et al. first demonstrated that the loss of MMR protein was observed only in the undifferentiated component [20]. In addition, Ono et al observed that in their cases there was MMR protein deficiency and that this was significantly associated with PD-L1 expression and the presence of tumour-infiltrating lymphocytes (CD8+), demonstrating that also EDC could be a target for immune checkpoint inhibitors. Due to the absence of PD-L1 expression in the well-differentiated component, Ono et al suggested that for this neoplasm the use of immune checkpoint inhibitors in combination with other
conventional chemotherapeutic agents, such as paclitaxel plus carboplatin and cisplatin, which could provide more satisfactory results, having an action on the growth of the well-differentiated component, too [79]. Immune checkpoint inhibitors, such as Pembrolizumab plus Lenvatinib have been used in KEYNOTE-146, a trial which included one case of EDC [82] or Dostarlimab, another anti PD-1 antibody, which showed an objective response in patients with recurrent or advanced MMRd Ecs [83]. Although a therapeutic strategy with a target for immune checkpoint inhibitors could improve the prognosis of patients affected by EDC, given that it is effective on the more aggressive undifferentiated component [26], it is important to keep in mind that immune checkpoint inhibitors can cause many adverse events involving multiple tissues and organs causing anaemia, lymphopenia, cutaneous rash, diarrhoea, hypoalbuminemia, dizziness, insomnia, headache and dyspnoea [80].
POLE domain mutations
In the TCGA classification, the POLE mutated/ultramutated group of endometrial cancers has been reported as a subtype of neoplasm that shows an excellent prognosis independently of other clinicopathological variables, such as high-grade tumours [64][65] and an advanced development stage[66]. Espinosa et al were the first to report the presence of POLE exonuclease mutations in one case of EDC [84] and then other additional cases were reported by Rosa-Rosa et al. [71] and later again by Espinosa et al. [85]. In the series of 18 cases of EDC, Rosa-Rosa et al. observed two POLE-mutated cases. In the series of POLE-mutated undifferentiated and dedifferentiated endometrial carcinomas reported by Espinosa et al [85] the neoplasms were more frequently stage I and the patients had a better prognosis than for other carcinomas lacking this mutational status. Of note, in their study, Espinosa et al [85], in evaluating the survival rate, observed that patients affected by EDCs with POLE domain mutations had a better prognosis than patients affected by advanced colorectal cancer with POLE domain mutations, who had statistically significant increase in mortality despite adjuvant therapy treatment [85]. Although in the series of high-grade endometrial carcinoma with POLE mutation of Concin et al. and Yu et al., there were only a few cases of EDC and EUC, these Authors suggested that in cases with concomitant POLEMut and P53abn the patients could be managed like patients with POLEmut neoplasms, since the prognosis remained good [86][87].
Switch/Sucrose Non-Fermentable (SWI/SNF) complex
SWI/SNF complex is a family of ATP chromatin remodelling complexes, present in eukaryotes and basically it is a group of proteins capable of remodelling the way DNA is packaged. In fact, this complex can stimulate ATPase activity that can destabilize histone-DNA interactions causing structural change and nucleosome rearrangement. As a result of this nucleosome rearrangement, some genes may be activated or repressed [88]. It has been demonstrated that neoplastic cells with SWI/SNF subunit mutations have disrupted chromatin structures and fail to express many genes. The subunits that are frequently mutations in mammalian malignancies are ARID1A [89], PBRM1 [90], SMARCB1 [91], SMARCA4 [92] and ARID2 [93]. In UEC/EDC, the protein products of core components of the SWI/SNF chromatin-remodelling complex that can be lost are {e.g., SMARCB1 (INI1) SMARCA2 (BRM), SMARCA4 (BRG-1), ARID1A,
ARID1B. More commonly, in the undifferentiated component of EDC, SMARCA4 (BRG-1), ARID1A and ARID1B are inactivated, with an absence of their expression on immunohistochemical analysis [27][30][94][95][96][97].
Neoplasms with core SWI/SNF-deficiency characterizes an extremely aggressive group of undifferentiated cancers which have a rapid disease progression that is refractory to conventional platinum/taxane-based chemotherapy [95]. In fact, Tessier-Cloutier et al, in their study, observed that at initial presentation, 55% of EDC with SWI/SNF deficiency had extrauterine spread in contrast to 38% of EDC with SWI/SNF-intact tumours. However, these Authors observed that for prognosis it is also important to evaluate the POLE status, given that all patients with mutated POLE showed a better prognosis with longer survival rates [95]. Until now there are no therapeutic applications in Ec with SWI/SNF mutations. However, due to the fact that the SWI/SNF subunit mutations are observed in a wide types of malignant neoplasms [98][99][100] and that there are drugs that target these genetic alterations, probably the same targeted therapies could enhance anti-cancer treatment effectiveness and provide new insights for therapeutic strategies in EDC, too. AU-15330 is a proteolysis-targeting chimera (PROTAC) degrader of the SWI/SNF ATPase subunits, SMARCA2 and SMARCA4. AU-15330 induces potent inhibition of tumour growth in xenograft models of prostate cancer and increased effectiveness of the androgen receptor enzalutamide, causing disease remission in castration-resistant prostate cancer (CRPC) models without toxicity [99].
Another therapeutic strategy that can be considered for treatment of EDC/UDC is Aurora A, which has been reported as a therapeutic target in ARID1A-deficient colorectal cancer cells [106].
In addition it seems worth investigating the use of Tazemetostat for the treatment of EDC/UEC, since this drug is well tolerated in patients with advanced epithelioid sarcoma with a loss of INI1/SMARCCB1 [101], and its effectiveness increases in association with doxorubicin (ClinicalTrials.gov identifier: NCT042049441).
Tp53 mutations
Tp53 mutations represent other types that can be observed in EDC [70][71][76]. In their study, Rosa-Rosa et al observed, by immunohistochemical analysis, that p53 staining was positive (aberrant) in the undifferentiated component and negative in the differentiated component, suggesting that this part of the neoplasm may be developing through a “serous-like” pathway [71]. Usually, the mutation of Tp53 can be evaluated using immunohistochemical analysis, which can be considered a surrogate method for molecular study. In fact, the majority of Tp53 mutations that can be observed by immunohistochemistry, are missense mutations and they can be demonstrated as detection of overexpressed protein. However, it is important to keep in mind that there are studies which have revealed that the nonsense TP53 mutations result in an absence of immunoreactivity due to the lack of gene product [102][103]. Moreover, it is important for immunohistochemical analyses to consider that this must be performed accurately without problems due to bad fixation and bad antigen preservation [69]. In addition, it is important to consider any morphological and genetic intra-tumoral heterogeneity that the neoplasm might have [22][53][70][71][72][73]. Since Tp53 mutation represents another genetic alteration of EDC [70][71][76] agents targeting the mutant p53 pathway could be considered for treatment of this malignancy. Among these targeting agents, it is worth considering APR-246 (Eprenetapop). There are Authors who have demonstrated that APR- 246 is well tolerated with a clinical response and remissions because it is capable of restoring wild-type p53 function in malignant cells administered in combination with azacytidine in patients with TP53-mutant myeloproliferative neoplasms, such as acute myeloid leukaemia (ClinicalTrials.gov identifier: NCT03072043) [104]. In addition, Adavosertib, which is a potent antitumor kinase inhibitor, in combination with carboplatin in advanced TP53 mutated ovarian cancer has been used at phase II (ClinicalTrials.gov identifier: NCT01164995) [105].
In conclusion, through researchers' review of the literature, it is clear that EDC is an extremely aggressive neoplasm with poor prognosis. On pathological analysis, this rare malignancy can be misdiagnosed, and its incidence therefore could be underestimated. However, a large number of studies in the last few years have contributed to improving pathological diagnoses by using immunohistochemical analyses. The use of immunohistochemical analysis with specific markers, which can be considered surrogates of molecular techniques, have also contributed to improving knowledge of the molecular and prognostic features of this malignancy. In addition, since EDC has a specific mutation, a therapeutic approach with potent targeting drugs could be investigated for its treatment, especially for more aggressive tumours, with a worse prognosis, such as those with SWI/SNF complex mutations. Researchers suggest combining pathological classification and surrogate TCGA molecular classification, options to improve the assessment of prognosis. In addition, further molecular analyses should be investigated to establish whether genetic alterations could have an impact on immunohistochemical expression of their surrogates. On pathological examination, in using immunohistochemistry, which can replace molecular studies, it is important that this is performed accurately without any problems due to bad fixation and bad antigen preservation [67] to avoid false negative results.
This entry is adapted from the peer-reviewed paper 10.3390/cancers15215155